Stream

Looking for reflexive high-beta trades.

[Stream @CryptoStreamHub]

I changed my view on Bitcoin.

My thesis shifted from "Supply Overhang is holding it hostage" towards "Strength despite unusually high visible supply."

Recently we observed unusual (but expected) selling behavior from Strategy and MARA among some other smaller participants, yet Bitcoin failed to trade materially lower while the supply overhang now seems to be improving quickly.

Strategy & Saylor:

- STRC lost its $100 peg & traded below $75
- Strategy sold ~7,000 Bitcoin (~$430m)
- STRC has since recovered to ~$94, closing back in on its $100 peg
- Strategy's USD reserve has grown to $4.65bn

While this situation unfolded Bitcoin barely even flinched.

More importantly, the Saylor supply-overhang bear case has now weakened substantially as STRC moves back towards $100 and Strategy's cash reserve continues to grow.

MARA sold even more:

MARA sold 23,093 BTC for ~$1.6bn in H1 while aggressively cleaning up its balance sheet, repurchasing ~$1bn of convertible notes and reducing total debt from $3.6 bn to $2.4 bn.

At the end of Q2 they still owned 35,577 BTC, which theoretically represented another major source of potential supply.
But something important changed last week:

MARA has since raised $600m of incremental financing against its Bitcoin holdings, pledging 18,750 BTC as collateral rather than selling them to help finance its infrastructure expansion.

That's more than half of its remaining Bitcoin treasury.

This makes the situation look much less like MARA is liquidating its Bitcoin treasury because it lost conviction.

Instead, MARA monetized a significant portion of its treasury while cleaning up the balance sheet, and has now started using a large part of the remaining Bitcoin as collateral to finance its infrastructure expansion.

MARA used to be one of the largest potential corporate sources of Bitcoin supply.

The probability that all 35k remaining BTC simply get dumped into the market just decreased substantially.

[Stream @CryptoStreamHub]

Honestly?

I don't think this is a very nuanced take.

With all the respect to Kyle but he basically says: "It already ran up a lot so there is a risk that it will stop to go up."

1) This essentially says noting
2) It's not a great framework
3) Doesn't even need to be correct

I agree that there is a risk of a technical positioning unwind, but we just had this.

[Stream @CryptoStreamHub]

The Saylor situation can be understood by the following:

Q1:

- Debt ~$8.2B
- USD Reserve ~$2.1B
- STRC works fine

Then: STRC collapses to $75
-> Strategy has to sell BTC

Now:

- Debt $6.7B
- USD Reserve $4.65B
- STRC ~94 closing $100

Strategy didn't only "repair" STRC they also reduced convertible debt and grew their USD reserve while only selling $400m of BTC.

He now has over 2 years of runway to cover his preferred dividends and debt interest.

Far from optimal but the picture is improving.

[Stream @CryptoStreamHub]

I changed my view on Bitcoin.

My thesis shifted from "Supply Overhang is holding it hostage" towards "Strength despite unusually high visible supply."

Recently we observed unusual (but expected) selling behavior from Strategy and MARA among some other smaller participants, yet Bitcoin failed to trade materially lower while the supply overhang now seems to be improving quickly.

Strategy & Saylor:

- STRC lost its $100 peg & traded below $75
- Strategy sold ~7,000 Bitcoin (~$430m)
- STRC has since recovered to ~$94, closing back in on its $100 peg
- Strategy's USD reserve has grown to $4.65bn

While this situation unfolded Bitcoin barely even flinched.

More importantly, the Saylor supply-overhang bear case has now weakened substantially as STRC moves back towards $100 and Strategy's cash reserve continues to grow.

MARA sold even more:

MARA sold 23,093 BTC for ~$1.6bn in H1 while aggressively cleaning up its balance sheet, repurchasing ