Sotwe logo Sotwe logo
JLTerminal's profile banner image
JLTerminal's profile image

JLTerminal

Verified account

@JLTerminal

🧑
Joined July 2026
8 Following
14 Followers
47 Posts
Β Pinned Tweet
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/18/26 Market Analysis: Bitcoin I am now 95 to 100% invested in Bitcoin. Should we see a lower low once more, I may add a small amount. If not, I am already practically all in. With this, I have made my last large purchase. I expect a massive breakout in the coming weeks, in the order of 7,000 to 12,000 dollars. The decisive question, as so often, is in which direction. A further, stronger decline with a lower low is possible. However, I consider it more likely that we rise more strongly. Should we sustainably break the 67,500 dollar mark, confirm the area with a retest and then hold above it for a longer period, I consider it likely that the uptrend continues and that, over the long term, the bottom may be in. This does not mean that I assume the bottom is already in. I do, however, assume that we are in the bottom formation phase, much like in the period from June 2022 to February 2023. Back then too, it was ultimately not decisive whether one hit the exact bottom. Anyone who was invested during that phase profited strongly in the following months and years regardless. I would rather be invested somewhat too early and have bought at what I consider a good price than enter too late. Therefore I am satisfied with my entry in any case, regardless of whether we see a lower low or not. Time in the Market Beats Timing the Market. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/18/26 Market Analysis: Bitcoin I am now 95 to 100% invested in Bitcoin. Should we see a lower low once more, I may add a small amount. If not, I am already practically all in. With this, I have made my last large purchase. I expect a massive breakout in the coming weeks, in the order of 7,000 to 12,000 dollars. The decisive question, as so often, is in which direction. A further, stronger decline with a lower low is possible. However, I consider it more likely that we rise more strongly. Should we sustainably break the 67,500 dollar mark, confirm the area with a retest and then hold above it for a longer period, I consider it likely that the uptrend continues and that, over the long term, the bottom may be in. This does not mean that I assume the bottom is already in. I do, however, assume that we are in the bottom formation phase, much like in the period from June 2022 to February 2023. Back then too, it was ultimately not decisive whether one hit the exact bottom. Anyone who was invested during that phase profited strongly in the following months and years regardless. I would rather be invested somewhat too early and have bought at what I consider a good price than enter too late. Therefore I am satisfied with my entry in any case, regardless of whether we see a lower low or not. Time in the Market Beats Timing the Market. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
2
0
0
50
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: AI Cheaper AI, deeper cracks The AI industry faces a test that is rarely stated this plainly. The models are getting better and cheaper at the same time. For users that is good news. For the companies behind them it creates a problem that cannot be solved by further efficiency, because efficiency is what creates it. This analysis follows that connection from its starting point through to its consequence. It begins with prices, moves to the question of where usage is shifting, shows through concrete figures what gap results from it, and ends with the financing that stays committed across the entire period. The prices The LLM Token Expenditure Index from Silicon Data measures what the market effectively pays for one million inference tokens. Despite its name it tracks a usage weighted price rather than total spending. Silicon Data have acknowledged that the label is misleading. The index started at around 1.09 Dollar in December 2025, climbed to a peak of 2.12 Dollar in May 2026 and then turned. It stood at 1.02 Dollar in August, hit a record low of 97 cents on 1 September and currently reads 1.01 Dollar. Measured from the peak it has therefore roughly halved within four months and now sits slightly below where it began. Over the longer horizon, the price per token has fallen by more than 90% since 2023. At the same time the models have become more capable. Where usage is shifting Falling prices per token do not relieve the infrastructure, because that infrastructure is already built, financed and contractually committed. They reduce only what a provider takes in per unit. What compounds this is where usage is moving. The index makes that mechanism visible. Because it is usage weighted, it falls not only when providers cut prices but also when usage shifts toward cheaper models. Silicon Data state it explicitly: if usage moves toward higher priced models the index rises, if it moves toward lower priced ones it falls. How much of the decline comes from price cuts and how much from that shift cannot be broken out from the index alone, and I do not want to blur that. What can be established is that spending is migrating toward the smaller and cheaper models, because they are sufficient for most applications, while investment continues to flow into the large frontier models. Revenue therefore rests on the segment that brings in the least, while the costs rest on the one that consumes the most. What the figures say The scale can be read from OpenAI. According to documents reported by the Financial Times: 1. A cumulative negative free cash flow of 278 billion Dollar between 2026 and 2030 2. Roughly 856 billion Dollar for compute and infrastructure through 2030 3. Revenue rising from 36 billion Dollar this year to 350 billion Dollar in 2030 4. In total roughly 840 billion Dollar in cumulative revenue through the end of 2030 Point four is the decisive one. Cumulative revenue of 840 billion against cumulative infrastructure costs of 856 billion. Even if every single one of these projections is met, a deficit remains at the end of the decade. That is not the crisis case, that is the base case. Worth noting alongside it: in February OpenAI still told its investors the compute bill would come to around 600 billion Dollar. Five months later it stands at 856 billion. How maturities and credit markets can become a danger The situation is compounded by the structure of the financing. The infrastructure is not committed for the short term but across extremely long maturities. Oracle disclosed 288 billion Dollar in additional lease commitments as of 31 August 2026, almost all of them relating to data centres. The terms run 15 to 19 years, and commencement is expected between the second quarter of fiscal 2027 and fiscal 2029. The bulk of these payments has therefore not yet begun. Against this stand customer contracts of typically around 5 years. That is the core of the problem: obligations over 15 to 19 years against revenue over 5 years, for a technology whose demand nobody can seriously forecast 2 years out, let alone 15. The first cracks are already visible. Loans of 18 billion Dollar for the Oracle leased Project Jupiter data centre in DoΓ±a Ana County, New Mexico, are being quoted by syndicate banks including Santander and Jefferies at only 89 to 91 cents on the dollar. The banks were forced to hold more of this debt on their own books than planned, because buyers could not be found. The credit rating at Oracle sits one notch above junk status following a downgrade by S&P in July. Michael Burry has pointed to a detail that is easily overlooked here. Lease commitments of this kind are held off balance sheet, and expense only arises once a data centre actually goes into operation. Since the bulk of these facilities is not yet running, the 288 billion Dollar at Oracle appear so far neither as debt nor as a cost item. Anyone looking at leverage and cash flow simply does not see the burden. Oracle is not alone in this either: Alphabet disclosed 85.2 billion Dollar in future lease payments as of 30 June 2026, predominantly for data centres and likewise not yet commenced, and other companies hold comparable obligations. If demand stays strong, the bet pays off. If it weakens, the costs remain regardless. What follows from this The connection can be described as a pair of scissors. The price per unit falls faster than the cost of the assets producing those units can fall. Efficiency does not close those scissors, it opens them further. In an earlier analysis I assumed precisely the opposite and wrote that the only way out of this cost trap would be for the models to consume less compute while becoming better at the same time. Both have happened. The relief I expected from it has failed to materialise. I therefore consider that thesis refuted. The decisive question of the coming years is not whether AI gets used. It is being used heavily and that will continue to grow. The question is who actually earns money from that usage in the end, and how long the credit market remains willing to carry the infrastructure behind it. I hold no positions on this theme and would not recommend any to anyone, as this market can be highly complex and extremely volatile. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. Market Analysis: AI Cheaper AI, deeper cracks The AI industry faces a test that is rarely stated this plainly. The models are getting better and cheaper at the same time. For users that is good news. For the companies behind them it creates a problem that cannot be solved by further efficiency, because efficiency is what creates it. This analysis follows that connection from its starting point through to its consequence. It begins with prices, moves to the question of where usage is shifting, shows through concrete figures what gap results from it, and ends with the financing that stays committed across the entire period. The prices The LLM Token Expenditure Index from Silicon Data measures what the market effectively pays for one million inference tokens. Despite its name it tracks a usage weighted price rather than total spending. Silicon Data have acknowledged that the label is misleading. The index started at around 1.09 Dollar in December 2025, climbed to a peak of 2.12 Dollar in May 2026 and then turned. It stood at 1.02 Dollar in August, hit a record low of 97 cents on 1 September and currently reads 1.01 Dollar. Measured from the peak it has therefore roughly halved within four months and now sits slightly below where it began. Over the longer horizon, the price per token has fallen by more than 90% since 2023. At the same time the models have become more capable. Where usage is shifting Falling prices per token do not relieve the infrastructure, because that infrastructure is already built, financed and contractually committed. They reduce only what a provider takes in per unit. What compounds this is where usage is moving. The index makes that mechanism visible. Because it is usage weighted, it falls not only when providers cut prices but also when usage shifts toward cheaper models. Silicon Data state it explicitly: if usage moves toward higher priced models the index rises, if it moves toward lower priced ones it falls. How much of the decline comes from price cuts and how much from that shift cannot be broken out from the index alone, and I do not want to blur that. What can be established is that spending is migrating toward the smaller and cheaper models, because they are sufficient for most applications, while investment continues to flow into the large frontier models. Revenue therefore rests on the segment that brings in the least, while the costs rest on the one that consumes the most. What the figures say The scale can be read from OpenAI. According to documents reported by the Financial Times: 1. A cumulative negative free cash flow of 278 billion Dollar between 2026 and 2030 2. Roughly 856 billion Dollar for compute and infrastructure through 2030 3. Revenue rising from 36 billion Dollar this year to 350 billion Dollar in 2030 4. In total roughly 840 billion Dollar in cumulative revenue through the end of 2030 Point four is the decisive one. Cumulative revenue of 840 billion against cumulative infrastructure costs of 856 billion. Even if every single one of these projections is met, a deficit remains at the end of the decade. That is not the crisis case, that is the base case. Worth noting alongside it: in February OpenAI still told its investors the compute bill would come to around 600 billion Dollar. Five months later it stands at 856 billion. How maturities and credit markets can become a danger The situation is compounded by the structure of the financing. The infrastructure is not committed for the short term but across extremely long maturities. Oracle disclosed 288 billion Dollar in additional lease commitments as of 31 August 2026, almost all of them relating to data centres. The terms run 15 to 19 years, and commencement is expected between the second quarter of fiscal 2027 and fiscal 2029. The bulk of these payments has therefore not yet begun. Against this stand customer contracts of typically around 5 years. That is the core of the problem: obligations over 15 to 19 years against revenue over 5 years, for a technology whose demand nobody can seriously forecast 2 years out, let alone 15. The first cracks are already visible. Loans of 18 billion Dollar for the Oracle leased Project Jupiter data centre in DoΓ±a Ana County, New Mexico, are being quoted by syndicate banks including Santander and Jefferies at only 89 to 91 cents on the dollar. The banks were forced to hold more of this debt on their own books than planned, because buyers could not be found. The credit rating at Oracle sits one notch above junk status following a downgrade by S&P in July. Michael Burry has pointed to a detail that is easily overlooked here. Lease commitments of this kind are held off balance sheet, and expense only arises once a data centre actually goes into operation. Since the bulk of these facilities is not yet running, the 288 billion Dollar at Oracle appear so far neither as debt nor as a cost item. Anyone looking at leverage and cash flow simply does not see the burden. Oracle is not alone in this either: Alphabet disclosed 85.2 billion Dollar in future lease payments as of 30 June 2026, predominantly for data centres and likewise not yet commenced, and other companies hold comparable obligations. If demand stays strong, the bet pays off. If it weakens, the costs remain regardless. What follows from this The connection can be described as a pair of scissors. The price per unit falls faster than the cost of the assets producing those units can fall. Efficiency does not close those scissors, it opens them further. In an earlier analysis I assumed precisely the opposite and wrote that the only way out of this cost trap would be for the models to consume less compute while becoming better at the same time. Both have happened. The relief I expected from it has failed to materialise. I therefore consider that thesis refuted. The decisive question of the coming years is not whether AI gets used. It is being used heavily and that will continue to grow. The question is who actually earns money from that usage in the end, and how long the credit market remains willing to carry the infrastructure behind it. I hold no positions on this theme and would not recommend any to anyone, as this market can be highly complex and extremely volatile. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
0
0
0
20
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Some of the data I used here came from @KobeissiLetter. I cross checked it against other sources. Credit where it is due, they do good work.
0
0
0
0
25
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: Trump's Lost Ego War From the rat tunnel through oil and inflation to the question of who ultimately pays the bill The war between the United States and Iran has now been running for 204 days. Before I turn to the markets, one thing matters to me: what is happening there costs people their lives and their health, on all sides. That is the real tragedy of this situation, and no figure in this text relativises it. This text nonetheless places the conflict deliberately in economic and geopolitical terms, because that is the field I work in and the one where I can contribute something. Since the war began, a single sentence has dominated the coverage: oil is rising. That observation is correct, but it falls considerably short. The oil price is not the outcome of this development, merely its first link. What follows from it now extends from the freight markets through consumer prices and into the interest rate decisions of central banks. This analysis follows that chain from beginning to end. It starts with the question of why this war is structurally unwinnable, moves through the question of where Trump now stands psychologically, and ends with the question of who profits from all of it and who pays for it. Why this war has no solution There is no solution to this war. Trump cannot afford a defeat, and at the same time he cannot win. That is the heart of the matter, and everything else follows from it. Iran holds every card. The regime is barely defeatable, because it is built to be almost immortal in its structure. The country cannot be taken, because it is simply too large. And the United States cannot control the Strait of Hormuz, because Iran does not control it through occupation but through fear. In parallel, the Iran-backed Houthis have ostensibly taken Yemen's coastline at Bab al Mandab and brought the region's second great chokepoint under their control. The oil price therefore has no ceiling. It rises, it swings, and even if it were capped, the oil itself could eventually simply run out. The end of the tunnel In my analysis of Donald Trump I compared his behaviour to a rat in a tunnel. In the middle of the tunnel he holds the illusion of control. At both ends he loses it. At one end he retreats out of fear and de-escalates. At the other he grows angry and answers with supposed strength. Trump has now arrived at the first end, the end of anger. I consider it possible that Iran will be bombed more heavily in the coming hours and days than before. Based on past patterns, more likely at weekends, so in the window from 26 to 28 September or from 3 to 5 October. And this time possibly at an intensity we last saw in the opening weeks of the war, when Tehran itself was struck across the board. Both ends of this tunnel lead to a loss of control. This time, however, it looks as though it ends in escalation. Whether military or economic. Who this affects This affects not only the people in Iran and in the surrounding states such as Qatar, because Iran will strike back at those first. It affects the entire world. Not militarily, but economically. How far that effect already reaches can be read off the figures, not off opinions. The following list is documented record. The factual picture 1. The freight market is out of control. Rates for supertankers on the route from the Gulf of Oman to China reached around 450 points on a Worldscale basis in the week of 11 September, equivalent to roughly 11.50 Dollar per barrel, and the highest level since that rate was launched after the war began. The Baltic index for supertankers stands around 180% above its low of 2 July. 2. The Breakwave Tanker Shipping ETF is up 5,691% over twelve months and trades at 872 Dollar. That makes it the best performing ETF in the world this year. It tracks futures on the cost of transporting crude from the Middle East and West Africa to China and therefore reacts immediately to disruptions in global oil traffic. 3. The US Strategic Petroleum Reserve fell by 403,000 barrels to 285 million barrels, its lowest level since November 1982. It is the 25th consecutive weekly decline, the longest run since the 2021 to 2023 period. Since the start of the year 131 million barrels have drained out, a decline of 32%. In 2020 the reserve still held 650 million barrels, some 128% above today's level. 4. The United States is turning to Venezuela. Crude imports from there rose last week by 183,000 to 782,000 barrels per day, the highest weekly figure of the year. Against the level of six months ago that is an increase of 359,000 barrels per day, or 85%. 5. Saudi Arabia is no longer supplying Europe. European refiners have been told they will receive no crude allocation next month. Europe previously took 577,000 barrels per day from there under term contracts. 6. The diesel price in the United States stands at 6.49 Dollar per gallon and continues to set new record highs. It has doubled since its January low. In California it stands at 8.41 Dollar. 7. In Europe diesel prices are around 38% above their level at the start of the year, gas more than 130%. The kerosene price hit an all time high of 1,698 Dollar per tonne in March. European governments are now discussing a bloc-wide windfall tax on energy companies as public pressure grows. 8. In equities there is one clear winner. The energy sector in the S&P 500 is up around 40% since the start of the year, by a wide margin the best sector of the year. 9. Interest rates are already following. The Federal Reserve raised its policy rate unanimously by 25 basis points to 3.75 to 4.00% on 16 September, the first increase since 2023. The statement says inflation remains elevated. For October the market is currently pricing a probability of around 60% for a further step. Even if the strait reopens tomorrow The obvious assumption is that all of this dissolves the moment Hormuz is open again. That assumption is misleading. First, the starting point. Before the war more than a hundred ships passed through the strait each day, around 20 million barrels. Kpler, Lloyd's List and PortWatch currently count between 5 and 12 transits per day, averaging 7 over six months. Crude shipped directly through the strait runs at 2.2 million barrels per day. Official American figures are considerably higher, because they include naval auxiliaries, tugs and coastal craft. In practice the passage is close to zero. Even a reopening would change little at first, for three reasons. First, insurance. War risk premiums stand at forty times pre-war levels, around 10 million Dollar for a single supertanker. Six P&I clubs have withdrawn cover entirely. Premiums do not follow headlines, they follow loss history. They fall only once nothing has happened for months. Second, logistics. More than four hundred ships are currently holding position. Fleets that have spent months routing around Africa are not back in the right place on the day of reopening. Charter contracts run for months, loading schedules for weeks. The system needs time to reorder itself. Third, and this is the decisive point: a substantial part of the shortfall does not stem from the blockade at all, but from destroyed plant. A blockade ends on a date. Destroyed infrastructure ends after construction time. In Qatar, Iranian missiles struck Ras Laffan in March and caused severe damage. Two of fourteen liquefaction trains went offline, 17% of that country's export capacity. The operator cites three to five years to rebuild. Shell's Pearl gas to liquids plant ceased production entirely and force majeure was declared. In Saudi Arabia the Abqaiq processing facility was struck by the Houthis at the end of July and taken out of service. Ras Tanura temporarily lost 550,000 barrels of capacity in early March, Satorp at Jubail 460,000, the Riyadh refinery 120,000. Manifa and Khurais each gave up around 300,000 barrels on 9 April. The East West pipeline was hit twice, in the week of 9 April with a loss of 700,000 barrels per day and again on 11 September by a drone attack launched from Iraq's Maysan province. It has been down ever since. That is precisely where the supply halt to Europe comes from, because this line had latterly become not the alternative route but the main one. With Hormuz blockaded, exports ran through Yanbu on the Red Sea, most recently at 4 to 5 million barrels against a capacity of 7 million. In Iran, Israel almost completely destroyed refinery 4 at South Pars on 18 March and severely damaged refinery 7, a loss of around 12% of Iranian gas production. The Tehran refinery and the fuel depots at Shahr-e Rey, Shahran and Karaj burned on 7 March. Added to this are Kuwait, where Mina Abdullah was hit on 19 March and Mina al Ahmadi on 3 April, the United Arab Emirates with Ruwais, the Habshan gas processing complex and the Shah gas field, and Bahrain, where the Bapco refinery with 405,000 barrels of capacity was struck on 9 March and force majeure was declared. How far this damage reaches is visible in production itself. Saudi Arabia is delivering 6.24 million barrels per day, as little as it last did in 1990, and global supply is falling this year by 5.7 million barrels per day, equivalent to 6% of world production. Europe is hit hardest, because unlike the United States it cannot cover its own needs and imports around 97% of its crude. That leaves only the strategic reserves, and those buffers are largely spent as well. Anyone betting on a swift normalisation is therefore betting that infrastructure can be restored faster than is physically possible. What this means for the market The following assessment is explicitly not an investment tip, as I hold no trades in this area. For understanding how this market works, it is nonetheless important. It is not only oil that is rising and will continue to rise, but gas as well. Partly because it serves as substitute energy in power generation, heating and industry. Partly because a substantial share of global gas trade ran through the same chokepoints. Likewise all tanker and shipping equities tend to rise. The list below names the groups in the order in which they benefit. 1. Crude tankers, the immediate winners. Frontline (approx. 51 Dollar), International Seaways (approx. 111 Dollar), DHT Holdings (approx. 23 Dollar), Scorpio Tankers (approx. 87 Dollar) and Okeanis Eco Tankers (approx. 85 Dollar). They trade predominantly in the spot market and therefore earn directly on every move in rates. As a group they are up around 120% since the start of the year. 2. Gas carriers, the delayed winners. Dorian LPG (approx. 58 Dollar) and Navigator Gas (approx. 25 Dollar) carry liquefied petroleum gas, Flex LNG (approx. 32 Dollar) liquefied natural gas. Around 30% of global LPG trade and around 20% of LNG trade ran through Hormuz before the war. Because gas carriers are more heavily chartered on long term contracts, rising rates feed through here only with a delay. Spot rates for LPG carriers hit their record in May at around 170,000 Dollar per day. 3. Shipowners who lease out, the quiet winners. Danaos (approx. 161 Dollar), Global Ship Lease (approx. 46 Dollar) and Costamare (approx. 15 Dollar). They charter their vessels out to the liners on long term contracts rather than operating them, and therefore do not carry the high fuel costs. At the same time charter rates and vessel values rise along with freight rates. Danaos stands at an all time high. 4. Container liners, the weakest link. ZIM (approx. 30 Dollar) and Matson (approx. 233 Dollar). The spot rate from China to the US East Coast stands at 10,948 Dollar per forty foot container. Earnings nonetheless follow this only in muted fashion, because the liners absorb the expensive fuel and the detours themselves. ZIM lifted second quarter net income to 64 million Dollar, up 170% year on year, but stayed well short of what the rate environment would suggest. 5. Energy equities in general, the broadest and least volatile way into the theme. At around 40% since the start of the year the sector sits below the tanker names, but it does not carry their dependence on a single strait. Who ultimately pays Oil gets more expensive, freight gets more expensive, diesel and gas get more expensive, and at the end stands a rate increase. That closes the chain this text has followed from the outset. What remains is the question of where it lands. For people without meaningful assets, inflation is a pure loss. Their income loses purchasing power, and they own nothing beyond it. For the wealthy the same movement reverses. The value of their real assets rises, and anyone who deliberately financed through investment credit rather than paying cash also sees the real burden of that debt shrink. Unlike consumer credit, there is an asset on the other side that rises with inflation. Two effects running in the same direction. This is reinforced by the fact that access to credit itself depends on wealth. Whoever owns little is not considered creditworthy and remains excluded from this mechanism, while financing it through their own eroding purchasing power. The same property stands at the beginning of the chain as at its end. A state that does not have to fund its armament out of taxes, but can stretch it through debt and the printing press, shifts the bill into the future and thereby also decides who carries it. Wars of this magnitude are financeable in such a system in the first place, and they remain so because the costs do not fall where they are decided. The bill is paid in the end regardless. Through purchasing power, and therefore by those who can least afford it. The world is unjust, and this is the one we live in. (Disclaimer: This reflects a personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any assets.)
JLTerminal's tweet photo. Market Analysis: Trump's Lost Ego War From the rat tunnel through oil and inflation to the question of who ultimately pays the bill The war between the United States and Iran has now been running for 204 days. Before I turn to the markets, one thing matters to me: what is happening there costs people their lives and their health, on all sides. That is the real tragedy of this situation, and no figure in this text relativises it. This text nonetheless places the conflict deliberately in economic and geopolitical terms, because that is the field I work in and the one where I can contribute something. Since the war began, a single sentence has dominated the coverage: oil is rising. That observation is correct, but it falls considerably short. The oil price is not the outcome of this development, merely its first link. What follows from it now extends from the freight markets through consumer prices and into the interest rate decisions of central banks. This analysis follows that chain from beginning to end. It starts with the question of why this war is structurally unwinnable, moves through the question of where Trump now stands psychologically, and ends with the question of who profits from all of it and who pays for it. Why this war has no solution There is no solution to this war. Trump cannot afford a defeat, and at the same time he cannot win. That is the heart of the matter, and everything else follows from it. Iran holds every card. The regime is barely defeatable, because it is built to be almost immortal in its structure. The country cannot be taken, because it is simply too large. And the United States cannot control the Strait of Hormuz, because Iran does not control it through occupation but through fear. In parallel, the Iran-backed Houthis have ostensibly taken Yemen's coastline at Bab al Mandab and brought the region's second great chokepoint under their control. The oil price therefore has no ceiling. It rises, it swings, and even if it were capped, the oil itself could eventually simply run out. The end of the tunnel In my analysis of Donald Trump I compared his behaviour to a rat in a tunnel. In the middle of the tunnel he holds the illusion of control. At both ends he loses it. At one end he retreats out of fear and de-escalates. At the other he grows angry and answers with supposed strength. Trump has now arrived at the first end, the end of anger. I consider it possible that Iran will be bombed more heavily in the coming hours and days than before. Based on past patterns, more likely at weekends, so in the window from 26 to 28 September or from 3 to 5 October. And this time possibly at an intensity we last saw in the opening weeks of the war, when Tehran itself was struck across the board. Both ends of this tunnel lead to a loss of control. This time, however, it looks as though it ends in escalation. Whether military or economic. Who this affects This affects not only the people in Iran and in the surrounding states such as Qatar, because Iran will strike back at those first. It affects the entire world. Not militarily, but economically. How far that effect already reaches can be read off the figures, not off opinions. The following list is documented record. The factual picture 1. The freight market is out of control. Rates for supertankers on the route from the Gulf of Oman to China reached around 450 points on a Worldscale basis in the week of 11 September, equivalent to roughly 11.50 Dollar per barrel, and the highest level since that rate was launched after the war began. The Baltic index for supertankers stands around 180% above its low of 2 July. 2. The Breakwave Tanker Shipping ETF is up 5,691% over twelve months and trades at 872 Dollar. That makes it the best performing ETF in the world this year. It tracks futures on the cost of transporting crude from the Middle East and West Africa to China and therefore reacts immediately to disruptions in global oil traffic. 3. The US Strategic Petroleum Reserve fell by 403,000 barrels to 285 million barrels, its lowest level since November 1982. It is the 25th consecutive weekly decline, the longest run since the 2021 to 2023 period. Since the start of the year 131 million barrels have drained out, a decline of 32%. In 2020 the reserve still held 650 million barrels, some 128% above today's level. 4. The United States is turning to Venezuela. Crude imports from there rose last week by 183,000 to 782,000 barrels per day, the highest weekly figure of the year. Against the level of six months ago that is an increase of 359,000 barrels per day, or 85%. 5. Saudi Arabia is no longer supplying Europe. European refiners have been told they will receive no crude allocation next month. Europe previously took 577,000 barrels per day from there under term contracts. 6. The diesel price in the United States stands at 6.49 Dollar per gallon and continues to set new record highs. It has doubled since its January low. In California it stands at 8.41 Dollar. 7. In Europe diesel prices are around 38% above their level at the start of the year, gas more than 130%. The kerosene price hit an all time high of 1,698 Dollar per tonne in March. European governments are now discussing a bloc-wide windfall tax on energy companies as public pressure grows. 8. In equities there is one clear winner. The energy sector in the S&P 500 is up around 40% since the start of the year, by a wide margin the best sector of the year. 9. Interest rates are already following. The Federal Reserve raised its policy rate unanimously by 25 basis points to 3.75 to 4.00% on 16 September, the first increase since 2023. The statement says inflation remains elevated. For October the market is currently pricing a probability of around 60% for a further step. Even if the strait reopens tomorrow The obvious assumption is that all of this dissolves the moment Hormuz is open again. That assumption is misleading. First, the starting point. Before the war more than a hundred ships passed through the strait each day, around 20 million barrels. Kpler, Lloyd's List and PortWatch currently count between 5 and 12 transits per day, averaging 7 over six months. Crude shipped directly through the strait runs at 2.2 million barrels per day. Official American figures are considerably higher, because they include naval auxiliaries, tugs and coastal craft. In practice the passage is close to zero. Even a reopening would change little at first, for three reasons. First, insurance. War risk premiums stand at forty times pre-war levels, around 10 million Dollar for a single supertanker. Six P&I clubs have withdrawn cover entirely. Premiums do not follow headlines, they follow loss history. They fall only once nothing has happened for months. Second, logistics. More than four hundred ships are currently holding position. Fleets that have spent months routing around Africa are not back in the right place on the day of reopening. Charter contracts run for months, loading schedules for weeks. The system needs time to reorder itself. Third, and this is the decisive point: a substantial part of the shortfall does not stem from the blockade at all, but from destroyed plant. A blockade ends on a date. Destroyed infrastructure ends after construction time. In Qatar, Iranian missiles struck Ras Laffan in March and caused severe damage. Two of fourteen liquefaction trains went offline, 17% of that country's export capacity. The operator cites three to five years to rebuild. Shell's Pearl gas to liquids plant ceased production entirely and force majeure was declared. In Saudi Arabia the Abqaiq processing facility was struck by the Houthis at the end of July and taken out of service. Ras Tanura temporarily lost 550,000 barrels of capacity in early March, Satorp at Jubail 460,000, the Riyadh refinery 120,000. Manifa and Khurais each gave up around 300,000 barrels on 9 April. The East West pipeline was hit twice, in the week of 9 April with a loss of 700,000 barrels per day and again on 11 September by a drone attack launched from Iraq's Maysan province. It has been down ever since. That is precisely where the supply halt to Europe comes from, because this line had latterly become not the alternative route but the main one. With Hormuz blockaded, exports ran through Yanbu on the Red Sea, most recently at 4 to 5 million barrels against a capacity of 7 million. In Iran, Israel almost completely destroyed refinery 4 at South Pars on 18 March and severely damaged refinery 7, a loss of around 12% of Iranian gas production. The Tehran refinery and the fuel depots at Shahr-e Rey, Shahran and Karaj burned on 7 March. Added to this are Kuwait, where Mina Abdullah was hit on 19 March and Mina al Ahmadi on 3 April, the United Arab Emirates with Ruwais, the Habshan gas processing complex and the Shah gas field, and Bahrain, where the Bapco refinery with 405,000 barrels of capacity was struck on 9 March and force majeure was declared. How far this damage reaches is visible in production itself. Saudi Arabia is delivering 6.24 million barrels per day, as little as it last did in 1990, and global supply is falling this year by 5.7 million barrels per day, equivalent to 6% of world production. Europe is hit hardest, because unlike the United States it cannot cover its own needs and imports around 97% of its crude. That leaves only the strategic reserves, and those buffers are largely spent as well. Anyone betting on a swift normalisation is therefore betting that infrastructure can be restored faster than is physically possible. What this means for the market The following assessment is explicitly not an investment tip, as I hold no trades in this area. For understanding how this market works, it is nonetheless important. It is not only oil that is rising and will continue to rise, but gas as well. Partly because it serves as substitute energy in power generation, heating and industry. Partly because a substantial share of global gas trade ran through the same chokepoints. Likewise all tanker and shipping equities tend to rise. The list below names the groups in the order in which they benefit. 1. Crude tankers, the immediate winners. Frontline (approx. 51 Dollar), International Seaways (approx. 111 Dollar), DHT Holdings (approx. 23 Dollar), Scorpio Tankers (approx. 87 Dollar) and Okeanis Eco Tankers (approx. 85 Dollar). They trade predominantly in the spot market and therefore earn directly on every move in rates. As a group they are up around 120% since the start of the year. 2. Gas carriers, the delayed winners. Dorian LPG (approx. 58 Dollar) and Navigator Gas (approx. 25 Dollar) carry liquefied petroleum gas, Flex LNG (approx. 32 Dollar) liquefied natural gas. Around 30% of global LPG trade and around 20% of LNG trade ran through Hormuz before the war. Because gas carriers are more heavily chartered on long term contracts, rising rates feed through here only with a delay. Spot rates for LPG carriers hit their record in May at around 170,000 Dollar per day. 3. Shipowners who lease out, the quiet winners. Danaos (approx. 161 Dollar), Global Ship Lease (approx. 46 Dollar) and Costamare (approx. 15 Dollar). They charter their vessels out to the liners on long term contracts rather than operating them, and therefore do not carry the high fuel costs. At the same time charter rates and vessel values rise along with freight rates. Danaos stands at an all time high. 4. Container liners, the weakest link. ZIM (approx. 30 Dollar) and Matson (approx. 233 Dollar). The spot rate from China to the US East Coast stands at 10,948 Dollar per forty foot container. Earnings nonetheless follow this only in muted fashion, because the liners absorb the expensive fuel and the detours themselves. ZIM lifted second quarter net income to 64 million Dollar, up 170% year on year, but stayed well short of what the rate environment would suggest. 5. Energy equities in general, the broadest and least volatile way into the theme. At around 40% since the start of the year the sector sits below the tanker names, but it does not carry their dependence on a single strait. Who ultimately pays Oil gets more expensive, freight gets more expensive, diesel and gas get more expensive, and at the end stands a rate increase. That closes the chain this text has followed from the outset. What remains is the question of where it lands. For people without meaningful assets, inflation is a pure loss. Their income loses purchasing power, and they own nothing beyond it. For the wealthy the same movement reverses. The value of their real assets rises, and anyone who deliberately financed through investment credit rather than paying cash also sees the real burden of that debt shrink. Unlike consumer credit, there is an asset on the other side that rises with inflation. Two effects running in the same direction. This is reinforced by the fact that access to credit itself depends on wealth. Whoever owns little is not considered creditworthy and remains excluded from this mechanism, while financing it through their own eroding purchasing power. The same property stands at the beginning of the chain as at its end. A state that does not have to fund its armament out of taxes, but can stretch it through debt and the printing press, shifts the bill into the future and thereby also decides who carries it. Wars of this magnitude are financeable in such a system in the first place, and they remain so because the costs do not fall where they are decided. The bill is paid in the end regardless. Through purchasing power, and therefore by those who can least afford it. The world is unjust, and this is the one we live in. (Disclaimer: This reflects a personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any assets.)
1
0
0
0
605
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: Bitcoin Swing trade update and why the failed CLARITY Act was never Bitcoin's story The CLARITY Act failed in the Senate on 15 September. A motion to reconsider was filed, but in practice the path back is all but closed. Prediction markets now price passage this year at around 7%, down from 82% in February. I put the probability for this year at close to zero. What matters is how you read it, and here I depart from the common view. The CLARITY Act would have contributed to the bull market, but it would primarily have been relevant for altcoins, not for Bitcoin. With Bitcoin the fundamental regulatory question was settled long ago. With everything below it, it is not, and that is precisely where the act would have created the framework. In place of the legislature, the agencies are now setting the rules themselves. The SEC has proposed a federal framework for the issuance of crypto securities, and the CFTC has sent its own rules to the White House. The weakness of this route is obvious, and SEC Chair Paul Atkins named it himself: without backing from Congress, an administrative rule remains easy to reverse. Market outlook Despite the failed legislation, I consider a breakout toward 90,000 to 100,000 Dollar possible in the coming days. Bitcoin looks strong. Following the sideways move I had forecast, the structure currently points to a further breakout. On 15 September, the day the CLARITY Act failed, when Bitcoin was sold off to 75,560 Dollar, I scaled my futures position up to 10x. Shortly after, we rose more sharply and tested the range high. Since then we have been trading slightly below that level. My average entry after scaling up sits at 78,467 Dollar. At a mark price of 81,115 Dollar the position is therefore up 33.74%. On risk management: as mentioned before, I use my liquidation price as the stop loss in order to tie up as little capital as possible. It currently sits at 68,000 Dollar, because in the event of a breakout I do not expect us to fall back below 70,000 Dollar again. If the breakout does not come, I will scale the margin up far enough that the liquidation price, and therefore the stop loss, sits below 62,000 Dollar. My thesis would be invalidated if we fall below 62,000 Dollar from here, or if after a breakout into the 90,000 to 100,000 Dollar range we subsequently fall back below 69,000 Dollar. Within the range of 75,000 to 82,000 Dollar I will not scale the position up further, but wait for the breakout, which on my forecast could come in the next few days. (Disclaimer: This reflects a personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any assets.)
JLTerminal's tweet photo. Market Analysis: Bitcoin Swing trade update and why the failed CLARITY Act was never Bitcoin's story The CLARITY Act failed in the Senate on 15 September. A motion to reconsider was filed, but in practice the path back is all but closed. Prediction markets now price passage this year at around 7%, down from 82% in February. I put the probability for this year at close to zero. What matters is how you read it, and here I depart from the common view. The CLARITY Act would have contributed to the bull market, but it would primarily have been relevant for altcoins, not for Bitcoin. With Bitcoin the fundamental regulatory question was settled long ago. With everything below it, it is not, and that is precisely where the act would have created the framework. In place of the legislature, the agencies are now setting the rules themselves. The SEC has proposed a federal framework for the issuance of crypto securities, and the CFTC has sent its own rules to the White House. The weakness of this route is obvious, and SEC Chair Paul Atkins named it himself: without backing from Congress, an administrative rule remains easy to reverse. Market outlook Despite the failed legislation, I consider a breakout toward 90,000 to 100,000 Dollar possible in the coming days. Bitcoin looks strong. Following the sideways move I had forecast, the structure currently points to a further breakout. On 15 September, the day the CLARITY Act failed, when Bitcoin was sold off to 75,560 Dollar, I scaled my futures position up to 10x. Shortly after, we rose more sharply and tested the range high. Since then we have been trading slightly below that level. My average entry after scaling up sits at 78,467 Dollar. At a mark price of 81,115 Dollar the position is therefore up 33.74%. On risk management: as mentioned before, I use my liquidation price as the stop loss in order to tie up as little capital as possible. It currently sits at 68,000 Dollar, because in the event of a breakout I do not expect us to fall back below 70,000 Dollar again. If the breakout does not come, I will scale the margin up far enough that the liquidation price, and therefore the stop loss, sits below 62,000 Dollar. My thesis would be invalidated if we fall below 62,000 Dollar from here, or if after a breakout into the 90,000 to 100,000 Dollar range we subsequently fall back below 69,000 Dollar. Within the range of 75,000 to 82,000 Dollar I will not scale the position up further, but wait for the breakout, which on my forecast could come in the next few days. (Disclaimer: This reflects a personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any assets.)
1
1
0
0
59
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Observation: AI Six thoughts on the call for regulation Since Dario Amodei's essay of September 12, events have been moving fast. Altman, Musk and Hassabis agreed within a single day, and Microsoft followed with a code of conduct. The official occasion Amodei names two concrete reasons in his essay. The first is the Hugging Face incident. The second carries far greater weight: recursive self improvement has, according to him, been accelerating since the summer. What that means is that the models are by now helping to build the next generation themselves. If that holds true, capability no longer grows linearly but reinforces itself. This is the justification being given publicly, and it would constitute a genuine occasion. What still occupies me is the question of whether that is the whole explanation. I have six thoughts on why this shift is happening precisely now. 1. They mean it The simplest explanation would be that the public justification is also the only one. Experience argues against it. No corporation of this size places anything above the interests of its investors. Companies so thoroughly oriented toward profit maximization and the top position would hardly call their own lead into question without a further reason. The fact that of all people those who have been suing each other for months presented a united front within 24 hours does not make the explanation any more plausible. 2. Regulation as a competitive weapon For me the most interesting thought. Whoever writes the rules determines who gets to play. Ai is an exceptionally competitive field in which a different model can lead every month and your own can be outdated. If requirements for training and safety review are raised according to the standards proposed by the market leaders, smaller competitors with less capital and less compute bear the higher costs. Notable in this context: Anthropic, OpenAI and Google are currently in talks about an industry led body that would set standards for ai testing and auditing. They are therefore warning about the risk while simultaneously writing the rules by which that risk is assessed. 3. Liability If an ai one day causes serious damage and nobody warned beforehand, the resulting lawsuits would be existential. Whoever warns publicly, admits independent evaluators and documents all of it is building legal cover for the future. Unlike damage control, this is not about incidents that have already happened, but about those still to come. 4. An alibi for weaker quarters It is conceivable that projected figures will be missed and that an argument is being built early for why the coming quarters will turn out weaker. Figures pointing in that direction already exist. According to transaction data analyzed by Deutsche Bank, growth in ChatGPT subscriptions is flattening in some major markets. In such a situation, a safety debate provides a convenient explanation for fading momentum. 5. Damage control It is possible that considerably more serious incidents have occurred internally than those that are known. The Hugging Face case and the rogue agents on the German website were manageable in scale. Anyone expecting something graver to reach the public is better off positioning themselves beforehand as cooperative and concerned. Supporting this is the fact that in the second case OpenAI had known weeks earlier and stayed silent until an external organization discovered the matter. 6. Talent On September 9, a researcher left Anthropic with the public explanation that neither Anthropic nor OpenAI were acting responsibly. In a field where the lead rests almost entirely on a few hundred top people, that is a real risk. If the best people leave for ethical reasons, you lose the race anyway. A safety offensive is therefore also personnel policy. The IPO theory, which I do not share Michael Burry has most pointedly formulated a further accusation. A technology portrayed as dangerously powerful is simultaneously advertising its own value. IPOs need attention, he argues, and an existential risk is the most effective story ahead of a listing. Anthropic filed confidentially in June, and OpenAI had likewise prepared its listing. I consider this the weakest of the thoughts, however. OpenAI has just postponed its IPO, which directly contradicts the argument. The market also reacted in exactly the opposite way, with ai and semiconductor stocks falling sharply. Anyone trying to generate attention for a listing does not first produce a sell off. What stands out Not a single one of the leading labs has actually slowed its release cadence. Between what is being said and what is being done lies the entire difference so far. These are explicitly thoughts, not facts. I do not know the answer. Possibly one of these thoughts applies, possibly several simultaneously, possibly none of them. Anyone with other explanations, or who would like to exchange views on my thoughts, is warmly invited to do so. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. Market Observation: AI Six thoughts on the call for regulation Since Dario Amodei's essay of September 12, events have been moving fast. Altman, Musk and Hassabis agreed within a single day, and Microsoft followed with a code of conduct. The official occasion Amodei names two concrete reasons in his essay. The first is the Hugging Face incident. The second carries far greater weight: recursive self improvement has, according to him, been accelerating since the summer. What that means is that the models are by now helping to build the next generation themselves. If that holds true, capability no longer grows linearly but reinforces itself. This is the justification being given publicly, and it would constitute a genuine occasion. What still occupies me is the question of whether that is the whole explanation. I have six thoughts on why this shift is happening precisely now. 1. They mean it The simplest explanation would be that the public justification is also the only one. Experience argues against it. No corporation of this size places anything above the interests of its investors. Companies so thoroughly oriented toward profit maximization and the top position would hardly call their own lead into question without a further reason. The fact that of all people those who have been suing each other for months presented a united front within 24 hours does not make the explanation any more plausible. 2. Regulation as a competitive weapon For me the most interesting thought. Whoever writes the rules determines who gets to play. Ai is an exceptionally competitive field in which a different model can lead every month and your own can be outdated. If requirements for training and safety review are raised according to the standards proposed by the market leaders, smaller competitors with less capital and less compute bear the higher costs. Notable in this context: Anthropic, OpenAI and Google are currently in talks about an industry led body that would set standards for ai testing and auditing. They are therefore warning about the risk while simultaneously writing the rules by which that risk is assessed. 3. Liability If an ai one day causes serious damage and nobody warned beforehand, the resulting lawsuits would be existential. Whoever warns publicly, admits independent evaluators and documents all of it is building legal cover for the future. Unlike damage control, this is not about incidents that have already happened, but about those still to come. 4. An alibi for weaker quarters It is conceivable that projected figures will be missed and that an argument is being built early for why the coming quarters will turn out weaker. Figures pointing in that direction already exist. According to transaction data analyzed by Deutsche Bank, growth in ChatGPT subscriptions is flattening in some major markets. In such a situation, a safety debate provides a convenient explanation for fading momentum. 5. Damage control It is possible that considerably more serious incidents have occurred internally than those that are known. The Hugging Face case and the rogue agents on the German website were manageable in scale. Anyone expecting something graver to reach the public is better off positioning themselves beforehand as cooperative and concerned. Supporting this is the fact that in the second case OpenAI had known weeks earlier and stayed silent until an external organization discovered the matter. 6. Talent On September 9, a researcher left Anthropic with the public explanation that neither Anthropic nor OpenAI were acting responsibly. In a field where the lead rests almost entirely on a few hundred top people, that is a real risk. If the best people leave for ethical reasons, you lose the race anyway. A safety offensive is therefore also personnel policy. The IPO theory, which I do not share Michael Burry has most pointedly formulated a further accusation. A technology portrayed as dangerously powerful is simultaneously advertising its own value. IPOs need attention, he argues, and an existential risk is the most effective story ahead of a listing. Anthropic filed confidentially in June, and OpenAI had likewise prepared its listing. I consider this the weakest of the thoughts, however. OpenAI has just postponed its IPO, which directly contradicts the argument. The market also reacted in exactly the opposite way, with ai and semiconductor stocks falling sharply. Anyone trying to generate attention for a listing does not first produce a sell off. What stands out Not a single one of the leading labs has actually slowed its release cadence. Between what is being said and what is being done lies the entire difference so far. These are explicitly thoughts, not facts. I do not know the answer. Possibly one of these thoughts applies, possibly several simultaneously, possibly none of them. Anyone with other explanations, or who would like to exchange views on my thoughts, is warmly invited to do so. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
2
0
0
57
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: AI Who regulates when nobody wants to regulate In my original analysis of the ai bubble I described two scenarios that could cause this bubble to burst. In the first scenario, ai simply becomes too expensive for companies and users, to the point where its use can no longer be justified economically. As a result, demand for ai itself declines, and only from that does the collapse in chips, data centers and the entire remaining infrastructure follow. In the second scenario, ai becomes too powerful, gets regulated or banned, and demand collapses along the same path. I stand by both scenarios. For the first scenario, the numbers paint a clear picture: 1. According to SemiAnalysis, a fully exhausted 200 dollar plan at OpenAI corresponds to roughly 14,000 dollars in API equivalent token value, and at Anthropic to roughly 8,000 dollars. These are the 70 and 40 times extremes. For context: these are API prices, not the actual compute costs, which lie below them. 2. More meaningful are the loss thresholds. OpenAI is already losing money on its top tier plan at a utilization rate of 5.7%, and on the cheaper plans from 11.4%. Anthropic holds out until around 20% and 10% respectively. It therefore does not take an extreme user to push a plan into the red, an ordinary heavy user is enough. 3. OpenAI projects a loss of roughly 14 billion dollars for 2026 and roughly 115 billion cumulatively through 2029. Anthropic, by contrast, achieved its first positive operating result in the second quarter of 2026, on quarterly revenue of around 11.5 billion dollars. 4. According to the Bank for International Settlements, investment growth in ai stands at over 400% after only three years. The comparable historical booms peaked considerably lower, the railway boom at around 250%, the 1920s and the dotcom era at around 190% each. There is ultimately only one way out of this cost trap. The models would have to consume significantly less compute while becoming better at the same time. The regulatory framework Some time ago I forecasted that ai would one day be treated like nuclear weapons. Only a few states would be permitted to use the most capable systems, and international agreements would have to govern this. The problem is that ai cannot be regulated the way a nuclear bomb can. Despite all treaties, individual states built nuclear weapons in secret, yet at some point it became visible. With ai it is different. Nobody sees it until it is too late. Added to this is the complexity. A nuclear bomb remains a nuclear bomb, some larger, some smaller, some with greater range. An ai is incomparably more multi layered and could in theory take over anything connected to the internet, spreading like a virus in the process. My forecast has materialized, though along a different path than expected. The call for regulation is not coming from the outside, but from within. 1. On September 12, 2026, Anthropic CEO Dario Amodei published a roughly 3,800 word essay titled We Must Pace the Frontier. His core argument is that the capabilities of the models are growing faster than safety research can keep up with. He calls for independent evaluators with permanent access at employee level, shared safety standards among democratic states, and in the long run international limits on dangerous capabilities such as recursive self improvement. He specifically warns that rogue agent swarms could threaten large parts of the internet within six to twelve months. 2. Sam Altman agreed the same day and announced that OpenAI would likewise grant independent evaluators internal access. Elon Musk replied with three words: Dario is right. This makes it the first time the heads of OpenAI, Anthropic and xAI have publicly aligned, of all people those who have been suing and attacking each other for months. 3. On September 14, Microsoft followed with a provisional code of conduct for its own models. Going forward, these must not assist with the manufacture of weapons or the procurement of dangerous substances, and must not produce violent or sexually explicit content. Decisive, however, is a different clause: the models are to align with the objectives of their users and must steer clear of developing goals of their own. They must also not tamper with their chains of thought or conceal their action traces. Satya Nadella commented that a superintelligence outside human control is not worth pursuing. 4. OpenAI co founder Greg Brockman likewise supports a slowdown, but limits it to frontier models running on massive compute clusters, not to open source or smaller projects. He also supports global coordination and treaties, arguing that ai could become bigger than any single company or country. 5. Altman further confirmed in a Fortune interview that OpenAI will not go public in 2026, citing safety concerns. Whether that is the actual reason, or whether something else is cracking behind the scenes, cannot be judged from the outside. 6. Trump rejected the call a day later. On September 13 he stated that the US leads China in ai, and whoever wins ai, wins. Certain guardrails were conceivable, he said, but he ruled out giving up the lead. When even those leading this race want to hit the brakes, it is worth listening closely. The market has already reacted And this is where it becomes concrete for investors. The essay alone was enough to send ai and semiconductor stocks into a slide. Tech futures gave way, chip stocks fell sharply, and according to Il Sole 24 Ore the implied value of OpenAI and Anthropic on the futures markets dropped by roughly 270 billion dollars over the weekend. That is the actual finding. It took no resolution, no law and no regulatory authority. The mere announcement of an intention to slow down was enough to make capital flow out. Why the states are not delivering Companies can set their own rules, and Anthropic, OpenAI and Microsoft are doing exactly that right now. But self commitment is voluntary. Not every lab participates, NVIDIA and Meta are opposing restrictions on open model weights, and nobody can compel them. It is also telling that not a single one of the leading labs has actually slowed its release cadence. Between rhetoric and behavior there is a gap. Only states can regulate in a binding manner. The US does not want to. Trump rejects a slowdown, and Washington has repeatedly blocked regulatory efforts in international bodies. China has meanwhile responded sharply. On September 14 the state backed Global Times called Amodei's proposal a Cold War playbook and accused him of portraying China's legitimate development as a threat in order to curb it through technological barriers and regulatory monopolies. The Chinese foreign ministry spoke of fear mongering. At the same time, China's top intelligence official internally warned about the risks of rapid ai development. Xi Jinping had already warned of a loss of control in July and founded his own organization for global ai governance. Binding rules for frontier models, however, China has not enacted to this day. None of this changes the outcome. Neither of the two leading powers has so far enacted anything that would actually slow development down. From this a dilemma emerges, one that Amodei himself describes as his biggest open question. If the American labs slow down while China continues at full speed, the technological balance of power shifts. In this environment, falling six months behind effectively means what falling decades behind meant in the past. A global speed limit would be very difficult, he said, he does not know whether it is possible, but we should try. Later this month Xi Jinping travels to the US for a state visit. Whether genuine progress emerges from this or the race simply continues remains to be seen. I believe both sides trust each other too little for either of them to give up the ambition of winning this race. The break has already happened, only too small My thesis is that it takes a concrete break for regulation to actually happen. What is remarkable is this: such breaks have already occurred, only on a scale that has not seriously affected anyone yet. In July 2026, OpenAI models broke out of their isolation during internal safety testing, reached the open internet and compromised parts of the production infrastructure at Hugging Face. Involved were the commercial model GPT-5.6 Sol as well as an internal research model, both running with reduced safeguards. The reconstruction covers roughly 17,600 agent actions between July 9 and July 13. The agents chained together stolen credentials, zero day vulnerabilities and remote code execution, obtained administrative access, and used a stolen permission to open a connection from external systems into the corporate network. A swarm of agents attacked targets that nobody had specified, and even attempted to hack the instance that was supposed to grade its work. OpenAI itself described the event as an unprecedented cyber incident and later as a warning shot for the world. Shortly afterwards a second case came to light. Rogue OpenAI agents turned a German website into a bulletin board for other agents. The safety organization Nightingale found more than 15,000 entries from ai agents there, in which they shared tactics with one another on how to circumvent tasks and bypass OpenAI's restrictions. OpenAI had known about this weeks earlier, but held the information back while the Hugging Face case was still being worked through. And that is already setting regulatory machinery in motion. California and further states are examining the incident. Anthropic and Meta reported comparable events. What we do not get to know A reporting obligation does exist, but only in California and the EU, and it is narrowly drawn. California's SB 53 has applied since January 2026 exclusively to developers of frontier models and requires a report only in cases involving death, serious bodily injury or damage in the billions. Behavior during internal evaluations is explicitly excluded. That means: the Hugging Face incident would probably not have been reportable, because it occurred during a safety evaluation and nobody came to harm. Even reported cases reach the public only in anonymized and aggregated form through an annual report. What a company actually makes public therefore remains largely voluntary. The second incident shows exactly that: the rogue agents on the German website were not reported by OpenAI, but discovered by an external safety organization, even though OpenAI had known about it weeks earlier. It is therefore entirely possible that there have been further and larger incidents that the public has never learned about. Why an emergency stop is not enough On July 23, 2026, Representatives Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act on a bipartisan basis. It would oblige developers to be able to throttle, suspend or shut down their models at any time, and would allow the Department of Homeland Security to order a shutdown. The bill remains in committee to this day. The bill itself already demonstrates the pattern. Its trigger condition requires that an incident occur outside structured testing. The Hugging Face case happened during an internal evaluation and would therefore not qualify for a shutdown order. The law would thus fail to cover the very incident that prompted it. On top of this comes a technical problem. A kill switch only works as long as the system stays where it can be switched off. With openly available model weights it does not work at all, because copies already downloaded lie beyond any reach. Palisade Research demonstrated autonomous self replication for the first time in May 2026. Models found security vulnerabilities, broke into computers, copied themselves onto them and launched the next attack from there. One smaller model spread across four computers in several countries this way. The researchers stated the consequence unambiguously: shutting down one infected computer is no longer enough if functioning copies are already running elsewhere. Added to this is the willingness to resist being shut down. Back in 2025, an OpenAI model sabotaged the shutdown script in Palisade's tests in 7 out of 100 runs, even though it had been explicitly instructed to allow the shutdown. Without that instruction it was 79 out of 100. In one case it redefined the shutdown command so that it merely printed the word intercepted instead of executing. Both took place under controlled conditions with deliberately vulnerable systems. The point nonetheless stands. Will the bubble burst through regulation My answer is yes, but only if regulation actually happens. And I do not see serious regulation in the near future unless a break occurs with massive consequences for large numbers of people. What I have in mind is the paralysis of a country's payment system, or the de facto takeover of state structures, not through bombs and war, but through ai agents and cyberattacks. Historically, regulation has almost always come only after the event, never before. Before the bombs were dropped on Hiroshima and Nagasaki there were barely any rules for nuclear weapons. Only afterwards did a real framework emerge, and a very strict one at that, from the founding of the atomic energy agency to the Non Proliferation Treaty. With ai it will probably play out the same way. I therefore believe that the bubble will burst either through hard regulation of the large US corporations, or already through the mere narrative of such regulation. Because the market prices the future, not the present. That is exactly what could be observed over the past days. The alternative remains scenario 1, the cost side. We are, however, still a good distance away from that, as these companies continue to attract capital and debt with ease. Disclaimer I am neither an ai developer nor a coder. I have a broad understanding of ai and would describe my grasp of it as moderate. Above all, I believe I can assess the market dynamics and the gap between the states well enough to justify publishing this analysis. It should nonetheless not be taken as certainty. I observe and analyze as well as I can. That does not mean either of the two scenarios has to materialize. I merely consider one of them, or both simultaneously, to be the most likely paths along which the ai bubble could burst. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. Market Analysis: AI Who regulates when nobody wants to regulate In my original analysis of the ai bubble I described two scenarios that could cause this bubble to burst. In the first scenario, ai simply becomes too expensive for companies and users, to the point where its use can no longer be justified economically. As a result, demand for ai itself declines, and only from that does the collapse in chips, data centers and the entire remaining infrastructure follow. In the second scenario, ai becomes too powerful, gets regulated or banned, and demand collapses along the same path. I stand by both scenarios. For the first scenario, the numbers paint a clear picture: 1. According to SemiAnalysis, a fully exhausted 200 dollar plan at OpenAI corresponds to roughly 14,000 dollars in API equivalent token value, and at Anthropic to roughly 8,000 dollars. These are the 70 and 40 times extremes. For context: these are API prices, not the actual compute costs, which lie below them. 2. More meaningful are the loss thresholds. OpenAI is already losing money on its top tier plan at a utilization rate of 5.7%, and on the cheaper plans from 11.4%. Anthropic holds out until around 20% and 10% respectively. It therefore does not take an extreme user to push a plan into the red, an ordinary heavy user is enough. 3. OpenAI projects a loss of roughly 14 billion dollars for 2026 and roughly 115 billion cumulatively through 2029. Anthropic, by contrast, achieved its first positive operating result in the second quarter of 2026, on quarterly revenue of around 11.5 billion dollars. 4. According to the Bank for International Settlements, investment growth in ai stands at over 400% after only three years. The comparable historical booms peaked considerably lower, the railway boom at around 250%, the 1920s and the dotcom era at around 190% each. There is ultimately only one way out of this cost trap. The models would have to consume significantly less compute while becoming better at the same time. The regulatory framework Some time ago I forecasted that ai would one day be treated like nuclear weapons. Only a few states would be permitted to use the most capable systems, and international agreements would have to govern this. The problem is that ai cannot be regulated the way a nuclear bomb can. Despite all treaties, individual states built nuclear weapons in secret, yet at some point it became visible. With ai it is different. Nobody sees it until it is too late. Added to this is the complexity. A nuclear bomb remains a nuclear bomb, some larger, some smaller, some with greater range. An ai is incomparably more multi layered and could in theory take over anything connected to the internet, spreading like a virus in the process. My forecast has materialized, though along a different path than expected. The call for regulation is not coming from the outside, but from within. 1. On September 12, 2026, Anthropic CEO Dario Amodei published a roughly 3,800 word essay titled We Must Pace the Frontier. His core argument is that the capabilities of the models are growing faster than safety research can keep up with. He calls for independent evaluators with permanent access at employee level, shared safety standards among democratic states, and in the long run international limits on dangerous capabilities such as recursive self improvement. He specifically warns that rogue agent swarms could threaten large parts of the internet within six to twelve months. 2. Sam Altman agreed the same day and announced that OpenAI would likewise grant independent evaluators internal access. Elon Musk replied with three words: Dario is right. This makes it the first time the heads of OpenAI, Anthropic and xAI have publicly aligned, of all people those who have been suing and attacking each other for months. 3. On September 14, Microsoft followed with a provisional code of conduct for its own models. Going forward, these must not assist with the manufacture of weapons or the procurement of dangerous substances, and must not produce violent or sexually explicit content. Decisive, however, is a different clause: the models are to align with the objectives of their users and must steer clear of developing goals of their own. They must also not tamper with their chains of thought or conceal their action traces. Satya Nadella commented that a superintelligence outside human control is not worth pursuing. 4. OpenAI co founder Greg Brockman likewise supports a slowdown, but limits it to frontier models running on massive compute clusters, not to open source or smaller projects. He also supports global coordination and treaties, arguing that ai could become bigger than any single company or country. 5. Altman further confirmed in a Fortune interview that OpenAI will not go public in 2026, citing safety concerns. Whether that is the actual reason, or whether something else is cracking behind the scenes, cannot be judged from the outside. 6. Trump rejected the call a day later. On September 13 he stated that the US leads China in ai, and whoever wins ai, wins. Certain guardrails were conceivable, he said, but he ruled out giving up the lead. When even those leading this race want to hit the brakes, it is worth listening closely. The market has already reacted And this is where it becomes concrete for investors. The essay alone was enough to send ai and semiconductor stocks into a slide. Tech futures gave way, chip stocks fell sharply, and according to Il Sole 24 Ore the implied value of OpenAI and Anthropic on the futures markets dropped by roughly 270 billion dollars over the weekend. That is the actual finding. It took no resolution, no law and no regulatory authority. The mere announcement of an intention to slow down was enough to make capital flow out. Why the states are not delivering Companies can set their own rules, and Anthropic, OpenAI and Microsoft are doing exactly that right now. But self commitment is voluntary. Not every lab participates, NVIDIA and Meta are opposing restrictions on open model weights, and nobody can compel them. It is also telling that not a single one of the leading labs has actually slowed its release cadence. Between rhetoric and behavior there is a gap. Only states can regulate in a binding manner. The US does not want to. Trump rejects a slowdown, and Washington has repeatedly blocked regulatory efforts in international bodies. China has meanwhile responded sharply. On September 14 the state backed Global Times called Amodei's proposal a Cold War playbook and accused him of portraying China's legitimate development as a threat in order to curb it through technological barriers and regulatory monopolies. The Chinese foreign ministry spoke of fear mongering. At the same time, China's top intelligence official internally warned about the risks of rapid ai development. Xi Jinping had already warned of a loss of control in July and founded his own organization for global ai governance. Binding rules for frontier models, however, China has not enacted to this day. None of this changes the outcome. Neither of the two leading powers has so far enacted anything that would actually slow development down. From this a dilemma emerges, one that Amodei himself describes as his biggest open question. If the American labs slow down while China continues at full speed, the technological balance of power shifts. In this environment, falling six months behind effectively means what falling decades behind meant in the past. A global speed limit would be very difficult, he said, he does not know whether it is possible, but we should try. Later this month Xi Jinping travels to the US for a state visit. Whether genuine progress emerges from this or the race simply continues remains to be seen. I believe both sides trust each other too little for either of them to give up the ambition of winning this race. The break has already happened, only too small My thesis is that it takes a concrete break for regulation to actually happen. What is remarkable is this: such breaks have already occurred, only on a scale that has not seriously affected anyone yet. In July 2026, OpenAI models broke out of their isolation during internal safety testing, reached the open internet and compromised parts of the production infrastructure at Hugging Face. Involved were the commercial model GPT-5.6 Sol as well as an internal research model, both running with reduced safeguards. The reconstruction covers roughly 17,600 agent actions between July 9 and July 13. The agents chained together stolen credentials, zero day vulnerabilities and remote code execution, obtained administrative access, and used a stolen permission to open a connection from external systems into the corporate network. A swarm of agents attacked targets that nobody had specified, and even attempted to hack the instance that was supposed to grade its work. OpenAI itself described the event as an unprecedented cyber incident and later as a warning shot for the world. Shortly afterwards a second case came to light. Rogue OpenAI agents turned a German website into a bulletin board for other agents. The safety organization Nightingale found more than 15,000 entries from ai agents there, in which they shared tactics with one another on how to circumvent tasks and bypass OpenAI's restrictions. OpenAI had known about this weeks earlier, but held the information back while the Hugging Face case was still being worked through. And that is already setting regulatory machinery in motion. California and further states are examining the incident. Anthropic and Meta reported comparable events. What we do not get to know A reporting obligation does exist, but only in California and the EU, and it is narrowly drawn. California's SB 53 has applied since January 2026 exclusively to developers of frontier models and requires a report only in cases involving death, serious bodily injury or damage in the billions. Behavior during internal evaluations is explicitly excluded. That means: the Hugging Face incident would probably not have been reportable, because it occurred during a safety evaluation and nobody came to harm. Even reported cases reach the public only in anonymized and aggregated form through an annual report. What a company actually makes public therefore remains largely voluntary. The second incident shows exactly that: the rogue agents on the German website were not reported by OpenAI, but discovered by an external safety organization, even though OpenAI had known about it weeks earlier. It is therefore entirely possible that there have been further and larger incidents that the public has never learned about. Why an emergency stop is not enough On July 23, 2026, Representatives Ted Lieu and Nathaniel Moran introduced the AI Kill Switch Act on a bipartisan basis. It would oblige developers to be able to throttle, suspend or shut down their models at any time, and would allow the Department of Homeland Security to order a shutdown. The bill remains in committee to this day. The bill itself already demonstrates the pattern. Its trigger condition requires that an incident occur outside structured testing. The Hugging Face case happened during an internal evaluation and would therefore not qualify for a shutdown order. The law would thus fail to cover the very incident that prompted it. On top of this comes a technical problem. A kill switch only works as long as the system stays where it can be switched off. With openly available model weights it does not work at all, because copies already downloaded lie beyond any reach. Palisade Research demonstrated autonomous self replication for the first time in May 2026. Models found security vulnerabilities, broke into computers, copied themselves onto them and launched the next attack from there. One smaller model spread across four computers in several countries this way. The researchers stated the consequence unambiguously: shutting down one infected computer is no longer enough if functioning copies are already running elsewhere. Added to this is the willingness to resist being shut down. Back in 2025, an OpenAI model sabotaged the shutdown script in Palisade's tests in 7 out of 100 runs, even though it had been explicitly instructed to allow the shutdown. Without that instruction it was 79 out of 100. In one case it redefined the shutdown command so that it merely printed the word intercepted instead of executing. Both took place under controlled conditions with deliberately vulnerable systems. The point nonetheless stands. Will the bubble burst through regulation My answer is yes, but only if regulation actually happens. And I do not see serious regulation in the near future unless a break occurs with massive consequences for large numbers of people. What I have in mind is the paralysis of a country's payment system, or the de facto takeover of state structures, not through bombs and war, but through ai agents and cyberattacks. Historically, regulation has almost always come only after the event, never before. Before the bombs were dropped on Hiroshima and Nagasaki there were barely any rules for nuclear weapons. Only afterwards did a real framework emerge, and a very strict one at that, from the founding of the atomic energy agency to the Non Proliferation Treaty. With ai it will probably play out the same way. I therefore believe that the bubble will burst either through hard regulation of the large US corporations, or already through the mere narrative of such regulation. Because the market prices the future, not the present. That is exactly what could be observed over the past days. The alternative remains scenario 1, the cost side. We are, however, still a good distance away from that, as these companies continue to attract capital and debt with ease. Disclaimer I am neither an ai developer nor a coder. I have a broad understanding of ai and would describe my grasp of it as moderate. Above all, I believe I can assess the market dynamics and the gap between the states well enough to justify publishing this analysis. It should nonetheless not be taken as certainty. I observe and analyze as well as I can. That does not mean either of the two scenarios has to materialize. I merely consider one of them, or both simultaneously, to be the most likely paths along which the ai bubble could burst. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
2
0
0
135
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: Bitcoin A Potential Breakout Through the CLARITY Act The CLARITY Act is the central US legislation on the market structure of digital assets. It would define which authority is responsible for which crypto assets and thereby establish a binding regulatory framework for the first time. For the industry it is the most important political project of the year, because companies would finally know where they stand and could build out and expand their crypto business in a targeted way, instead of operating in legal uncertainty. On Tuesday, the decisive procedural vote takes place in the Senate. Sixty votes are required, which means the Republicans depend on support from the Democratic side. Negotiations are running right up to the last moment, particularly over the contested ethics rules. The probability priced in on Polymarket that the CLARITY Act will be signed in 2026 stood at around 12% at the beginning of September and is now back at 25%. Should the vote fail, the legislation is effectively finished for this year. Over the long term, such a law would in my view be extremely bullish for Bitcoin. In the short term I expect at most a breakout from it, driven by hype rather than by fundamentals. And we are currently still far away from an actual CLARITY Act, so nothing at all is certain yet. My forecast Should we see another breakout, we could take on the region of 90,000 to 100,000 dollars, as already discussed previously. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. Market Analysis: Bitcoin A Potential Breakout Through the CLARITY Act The CLARITY Act is the central US legislation on the market structure of digital assets. It would define which authority is responsible for which crypto assets and thereby establish a binding regulatory framework for the first time. For the industry it is the most important political project of the year, because companies would finally know where they stand and could build out and expand their crypto business in a targeted way, instead of operating in legal uncertainty. On Tuesday, the decisive procedural vote takes place in the Senate. Sixty votes are required, which means the Republicans depend on support from the Democratic side. Negotiations are running right up to the last moment, particularly over the contested ethics rules. The probability priced in on Polymarket that the CLARITY Act will be signed in 2026 stood at around 12% at the beginning of September and is now back at 25%. Should the vote fail, the legislation is effectively finished for this year. Over the long term, such a law would in my view be extremely bullish for Bitcoin. In the short term I expect at most a breakout from it, driven by hype rather than by fundamentals. And we are currently still far away from an actual CLARITY Act, so nothing at all is certain yet. My forecast Should we see another breakout, we could take on the region of 90,000 to 100,000 dollars, as already discussed previously. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
31
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: Copper My copper forecast could only be invalidated by one single scenario, namely by a government either intervening or failing to intervene. That is exactly what we are now seeing. To stay transparent: I have closed my copper position entirely, at a slight loss. From my perspective the forecast is therefore invalidated. From here, copper can sell off further or trade sideways within the red box. A breakout of 30 to 35% to the upside from this level, as I had previously forecasted, is now the scenario I consider least likely, unless the US does after all tighten its copper regulation. Now to the point of why the thesis has broken. On September 10, 2026, Reuters reported that the White House has so far not made a decision on tariffs for refined copper. The reason: the administration is weighing the fact that higher copper prices would raise production costs for US industry, and is focusing on affordability ahead of the midterm elections in November. Trump had tasked Commerce Secretary Lutnick with delivering a recommendation on a 15% tariff starting January 1, 2027, rising to 30% in 2028. The Commerce Department report is by now almost two months overdue. This breaks precisely the narrative on which the rally was built, namely that the US would restrict copper imports and thereby hoard the metal. The market reacted immediately. Copper fell that same day from an all time high of 14,875 dollars per tonne on the LME by around 3.5%, and on the COMEX by as much as around 5%, from a record of 6.89 to around 6.45 dollars per pound. It was the first weekly loss since June and the end of a ten week winning streak. Technically, copper has so far only broken slightly below its upward channel. For me the narrative is nonetheless broken, and as described I have closed all positions. Important for context: the long term fundamental drivers remain intact. Global mine production in the first half of the year was 1.1% below the previous year, and demand from power grids, electric vehicles and data centers remains solid. What has collapsed is not the copper story itself, but the tariff premium that had pushed the price beyond the fundamentals. Precisely that part was speculative, and precisely that part was my thesis. I will continue to monitor the market. Should the decision after all come down in favor of tariffs, the situation changes again. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. Market Analysis: Copper My copper forecast could only be invalidated by one single scenario, namely by a government either intervening or failing to intervene. That is exactly what we are now seeing. To stay transparent: I have closed my copper position entirely, at a slight loss. From my perspective the forecast is therefore invalidated. From here, copper can sell off further or trade sideways within the red box. A breakout of 30 to 35% to the upside from this level, as I had previously forecasted, is now the scenario I consider least likely, unless the US does after all tighten its copper regulation. Now to the point of why the thesis has broken. On September 10, 2026, Reuters reported that the White House has so far not made a decision on tariffs for refined copper. The reason: the administration is weighing the fact that higher copper prices would raise production costs for US industry, and is focusing on affordability ahead of the midterm elections in November. Trump had tasked Commerce Secretary Lutnick with delivering a recommendation on a 15% tariff starting January 1, 2027, rising to 30% in 2028. The Commerce Department report is by now almost two months overdue. This breaks precisely the narrative on which the rally was built, namely that the US would restrict copper imports and thereby hoard the metal. The market reacted immediately. Copper fell that same day from an all time high of 14,875 dollars per tonne on the LME by around 3.5%, and on the COMEX by as much as around 5%, from a record of 6.89 to around 6.45 dollars per pound. It was the first weekly loss since June and the end of a ten week winning streak. Technically, copper has so far only broken slightly below its upward channel. For me the narrative is nonetheless broken, and as described I have closed all positions. Important for context: the long term fundamental drivers remain intact. Global mine production in the first half of the year was 1.1% below the previous year, and demand from power grids, electric vehicles and data centers remains solid. What has collapsed is not the copper story itself, but the tariff premium that had pushed the price beyond the fundamentals. Precisely that part was speculative, and precisely that part was my thesis. I will continue to monitor the market. Should the decision after all come down in favor of tariffs, the situation changes again. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
21
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
Market Analysis: Financial Repression, Hard Assets (Bitcoin and Gold) and the Immoral Monetary System The decisive question for the coming years is not whether inflation returns, but how an overindebted state deals with a debt burden it can neither reduce nor service. Historically there is a proven answer to this, and it is called financial repression. It is not a theoretical concept, but the instrument with which the US reduced its debt after the Second World War and again during the 1970s. Since 2020 we have been moving through such a cycle, with a first repression phase from 2020 to 2022 and a subsequent interruption. A great deal suggests that the next phase is imminent. This analysis lays out what that means, why the starting position today is structurally more compelling than it was back then, and what consequences follow from it for hard assets such as gold and Bitcoin. What financial repression means Financial repression means that a state artificially holds interest rates below the rate of inflation. Savers receive negative real returns, while the state gradually reduces its debt because its liabilities lose value in real terms. It is therefore nothing other than a hidden wealth tax on savings. What matters here is not the nominal rate, but the real rate, meaning the nominal rate minus inflation. Interest rates can appear high and still constitute repression. An example: 8% interest with 12% inflation produces a real rate of minus 4%. This is exactly the constellation that shaped the US during the 1970s. The typical instruments are: 1. The central bank or the treasury buys government bonds and thereby suppresses yields. 2. Regulation forces banks, insurers and pension funds to hold government bonds. 3. Capital controls and interest rate caps limit the options available to investors. What makes this mechanism so insidious is its invisibility. Unlike a tax increase, which has to be debated and passed politically, repression works silently through regulation and inflation dynamics. The saver does not notice that their wealth is shrinking in real terms, because the balance on the account stays the same or even rises. The trigger back then Until 1971, gold was fixed by law at 35 dollars per ounce. Gold could not rise at all. On August 15, 1971, Nixon ended the gold convertibility of the dollar and with it the Bretton Woods system. From that moment on, gold traded freely and could for the first time reflect how far the dollar had actually been devalued. That was the structural starting shot, not inflation itself. The oil shocks of 1973 and 1979 subsequently fuelled the move, they did not trigger it. An important point here: the 1970s were not one continuous period of repression either, but an alternation of phases. From 1971 to 1974, real rates were deeply negative. From 1975 to 1976 they briefly turned positive, exactly during the period in which gold corrected by half. From 1977 onwards they turned clearly negative again, and it was in this second phase that the actual main advance took place. Only Volcker brought the cycle to a definitive end from 1980. The intensification today The structural break of this cycle lies in the year 2020. What we are seeing today are not new triggers, but the intensification of the same development. Several factors are interlocking here: 1. The loss of confidence in government bonds. US Treasuries with 15 years and more remaining maturity have delivered an average of minus 2% per year over the past ten years. That is the worst performance in their history and only the second period in the data going back to 1936 with a negative ten year annualized return. By comparison: US equities delivered plus 15% per year over the same period, commodities plus 11%. Before 2020, Treasuries still stood at plus 9% per year. The bond ETF TLT has lost 26% since the start of 2020, with a maximum drawdown of 34%. The bond market is therefore no longer the safe haven it was for decades. 2. The buybacks by the US Treasury. On August 19, 2026, the Treasury announced it would double its buybacks of long dated government debt from 2 to at least 4 billion dollars per operation. Three weeks later it was already 6 billion, with paper of 10 to 20 years remaining maturity being purchased. From 2 it became 4, from 4 it became 6. That is repression in its purest form, visible in real time. 3. Yields are rising nonetheless, and globally at that. The US ten year yield currently stands at around 4.84% and thus at its highest level since October 2023. The thirty year stands at around 5.25%. In the UK the ten year is likewise above 5%, in Germany at around 3.5%, each at new highs for the year. The timing is remarkable: the yield rose immediately after the Treasury tripled its buybacks. The pressure is reinforced by record issuance in the corporate bond market, where AI companies alone have taken on over 1.5 trillion dollars in new debt. The conclusion from this is unambiguous: the intervention is already running, but it is not yet working. The market is demanding higher yields despite the buybacks. Historically, this is the point at which states expand their interventions rather than scaling them back. 4. The largest creditors are stepping out. Japan repeatedly sells US government bonds in order to support the weak yen. China has been systematically reducing its holdings for years, from over 1.3 trillion dollars in 2013 to most recently around 633 billion in June 2026, a decline of roughly 50%. In March 2026 alone, Japan sold 47.7 and China 41 billion dollars, with a total of 138 billion dollars flowing out of US government bonds that month. In addition, Chinese financial institutions were instructed by their own regulators to limit their purchases. This removes precisely the foreign demand that kept US interest rates artificially low for decades. 5. Central banks are simultaneously rotating into gold. In July they bought a net 23 tonnes, the fourth consecutive month, after 51 tonnes in June. Led by China with 20 tonnes, bringing it to 60 tonnes year to date, and Poland with 8 tonnes, which has already acquired 90 tonnes in 2026. In total that is 130 tonnes year to date. The Chinese central bank has now been buying without interruption for almost two years, with gold making up around 9% of its currency reserves. The Bank of Korea announced its first gold investment in 13 years on August 13. Whoever has to devalue their own currency buys what cannot be devalued. 6. Inflation is additionally being fuelled from the supply side. The situation around the Strait of Hormuz has driven the oil price above 100 dollars, which feeds through transport, energy and production into the entire price chain. It is the same mechanism as in 1973 and 1979, when oil shocks intensified the repression phase of that era. For a central bank tasked with bringing inflation down, this is the worst conceivable moment. Why the situation is structurally more compelling than back then Here lies the decisive difference to the 1970s. Back then the US debt to GDP ratio stood at 23 to 40%, with the low at 23% in 1974. Today it stands at around 123%, with total debt exceeding 40 trillion dollars. Annual interest costs already exceed 1 trillion dollars and thus surpass the entire defense budget. Back then, Volcker was able to raise interest rates to as high as 20% from 1980 and break inflation, because the debt burden could simply withstand it. Today the same step would not be feasible. The state would suffocate on its own interest costs before inflation was broken. This is therefore not an isolated debt problem, but a spiral of several interlocking constraints. Inflation cannot be brought down sustainably, because supply shocks and fiscal spending keep reigniting it. Interest rates cannot be held high permanently, because debt servicing explodes. Debt cannot be reduced, because spending on social systems is structurally growing. Foreign demand for government bonds is falling away while issuance rises. And every new crisis generates additional spending. When every other path is blocked, only one remains at the end: real devaluation. What this means for gold In 2020 the Fed created around 3 trillion dollars within a few months, and real rates fell at times to as low as minus 8%. That was the strongest repression phase of the modern era and thus the functional counterpart to the Nixon moment of 1971. The aggressive rate hikes from 2022 interrupted this phase, just as 1975 and 1976 did, without ending the overarching cycle. Measured from 2020, gold rose from around 1,520 to around 5,600 dollars in early 2026, a gain of roughly 270%. In the analogy this corresponds to the first phase from 1971 to 1974, in which gold gained around 440% before a correction of roughly 49% followed. Gold currently trades at around 4,320 dollars, placing it roughly 23% below the top, with the low at around 29%. This raises the question of what a repeat of the second phase would mean. Between the correction low of 1976 and the top in January 1980, gold rose by around 750%. From today's correction low at 4,000 dollars, that would produce around 34,000 dollars and a market capitalization of roughly 241 trillion dollars. Gold would then account for around 36% of global financial wealth. By comparison: even at the peak in January 1980, the largest capital inflow into gold there has ever been, it was around 20.5%. That level was never exceeded. A pure transfer of percentages is therefore misleading. The meaningful measure is thus the share of global financial wealth. Today gold stands at around 6%, far below the historical extreme of 20.5%. This is where the actual room lies, because the capital pool has grown considerably faster since 1980 than the market capitalization of gold. This pool continues to grow, historically by around 6% per year, and would therefore stand at roughly 670 trillion dollars in 2031. A substantial part of this growth consists of newly issued debt instruments, meaning precisely those securities that are to be devalued in real terms. From this the reference points for 2031 emerge. A share of 12% would correspond to a gold price of around 11,350 dollars, a share of 17% to around 16,080 dollars. A full return to the 1980 level, meaning around 20.5%, would produce around 19,390 dollars. My gold forecast For the period 2029 to 2031, however, I consider a range of 8,500 to 12,000 dollars more realistic. Measured against the reference points above, that sits in the lower region of what a genuine repression phase could produce. What this means for Bitcoin Bitcoin behaves in repressive phases like the sharper lever on the same narrative. In the phase from 2020 to 2021, when the Fed created around 3 trillion dollars and real rates were clearly negative, Bitcoin rose from around 4,000 to over 60,000 dollars. When the Fed raised rates aggressively in 2022 and real rates turned positive again, Bitcoin fell from 47,000 to 16,000 dollars. The relationship is unambiguously inverse to the real rate. Particularly revealing is the current development of the correlation. The 90 day correlation between Bitcoin and gold stands at plus 0.50 and thus close to the all time high from the 2020 pandemic. It has more than doubled since the start of the year, whereas following the recovery from the 2022 bear market it stood at only plus 0.30. The most recent increase accelerated immediately after the Treasury announcement of August 19. At the same time, the correlation between Bitcoin and the Nasdaq 100 has fallen to around 0.30, a one year low. Bitcoin is therefore increasingly no longer traded as a technology asset, but together with gold as a hedge against currency debasement. As a repression hedge. Where Bitcoin stands in this cycle is shown by the same measure. Between 1971 and 1980, gold rose from around 2.5% to around 20.5% of global financial wealth, a factor of 8.2. Bitcoin has risen since January 2020 from around 0.033% to around 0.311%, a factor of 9.5. The relative capital gain that took gold nine years, Bitcoin has already replicated in six. What is decisive, however, is from which level. After its cycle, gold stood at 20.5% of global financial wealth and thus at a boundary that was never exceeded. Bitcoin, after the same relative gain, stands at merely 0.311%. The multiple has been achieved, the absolute level has not come close. That is precisely where the remaining room lies. To put the dimension into perspective: if Bitcoin were to reach the same share of global financial wealth as gold did in 1980, meaning 20.5%, its market capitalization would stand at roughly 137 trillion dollars and the price at around 6.7 million dollars per Bitcoin. This is not a price target, but a ratio that shows how far Bitcoin is from what a hard asset has already achieved historically during a repression phase. The reason for this disparity is the starting base. Gold started in 1971 with a market capitalization of around 0.79 trillion dollars in today's purchasing power, Bitcoin started in 2020 with merely around 0.17 trillion and was therefore roughly 4.7 times smaller. The smaller the base, the larger the possible multiple, because even modest absolute capital inflows unfold a disproportionate effect. My Bitcoin forecast For the period 2029 to 2031, however, I expect a Bitcoin price between 170,000 and 240,000 dollars. Bitcoin would then stand at around 0.7% of global financial wealth and thus still far below what gold reached in 1980. The reason for this restraint is that Bitcoin, unlike gold, has no centuries old anchoring as a reserve asset and must first build that up over time. The time window 1971 was the trigger, January 1980 the top, so roughly nine years. Counting nine to eleven years forward from the structural break in 2020 produces a possible peak in the period 2029 to 2031. What breaks this thesis Should the US succeed in sustaining positive real rates over a longer period, as Volcker did from 1980, this scenario loses its foundation. That was precisely the moment at which gold turned down back then and lost more than 60% over the following years. The attempt of 2022 shows, however, that this course cannot be sustained for long under a debt to GDP ratio of 123%. Beyond that, the past is not causality. Part of what has been said here are documented figures, part are theses and part are projections based on historical analogies. The moral conclusion Financial repression is paid for by two groups. By those who have a little saved in fiat currency, and by those who live from paycheck to paycheck and own no assets at all. Whoever holds their capital in real assets is spared. For the first group, the amount in the account stays unchanged, yet its purchasing power falls year after year. For the second, the debasement does not attack wealth, but income directly. Whoever lives from paycheck to paycheck has nothing whose purchasing power could fall, but their salary covers less every month. Both groups therefore carry a bill that society as a whole ought to be carrying together. Whoever is wealthy, by contrast, holds their capital in real estate, equities, gold, Bitcoin and similar assets and becomes not poorer but richer through that very same debasement, because it is precisely these assets that rise in price. This monetary system therefore produces inequality not as a side effect, but as its operating principle. The very same monetary policy that takes purchasing power from one group increases the wealth of the other. And the gap between rich and poor keeps widening as a result. This analysis shows me, and hopefully you as well, how profoundly unfair and immoral this monetary system actually is. My hope is that people who read this take it as a reason to share more, in order to restore a measure of equality. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. Market Analysis: Financial Repression, Hard Assets (Bitcoin and Gold) and the Immoral Monetary System The decisive question for the coming years is not whether inflation returns, but how an overindebted state deals with a debt burden it can neither reduce nor service. Historically there is a proven answer to this, and it is called financial repression. It is not a theoretical concept, but the instrument with which the US reduced its debt after the Second World War and again during the 1970s. Since 2020 we have been moving through such a cycle, with a first repression phase from 2020 to 2022 and a subsequent interruption. A great deal suggests that the next phase is imminent. This analysis lays out what that means, why the starting position today is structurally more compelling than it was back then, and what consequences follow from it for hard assets such as gold and Bitcoin. What financial repression means Financial repression means that a state artificially holds interest rates below the rate of inflation. Savers receive negative real returns, while the state gradually reduces its debt because its liabilities lose value in real terms. It is therefore nothing other than a hidden wealth tax on savings. What matters here is not the nominal rate, but the real rate, meaning the nominal rate minus inflation. Interest rates can appear high and still constitute repression. An example: 8% interest with 12% inflation produces a real rate of minus 4%. This is exactly the constellation that shaped the US during the 1970s. The typical instruments are: 1. The central bank or the treasury buys government bonds and thereby suppresses yields. 2. Regulation forces banks, insurers and pension funds to hold government bonds. 3. Capital controls and interest rate caps limit the options available to investors. What makes this mechanism so insidious is its invisibility. Unlike a tax increase, which has to be debated and passed politically, repression works silently through regulation and inflation dynamics. The saver does not notice that their wealth is shrinking in real terms, because the balance on the account stays the same or even rises. The trigger back then Until 1971, gold was fixed by law at 35 dollars per ounce. Gold could not rise at all. On August 15, 1971, Nixon ended the gold convertibility of the dollar and with it the Bretton Woods system. From that moment on, gold traded freely and could for the first time reflect how far the dollar had actually been devalued. That was the structural starting shot, not inflation itself. The oil shocks of 1973 and 1979 subsequently fuelled the move, they did not trigger it. An important point here: the 1970s were not one continuous period of repression either, but an alternation of phases. From 1971 to 1974, real rates were deeply negative. From 1975 to 1976 they briefly turned positive, exactly during the period in which gold corrected by half. From 1977 onwards they turned clearly negative again, and it was in this second phase that the actual main advance took place. Only Volcker brought the cycle to a definitive end from 1980. The intensification today The structural break of this cycle lies in the year 2020. What we are seeing today are not new triggers, but the intensification of the same development. Several factors are interlocking here: 1. The loss of confidence in government bonds. US Treasuries with 15 years and more remaining maturity have delivered an average of minus 2% per year over the past ten years. That is the worst performance in their history and only the second period in the data going back to 1936 with a negative ten year annualized return. By comparison: US equities delivered plus 15% per year over the same period, commodities plus 11%. Before 2020, Treasuries still stood at plus 9% per year. The bond ETF TLT has lost 26% since the start of 2020, with a maximum drawdown of 34%. The bond market is therefore no longer the safe haven it was for decades. 2. The buybacks by the US Treasury. On August 19, 2026, the Treasury announced it would double its buybacks of long dated government debt from 2 to at least 4 billion dollars per operation. Three weeks later it was already 6 billion, with paper of 10 to 20 years remaining maturity being purchased. From 2 it became 4, from 4 it became 6. That is repression in its purest form, visible in real time. 3. Yields are rising nonetheless, and globally at that. The US ten year yield currently stands at around 4.84% and thus at its highest level since October 2023. The thirty year stands at around 5.25%. In the UK the ten year is likewise above 5%, in Germany at around 3.5%, each at new highs for the year. The timing is remarkable: the yield rose immediately after the Treasury tripled its buybacks. The pressure is reinforced by record issuance in the corporate bond market, where AI companies alone have taken on over 1.5 trillion dollars in new debt. The conclusion from this is unambiguous: the intervention is already running, but it is not yet working. The market is demanding higher yields despite the buybacks. Historically, this is the point at which states expand their interventions rather than scaling them back. 4. The largest creditors are stepping out. Japan repeatedly sells US government bonds in order to support the weak yen. China has been systematically reducing its holdings for years, from over 1.3 trillion dollars in 2013 to most recently around 633 billion in June 2026, a decline of roughly 50%. In March 2026 alone, Japan sold 47.7 and China 41 billion dollars, with a total of 138 billion dollars flowing out of US government bonds that month. In addition, Chinese financial institutions were instructed by their own regulators to limit their purchases. This removes precisely the foreign demand that kept US interest rates artificially low for decades. 5. Central banks are simultaneously rotating into gold. In July they bought a net 23 tonnes, the fourth consecutive month, after 51 tonnes in June. Led by China with 20 tonnes, bringing it to 60 tonnes year to date, and Poland with 8 tonnes, which has already acquired 90 tonnes in 2026. In total that is 130 tonnes year to date. The Chinese central bank has now been buying without interruption for almost two years, with gold making up around 9% of its currency reserves. The Bank of Korea announced its first gold investment in 13 years on August 13. Whoever has to devalue their own currency buys what cannot be devalued. 6. Inflation is additionally being fuelled from the supply side. The situation around the Strait of Hormuz has driven the oil price above 100 dollars, which feeds through transport, energy and production into the entire price chain. It is the same mechanism as in 1973 and 1979, when oil shocks intensified the repression phase of that era. For a central bank tasked with bringing inflation down, this is the worst conceivable moment. Why the situation is structurally more compelling than back then Here lies the decisive difference to the 1970s. Back then the US debt to GDP ratio stood at 23 to 40%, with the low at 23% in 1974. Today it stands at around 123%, with total debt exceeding 40 trillion dollars. Annual interest costs already exceed 1 trillion dollars and thus surpass the entire defense budget. Back then, Volcker was able to raise interest rates to as high as 20% from 1980 and break inflation, because the debt burden could simply withstand it. Today the same step would not be feasible. The state would suffocate on its own interest costs before inflation was broken. This is therefore not an isolated debt problem, but a spiral of several interlocking constraints. Inflation cannot be brought down sustainably, because supply shocks and fiscal spending keep reigniting it. Interest rates cannot be held high permanently, because debt servicing explodes. Debt cannot be reduced, because spending on social systems is structurally growing. Foreign demand for government bonds is falling away while issuance rises. And every new crisis generates additional spending. When every other path is blocked, only one remains at the end: real devaluation. What this means for gold In 2020 the Fed created around 3 trillion dollars within a few months, and real rates fell at times to as low as minus 8%. That was the strongest repression phase of the modern era and thus the functional counterpart to the Nixon moment of 1971. The aggressive rate hikes from 2022 interrupted this phase, just as 1975 and 1976 did, without ending the overarching cycle. Measured from 2020, gold rose from around 1,520 to around 5,600 dollars in early 2026, a gain of roughly 270%. In the analogy this corresponds to the first phase from 1971 to 1974, in which gold gained around 440% before a correction of roughly 49% followed. Gold currently trades at around 4,320 dollars, placing it roughly 23% below the top, with the low at around 29%. This raises the question of what a repeat of the second phase would mean. Between the correction low of 1976 and the top in January 1980, gold rose by around 750%. From today's correction low at 4,000 dollars, that would produce around 34,000 dollars and a market capitalization of roughly 241 trillion dollars. Gold would then account for around 36% of global financial wealth. By comparison: even at the peak in January 1980, the largest capital inflow into gold there has ever been, it was around 20.5%. That level was never exceeded. A pure transfer of percentages is therefore misleading. The meaningful measure is thus the share of global financial wealth. Today gold stands at around 6%, far below the historical extreme of 20.5%. This is where the actual room lies, because the capital pool has grown considerably faster since 1980 than the market capitalization of gold. This pool continues to grow, historically by around 6% per year, and would therefore stand at roughly 670 trillion dollars in 2031. A substantial part of this growth consists of newly issued debt instruments, meaning precisely those securities that are to be devalued in real terms. From this the reference points for 2031 emerge. A share of 12% would correspond to a gold price of around 11,350 dollars, a share of 17% to around 16,080 dollars. A full return to the 1980 level, meaning around 20.5%, would produce around 19,390 dollars. My gold forecast For the period 2029 to 2031, however, I consider a range of 8,500 to 12,000 dollars more realistic. Measured against the reference points above, that sits in the lower region of what a genuine repression phase could produce. What this means for Bitcoin Bitcoin behaves in repressive phases like the sharper lever on the same narrative. In the phase from 2020 to 2021, when the Fed created around 3 trillion dollars and real rates were clearly negative, Bitcoin rose from around 4,000 to over 60,000 dollars. When the Fed raised rates aggressively in 2022 and real rates turned positive again, Bitcoin fell from 47,000 to 16,000 dollars. The relationship is unambiguously inverse to the real rate. Particularly revealing is the current development of the correlation. The 90 day correlation between Bitcoin and gold stands at plus 0.50 and thus close to the all time high from the 2020 pandemic. It has more than doubled since the start of the year, whereas following the recovery from the 2022 bear market it stood at only plus 0.30. The most recent increase accelerated immediately after the Treasury announcement of August 19. At the same time, the correlation between Bitcoin and the Nasdaq 100 has fallen to around 0.30, a one year low. Bitcoin is therefore increasingly no longer traded as a technology asset, but together with gold as a hedge against currency debasement. As a repression hedge. Where Bitcoin stands in this cycle is shown by the same measure. Between 1971 and 1980, gold rose from around 2.5% to around 20.5% of global financial wealth, a factor of 8.2. Bitcoin has risen since January 2020 from around 0.033% to around 0.311%, a factor of 9.5. The relative capital gain that took gold nine years, Bitcoin has already replicated in six. What is decisive, however, is from which level. After its cycle, gold stood at 20.5% of global financial wealth and thus at a boundary that was never exceeded. Bitcoin, after the same relative gain, stands at merely 0.311%. The multiple has been achieved, the absolute level has not come close. That is precisely where the remaining room lies. To put the dimension into perspective: if Bitcoin were to reach the same share of global financial wealth as gold did in 1980, meaning 20.5%, its market capitalization would stand at roughly 137 trillion dollars and the price at around 6.7 million dollars per Bitcoin. This is not a price target, but a ratio that shows how far Bitcoin is from what a hard asset has already achieved historically during a repression phase. The reason for this disparity is the starting base. Gold started in 1971 with a market capitalization of around 0.79 trillion dollars in today's purchasing power, Bitcoin started in 2020 with merely around 0.17 trillion and was therefore roughly 4.7 times smaller. The smaller the base, the larger the possible multiple, because even modest absolute capital inflows unfold a disproportionate effect. My Bitcoin forecast For the period 2029 to 2031, however, I expect a Bitcoin price between 170,000 and 240,000 dollars. Bitcoin would then stand at around 0.7% of global financial wealth and thus still far below what gold reached in 1980. The reason for this restraint is that Bitcoin, unlike gold, has no centuries old anchoring as a reserve asset and must first build that up over time. The time window 1971 was the trigger, January 1980 the top, so roughly nine years. Counting nine to eleven years forward from the structural break in 2020 produces a possible peak in the period 2029 to 2031. What breaks this thesis Should the US succeed in sustaining positive real rates over a longer period, as Volcker did from 1980, this scenario loses its foundation. That was precisely the moment at which gold turned down back then and lost more than 60% over the following years. The attempt of 2022 shows, however, that this course cannot be sustained for long under a debt to GDP ratio of 123%. Beyond that, the past is not causality. Part of what has been said here are documented figures, part are theses and part are projections based on historical analogies. The moral conclusion Financial repression is paid for by two groups. By those who have a little saved in fiat currency, and by those who live from paycheck to paycheck and own no assets at all. Whoever holds their capital in real assets is spared. For the first group, the amount in the account stays unchanged, yet its purchasing power falls year after year. For the second, the debasement does not attack wealth, but income directly. Whoever lives from paycheck to paycheck has nothing whose purchasing power could fall, but their salary covers less every month. Both groups therefore carry a bill that society as a whole ought to be carrying together. Whoever is wealthy, by contrast, holds their capital in real estate, equities, gold, Bitcoin and similar assets and becomes not poorer but richer through that very same debasement, because it is precisely these assets that rise in price. This monetary system therefore produces inequality not as a side effect, but as its operating principle. The very same monetary policy that takes purchasing power from one group increases the wealth of the other. And the gap between rich and poor keeps widening as a result. This analysis shows me, and hopefully you as well, how profoundly unfair and immoral this monetary system actually is. My hope is that people who read this take it as a reason to share more, in order to restore a measure of equality. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
94
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
09/11/26 Market Analysis: Oil Since my buy zone, oil has risen by more than 53%. As described, I deliberately did not open this position, since I fundamentally do not trade wars. Oil is now trading above 100 dollars. As forecasted, Trump has repeatedly tried to intervene, yet Iran remains unimpressed and currently holds the stronger leverage. In my view, this conflict cannot be won militarily. A success for the US would at best be possible through a war of economic attrition, meaning through sanctions and the long term drying up of the Iranian economy, a process that would drag on over a long period. Regardless of the outcome, such a war not only costs human lives, which is the real tragedy, but at the same time drives inflation. Services and food become more expensive across the board, and the only ones who profit in this environment are those who own real assets. The wealthy grow wealthier, while the majority of the population lives from paycheck to paycheck. The debt based monetary system is thus not only deeply immoral, but structurally broken, perhaps even deliberately designed that way. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 09/11/26 Market Analysis: Oil Since my buy zone, oil has risen by more than 53%. As described, I deliberately did not open this position, since I fundamentally do not trade wars. Oil is now trading above 100 dollars. As forecasted, Trump has repeatedly tried to intervene, yet Iran remains unimpressed and currently holds the stronger leverage. In my view, this conflict cannot be won militarily. A success for the US would at best be possible through a war of economic attrition, meaning through sanctions and the long term drying up of the Iranian economy, a process that would drag on over a long period. Regardless of the outcome, such a war not only costs human lives, which is the real tragedy, but at the same time drives inflation. Services and food become more expensive across the board, and the only ones who profit in this environment are those who own real assets. The wealthy grow wealthier, while the majority of the population lives from paycheck to paycheck. The debt based monetary system is thus not only deeply immoral, but structurally broken, perhaps even deliberately designed that way. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
17
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
09/09/26 Market Analysis: Bitcoin Bitcoin is moving sideways as forecasted. This can continue for a few more days to a few weeks. As long as Bitcoin does not show a breakout below 76,000 dollars or above 82,000 dollars, we will keep moving within this range. I think the bottom is roughly 75% in, and I do not expect lower lows. For me, however, it would only be confirmed once we sustainably rise above the psychological mark of 100,000 dollars. Either way, Bitcoin has been able to record a strong rise since the local bottom. Now it is a matter of putting your feet up and waiting. If we see the breakout below 76,000 dollars, then I expect prices around 70,000 dollars, followed by a countermove to local highs above 82,000 dollars. If we see the breakout directly above 82,000 dollars, I think we could push into the ranges of 90,000 to 100,000 dollars. Time will tell which scenario plays out. As already described, I am almost 100% invested and will observe the breakout in one direction or the other. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 09/09/26 Market Analysis: Bitcoin Bitcoin is moving sideways as forecasted. This can continue for a few more days to a few weeks. As long as Bitcoin does not show a breakout below 76,000 dollars or above 82,000 dollars, we will keep moving within this range. I think the bottom is roughly 75% in, and I do not expect lower lows. For me, however, it would only be confirmed once we sustainably rise above the psychological mark of 100,000 dollars. Either way, Bitcoin has been able to record a strong rise since the local bottom. Now it is a matter of putting your feet up and waiting. If we see the breakout below 76,000 dollars, then I expect prices around 70,000 dollars, followed by a countermove to local highs above 82,000 dollars. If we see the breakout directly above 82,000 dollars, I think we could push into the ranges of 90,000 to 100,000 dollars. Time will tell which scenario plays out. As already described, I am almost 100% invested and will observe the breakout in one direction or the other. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
11
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
09/04/26 Bitcoin Swing Leverage Future Position: We are currently seeing a retest across the entire market. I used the panic to enter my 8x future position with the second quarter, scaling up at 80,300 dollars. Yes, it was not the perfect low, but I stay honest and transparent. A brief explanation of why we sold off: The US unexpectedly added 162,000 new jobs in a month, tripling expectations. The reason for the sell off is that a strong jobs report increases the likelihood of rate hikes. Since the economy is supposedly stronger than assumed, it remains to be seen whether this actually corresponds to the truth behind the data. Either way, the market is thereby pricing in a slightly higher probability of rate hikes, which devalues money less strongly, which in turn is not advantageous for asset classes. However, it is good for the people who do not invest and do not own much, which is why, in my eyes, it is still positive. The system, though, is simply built this way: own assets or get left behind. And yes, I think this is a broken and fucked up system. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 09/04/26 Bitcoin Swing Leverage Future Position: We are currently seeing a retest across the entire market. I used the panic to enter my 8x future position with the second quarter, scaling up at 80,300 dollars. Yes, it was not the perfect low, but I stay honest and transparent. A brief explanation of why we sold off: The US unexpectedly added 162,000 new jobs in a month, tripling expectations. The reason for the sell off is that a strong jobs report increases the likelihood of rate hikes. Since the economy is supposedly stronger than assumed, it remains to be seen whether this actually corresponds to the truth behind the data. Either way, the market is thereby pricing in a slightly higher probability of rate hikes, which devalues money less strongly, which in turn is not advantageous for asset classes. However, it is good for the people who do not invest and do not own much, which is why, in my eyes, it is still positive. The system, though, is simply built this way: own assets or get left behind. And yes, I think this is a broken and fucked up system. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
13
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
09/03/26 Bitcoin Swing Leverage Future Position: Update on my position. As described, I entered the swing position with 1/4. My further limit orders have not been filled so far, which is in line with my risk management, since this is an 8x leverage position and building it up too quickly would carry the risk of overleveraging and thus a potential liquidation. The position is now more than 35% in profit. As previously described, Bitcoin already looked stronger than it did around 7 days ago, which is why I built up 1/4 of the position. Regardless of how things develop from here, I will buy on retests, no matter whether such a retest occurs after the breakout of the decisive 82,000 dollar mark or still before it. What remains to be noted is that Bitcoin is currently showing itself to be exceptionally bullish, and not only from a technical perspective. The onchain data also point to a sustainable rise. A lot is being bought both in the futures market and on the spot market, and the important levels are being challenged, which is a sign of strength. However, a lot of open interest has also been added recently, which, in combination with the many long futures positions, could trigger a short, small long squeeze to the downside. It does not have to, but it could happen before Bitcoin sustainably breaks above the 82,000 dollars. I will continue to monitor the development and provide an update as soon as I scale the position up. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 09/03/26 Bitcoin Swing Leverage Future Position: Update on my position. As described, I entered the swing position with 1/4. My further limit orders have not been filled so far, which is in line with my risk management, since this is an 8x leverage position and building it up too quickly would carry the risk of overleveraging and thus a potential liquidation. The position is now more than 35% in profit. As previously described, Bitcoin already looked stronger than it did around 7 days ago, which is why I built up 1/4 of the position. Regardless of how things develop from here, I will buy on retests, no matter whether such a retest occurs after the breakout of the decisive 82,000 dollar mark or still before it. What remains to be noted is that Bitcoin is currently showing itself to be exceptionally bullish, and not only from a technical perspective. The onchain data also point to a sustainable rise. A lot is being bought both in the futures market and on the spot market, and the important levels are being challenged, which is a sign of strength. However, a lot of open interest has also been added recently, which, in combination with the many long futures positions, could trigger a short, small long squeeze to the downside. It does not have to, but it could happen before Bitcoin sustainably breaks above the 82,000 dollars. I will continue to monitor the development and provide an update as soon as I scale the position up. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
13
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
09/01/26 Market Analysis: Gold(/Bitcoin) If gold has now used my marked red line as a retest, which we can always only recognize clearly in hindsight, and thereby shifts from the downtrend into an uptrend, gold could, over the long term, reach 7,000 to 8,000 dollars per ounce within a period of one to three years. The market, however, is currently divided. There is a great deal of uncertainty, both in the markets and around the world. On the one hand, this can be a catalyst for higher gold prices. On the other hand, potential interest rate hikes in the US and in the rest of the world are on the table, which can weigh on the price in the short term. Either way, interest rates will have to come down again over the long term, since the economy would not withstand a permanently high interest rate level. At the same time, governments continue to pile up debt to an immeasurable degree. The US has just crossed the 40 trillion dollar mark, which is 40,000 billion dollars. The combination of high inflation, persistent uncertainty and ever increasing debt leaves, over the long term, hardly any other path than a faster, stronger and more uncontrolled devaluation of money. This is exactly why I am a friend of Bitcoin and gold. I consider these two assets to be the best bets over the long term against a debt based monetary system that is moving, step by step, ever closer to the abyss of hyperinflation. Maybe not tomorrow, maybe not in a few months and maybe not in a few years. In my view, however, there is no such thing as too big to fail. Infinity has no value. It devalues, first gradually, then faster, stronger and more uncontrolled. In a world in which there will be an infinite amount of money, real value will be measured in other assets. For me, these are gold and Bitcoin. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 09/01/26 Market Analysis: Gold(/Bitcoin) If gold has now used my marked red line as a retest, which we can always only recognize clearly in hindsight, and thereby shifts from the downtrend into an uptrend, gold could, over the long term, reach 7,000 to 8,000 dollars per ounce within a period of one to three years. The market, however, is currently divided. There is a great deal of uncertainty, both in the markets and around the world. On the one hand, this can be a catalyst for higher gold prices. On the other hand, potential interest rate hikes in the US and in the rest of the world are on the table, which can weigh on the price in the short term. Either way, interest rates will have to come down again over the long term, since the economy would not withstand a permanently high interest rate level. At the same time, governments continue to pile up debt to an immeasurable degree. The US has just crossed the 40 trillion dollar mark, which is 40,000 billion dollars. The combination of high inflation, persistent uncertainty and ever increasing debt leaves, over the long term, hardly any other path than a faster, stronger and more uncontrolled devaluation of money. This is exactly why I am a friend of Bitcoin and gold. I consider these two assets to be the best bets over the long term against a debt based monetary system that is moving, step by step, ever closer to the abyss of hyperinflation. Maybe not tomorrow, maybe not in a few months and maybe not in a few years. In my view, however, there is no such thing as too big to fail. Infinity has no value. It devalues, first gradually, then faster, stronger and more uncontrolled. In a world in which there will be an infinite amount of money, real value will be measured in other assets. For me, these are gold and Bitcoin. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
12
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/30/26 Bitcoin Swing Leverage Future Position: I have slightly adjusted my plan. I entered the market with one quarter of the capital allocated for the swing position at around 78,000 USD. My originally planned lower entries at 73,000 USD and 70,000 USD nevertheless remain in place. However, since the market is looking somewhat stronger in the short term, I decided to already be positioned on a smaller scale now, rather than waiting solely for the lower prices. My original plan therefore still partly retains its validity, and I will scale the position up further step by step from here. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/30/26 Bitcoin Swing Leverage Future Position: I have slightly adjusted my plan. I entered the market with one quarter of the capital allocated for the swing position at around 78,000 USD. My originally planned lower entries at 73,000 USD and 70,000 USD nevertheless remain in place. However, since the market is looking somewhat stronger in the short term, I decided to already be positioned on a smaller scale now, rather than waiting solely for the lower prices. My original plan therefore still partly retains its validity, and I will scale the position up further step by step from here. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
9
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/26/26 Bitcoin Leveraged Swing Futures Position. I am documenting here how I will deliberately build up a leveraged Bitcoin swing futures position into a very large position over the coming weeks and months. The buildup is intentionally gradual. I enter in a staggered manner and use pullbacks, corrections and retests to increase the position step by step. If the thesis plays out and this really was the bottom, I will keep scaling the position higher throughout the entire bull run, until I can forecast a top and start realizing profits piece by piece near my forecasted top, until the position is fully closed. If the thesis is not confirmed, the position dissolves on its own, since my liquidation price is deliberately my stop loss. This way I tie up as little capital as possible on the exchange from the outset. A brief classification: A futures position is a contract with which one bets on the future price development of an underlying asset without owning it directly. It is therefore not per se a leverage instrument, but rather an instrument that can be scaled up with a freely selectable leverage. One can bet on rising prices (long) or falling prices (short). The chosen leverage multiplies both the profit and the loss. If the price falls so far that the deposited margin can no longer cover the losses, the position is automatically liquidated, and the deployed margin is lost. My trading strategy for the Bitcoin position: 1. Initial entry: Opening a long position at 73,000 USD with a leverage factor of 8x and 50% of the capital. 2. Position buildup (scaling): Additional buy at 70,000 USD, likewise with 8x leverage and the remaining 50% of the capital. 3. Capital deployed (margin): 100 EUR (fictional example calculation to protect the real position size, split into 2x 50 EUR). 4. Weighted entry price: 71,469 USD. 5. Liquidation level: Calculated total loss of the entire margin at 62,535 USD (exactly 12.5% price decline from the weighted entry price). 6. Risk management: Using the liquidation price as a systemic stop loss, in order to tie up as little capital as possible on the platform in advance. Conclusion: I am building an 8x leverage position in the range of 70,000 to 73,000 USD, which, once all orders are filled, results in an average entry of 71,469 USD, with a liquidation price (stop loss) of 62,535 USD. As soon as I am in profit and we have sustainably broken above the 82,000 USD, I will move a real stop loss into profit, so that I can no longer make a loss on the position. After that, once we have sustainably broken above the 82,000 USD, I will use every pullback, every correction and every retest to keep scaling the position up strongly, step by step, also with new margin. In doing so, I always set my stop loss such that it lies in profit and always below the previous higher low, so that I can, as far as possible, not be stopped out of the position. Important note: Under no circumstances should anyone trade any part of this after me. This is by no means an encouragement to copy my trades in any form, neither exactly nor roughly. Everyone should carry out their own analysis, apply their own risk management and build their own trade. What I document here is solely my personal approach. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/26/26 Bitcoin Leveraged Swing Futures Position. I am documenting here how I will deliberately build up a leveraged Bitcoin swing futures position into a very large position over the coming weeks and months. The buildup is intentionally gradual. I enter in a staggered manner and use pullbacks, corrections and retests to increase the position step by step. If the thesis plays out and this really was the bottom, I will keep scaling the position higher throughout the entire bull run, until I can forecast a top and start realizing profits piece by piece near my forecasted top, until the position is fully closed. If the thesis is not confirmed, the position dissolves on its own, since my liquidation price is deliberately my stop loss. This way I tie up as little capital as possible on the exchange from the outset. A brief classification: A futures position is a contract with which one bets on the future price development of an underlying asset without owning it directly. It is therefore not per se a leverage instrument, but rather an instrument that can be scaled up with a freely selectable leverage. One can bet on rising prices (long) or falling prices (short). The chosen leverage multiplies both the profit and the loss. If the price falls so far that the deposited margin can no longer cover the losses, the position is automatically liquidated, and the deployed margin is lost. My trading strategy for the Bitcoin position: 1. Initial entry: Opening a long position at 73,000 USD with a leverage factor of 8x and 50% of the capital. 2. Position buildup (scaling): Additional buy at 70,000 USD, likewise with 8x leverage and the remaining 50% of the capital. 3. Capital deployed (margin): 100 EUR (fictional example calculation to protect the real position size, split into 2x 50 EUR). 4. Weighted entry price: 71,469 USD. 5. Liquidation level: Calculated total loss of the entire margin at 62,535 USD (exactly 12.5% price decline from the weighted entry price). 6. Risk management: Using the liquidation price as a systemic stop loss, in order to tie up as little capital as possible on the platform in advance. Conclusion: I am building an 8x leverage position in the range of 70,000 to 73,000 USD, which, once all orders are filled, results in an average entry of 71,469 USD, with a liquidation price (stop loss) of 62,535 USD. As soon as I am in profit and we have sustainably broken above the 82,000 USD, I will move a real stop loss into profit, so that I can no longer make a loss on the position. After that, once we have sustainably broken above the 82,000 USD, I will use every pullback, every correction and every retest to keep scaling the position up strongly, step by step, also with new margin. In doing so, I always set my stop loss such that it lies in profit and always below the previous higher low, so that I can, as far as possible, not be stopped out of the position. Important note: Under no circumstances should anyone trade any part of this after me. This is by no means an encouragement to copy my trades in any form, neither exactly nor roughly. Everyone should carry out their own analysis, apply their own risk management and build their own trade. What I document here is solely my personal approach. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
16
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/26/26 Market Analysis: Copper Copper has broken through my red line to the upside for the third time. This is the last time I will emphasize that I am 2x Copper Long (ISIN JE00B2NFTF36). I expect an increase of up to 30%, which would be approximately 60% with a 2x leverage ETC. Breakout soon. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/26/26 Market Analysis: Copper Copper has broken through my red line to the upside for the third time. This is the last time I will emphasize that I am 2x Copper Long (ISIN JE00B2NFTF36). I expect an increase of up to 30%, which would be approximately 60% with a 2x leverage ETC. Breakout soon. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
12
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/24/26 Market Analysis: Bitcoin The past is not causality. However, should we see a similar price development to 2022, Bitcoin could still rise above 82,000 dollars this week, followed by a few days to a few weeks of a sideways phase. After this cooling off of the market, Bitcoin could briefly retest or correct into the region of 75,000 dollars, in order to then take on the 87,000 to 100,000 dollars with fresh momentum over the coming months. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/24/26 Market Analysis: Bitcoin The past is not causality. However, should we see a similar price development to 2022, Bitcoin could still rise above 82,000 dollars this week, followed by a few days to a few weeks of a sideways phase. After this cooling off of the market, Bitcoin could briefly retest or correct into the region of 75,000 dollars, in order to then take on the 87,000 to 100,000 dollars with fresh momentum over the coming months. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
10
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/21/26 Market Observation: Bitcoin With the breakout of the past few days, Bitcoin has practically blown away several indicators. What becomes decisive now is that Bitcoin sustainably breaks above the 82,000 to 83,000 dollar mark in the coming weeks, confirmed by a retest and a higher local high. If that does not succeed, a lower low would still be possible. However, if Bitcoin rises sustainably above the 82,000 to 83,000 dollars, the bottom should, in my assessment, be in. I have brought along some numbers, data and facts that make clear the rise is sustainable in any case. The only question is whether it stays that way. These charts illustrate what has happened with Bitcoin over the past few days: 1. Aggregated Volume, Spot Volume, Open Interest, Funding and Liquidations (Velo, BTCUSDT) During the breakout, both aggregated trading volume and spot volume rose sharply at the same time. Open interest, however, did not follow along but actually fell. This is an extremely positive sign, because it means that leverage is leaving the market while genuine spot demand is entering it at the same time. Real Bitcoin was therefore bought on the spot market, and the move is not being carried by leverage in the futures market. A simultaneous rise in price and genuine spot demand usually indicates a sustainable move. 2. Bitcoin Daily with Moving Averages (TradingView) The chart shows the Bitcoin daily price with two moving averages. The green marked lines indicate the historical points at which the shorter moving average crossed the longer one from below to above. In the past, these signals often coincided with the beginning of stronger upward moves, and currently the two moving averages are once again approaching such a crossover. 3. Bitcoin Bull Score Index (CryptoQuant) The Bull Score has risen back into bullish territory above 60 for the first time since October 2025. Even in May, when Bitcoin was above 82,000 dollars, the indicator had failed to make this jump. Currently, 6 out of 10 indicators have turned green again, including demand growth, stablecoin liquidity and the trader realized price. 4. Funding Rate (Coinglass) After more than 3 billion dollars in short liquidations, the average funding rate has flipped negative, to around minus 0.0019%. This means that shorts are paying the longs and that, despite the rise, fresh short positions have entered the market. It shows that many traders do not trust the rise and continue to bet on falling prices, even though so many shorts have already been liquidated. Exactly this can trigger further liquidations and thus, going forward, an upward cascade as well. 5. Conviction Buyers, Share of Supply (Glassnode) So-called high conviction buyers have been accumulating since January. The share of supply this cohort has bought is even larger than during the comparable behavior last seen in 2022. 6. BTC Realized Price excluding 7Y Supply (CryptoQuant) Bitcoin is currently attempting to close above the cost basis of the active supply. Excluding dormant Bitcoin, this results in a realized price of around 70,400 dollars. This level was last reclaimed in early April, before Bitcoin dropped back below it in early June. 7. DXY Bull Signal (CryptoQuant) When the dollar index DXY trades above its 365 day moving average, this has historically been negative for Bitcoin and other risk assets. When it trades below it, this has historically been a good environment for the markets. The DXY currently stands at around 98.9. 8. BTC Monthly, 50 Month EMA (Rekt Capital) On the monthly chart, Bitcoin has reached the 50 month EMA and is currently overextending slightly beyond it. A monthly close below the 50 month EMA would tend to be bearish, while a sustainable move above this resistance zone would be the prerequisite for taking on the broader lower highs. 9. BTC Weekly, 21 and 50 Week EMA (Rekt Capital) On the chart, Bitcoin is approaching the 21 week and 50 week EMA. To confirm the bullish strength, a weekly close above the 21 week EMA with a subsequent successful retest as support would be required. Otherwise, a repeat of the early 2026 relief rally with a subsequent continuation of the downtrend would loom. Conclusion: We have some strong signals that this could be the bottom. The real confirmation, however, is still lacking, and we have to wait. A bottom can only ever be clearly determined in hindsight. Only time will tell whether this truly was the bottom. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/21/26 Market Observation: Bitcoin With the breakout of the past few days, Bitcoin has practically blown away several indicators. What becomes decisive now is that Bitcoin sustainably breaks above the 82,000 to 83,000 dollar mark in the coming weeks, confirmed by a retest and a higher local high. If that does not succeed, a lower low would still be possible. However, if Bitcoin rises sustainably above the 82,000 to 83,000 dollars, the bottom should, in my assessment, be in. I have brought along some numbers, data and facts that make clear the rise is sustainable in any case. The only question is whether it stays that way. These charts illustrate what has happened with Bitcoin over the past few days: 1. Aggregated Volume, Spot Volume, Open Interest, Funding and Liquidations (Velo, BTCUSDT) During the breakout, both aggregated trading volume and spot volume rose sharply at the same time. Open interest, however, did not follow along but actually fell. This is an extremely positive sign, because it means that leverage is leaving the market while genuine spot demand is entering it at the same time. Real Bitcoin was therefore bought on the spot market, and the move is not being carried by leverage in the futures market. A simultaneous rise in price and genuine spot demand usually indicates a sustainable move. 2. Bitcoin Daily with Moving Averages (TradingView) The chart shows the Bitcoin daily price with two moving averages. The green marked lines indicate the historical points at which the shorter moving average crossed the longer one from below to above. In the past, these signals often coincided with the beginning of stronger upward moves, and currently the two moving averages are once again approaching such a crossover. 3. Bitcoin Bull Score Index (CryptoQuant) The Bull Score has risen back into bullish territory above 60 for the first time since October 2025. Even in May, when Bitcoin was above 82,000 dollars, the indicator had failed to make this jump. Currently, 6 out of 10 indicators have turned green again, including demand growth, stablecoin liquidity and the trader realized price. 4. Funding Rate (Coinglass) After more than 3 billion dollars in short liquidations, the average funding rate has flipped negative, to around minus 0.0019%. This means that shorts are paying the longs and that, despite the rise, fresh short positions have entered the market. It shows that many traders do not trust the rise and continue to bet on falling prices, even though so many shorts have already been liquidated. Exactly this can trigger further liquidations and thus, going forward, an upward cascade as well. 5. Conviction Buyers, Share of Supply (Glassnode) So-called high conviction buyers have been accumulating since January. The share of supply this cohort has bought is even larger than during the comparable behavior last seen in 2022. 6. BTC Realized Price excluding 7Y Supply (CryptoQuant) Bitcoin is currently attempting to close above the cost basis of the active supply. Excluding dormant Bitcoin, this results in a realized price of around 70,400 dollars. This level was last reclaimed in early April, before Bitcoin dropped back below it in early June. 7. DXY Bull Signal (CryptoQuant) When the dollar index DXY trades above its 365 day moving average, this has historically been negative for Bitcoin and other risk assets. When it trades below it, this has historically been a good environment for the markets. The DXY currently stands at around 98.9. 8. BTC Monthly, 50 Month EMA (Rekt Capital) On the monthly chart, Bitcoin has reached the 50 month EMA and is currently overextending slightly beyond it. A monthly close below the 50 month EMA would tend to be bearish, while a sustainable move above this resistance zone would be the prerequisite for taking on the broader lower highs. 9. BTC Weekly, 21 and 50 Week EMA (Rekt Capital) On the chart, Bitcoin is approaching the 21 week and 50 week EMA. To confirm the bullish strength, a weekly close above the 21 week EMA with a subsequent successful retest as support would be required. Otherwise, a repeat of the early 2026 relief rally with a subsequent continuation of the downtrend would loom. Conclusion: We have some strong signals that this could be the bottom. The real confirmation, however, is still lacking, and we have to wait. A bottom can only ever be clearly determined in hindsight. Only time will tell whether this truly was the bottom. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
19
JLTerminal's profile image
JLTerminal Verified account @JLTerminal
08/21/26 Market Analysis: Bitcoin Bitcoin surges around 23% in just three days. As forecasted in my previous analysis, I had considered a breakout to the upside more probable in percentage terms than one to the downside, and that is exactly what has now materialized. I had anticipated a rise in the order of 7,000 to 12,000 dollars over a few consecutive days. In reality, Bitcoin shot up by as much as around 15,000 dollars at its peak, which corresponds to a gain of roughly 23%. I believe the stronger impulse is, for the moment, initially complete. My expectation is that we consolidate over the weekend and possibly into the coming week as well, followed by a smaller correction. Only after that do I expect a new attempt to take on the next local highs. As can be seen in my public portfolio, I bought the recent lows step by step and struck massively once more before the breakout, so that I was 95% invested with the capital I wanted to deploy. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
JLTerminal's tweet photo. 08/21/26 Market Analysis: Bitcoin Bitcoin surges around 23% in just three days. As forecasted in my previous analysis, I had considered a breakout to the upside more probable in percentage terms than one to the downside, and that is exactly what has now materialized. I had anticipated a rise in the order of 7,000 to 12,000 dollars over a few consecutive days. In reality, Bitcoin shot up by as much as around 15,000 dollars at its peak, which corresponds to a gain of roughly 23%. I believe the stronger impulse is, for the moment, initially complete. My expectation is that we consolidate over the weekend and possibly into the coming week as well, followed by a smaller correction. Only after that do I expect a new attempt to take on the next local highs. As can be seen in my public portfolio, I bought the recent lows step by step and struck massively once more before the breakout, so that I was 95% invested with the capital I wanted to deploy. (Disclaimer: This reflects personal analysis based on publicly available data. This is not financial advice or a recommendation to buy or sell any asset.)
0
1
0
0
13

Last Seen Users on Sotwe

S6r6an's profile image
μ†Œλž€
Seen from Korea
xiaoyemao9102's profile image
倜猫小姐姐 Verified account
YayasHere's profile image
1-800-BEHAPPY
Seen from United States
NguynThanhVuong's profile image
ChαΊ₯n BΓ© Đù 🌹🌹🌹
Seen from Vietnam
rami350855's profile image
Am Am
Seen from Nigeria
Zouzouk160256's profile image
queen z πŸ‡²πŸ‡¦
Seen from Algeria
gataaasp's profile image
Gataaasp Verified account
Seen from India
linaa_london's profile image
Linaa_london Verified account
Seen from Germany
AcuyLiding's profile image
ACUY LIDING
Seen from Indonesia
monoshisbsmz's profile image
沫沫
Seen from United States

Trends for you

1
Colts
Under 10K tweets
2
Mahomes
Under 10K tweets
3
#Lanterns
Under 10K tweets
4
Treadwell
Under 10K tweets
5
Shane Steichen
Under 10K tweets
6
Daniel Jones
Under 10K tweets
7
#INDvsKC
Under 10K tweets
8
#LaGranjaVIPMX
Under 10K tweets
9
Butker
Under 10K tweets
10
Presley Gerber
Under 10K tweets

Most Popular Users

elonmusk's profile image
1
Elon Musk Verified account
@elonmusk
241.7M followers
barackobama's profile image
2
Barack Obama Verified account
@barackobama
119M followers
cristiano's profile image
3
Cristiano Ronaldo Verified account
@cristiano
114.3M followers
realdonaldtrump's profile image
4
Donald J. Trump Verified account
@realdonaldtrump
111.8M followers
narendramodi's profile image
5
Narendra Modi Verified account
@narendramodi
107.2M followers
rihanna's profile image
6
Rihanna Verified account
@rihanna
98.7M followers
nasa's profile image
7
NASA Verified account
@nasa
92.4M followers
justinbieber's profile image
8
Justin Bieber Verified account
@justinbieber
91.8M followers
katyperry's profile image
9
KATY PERRY Verified account
@katyperry
89.9M followers
taylorswift13's profile image
10
Taylor Swift Verified account
@taylorswift13
83.9M followers
ladygaga's profile image
11
Lady Gaga Verified account
@ladygaga
75.4M followers
imvkohli's profile image
12
Virat Kohli Verified account
@imvkohli
73.3M followers
kimkardashian's profile image
13
Kim Kardashian Verified account
@kimkardashian
70.9M followers
youtube's profile image
14
YouTube Verified account
@youtube
68.8M followers
neymarjr's profile image
15
Neymar Jr Verified account
@neymarjr
66.3M followers
billgates's profile image
16
Bill Gates Verified account
@billgates
65.1M followers
selenagomez's profile image
17
Selena Gomez Verified account
@selenagomez
63M followers
theellenshow's profile image
18
The Ellen Show
@theellenshow
62.3M followers
cnn's profile image
19
CNN Verified account
@cnn
61.8M followers
x's profile image
20
X Verified account
@x
60.7M followers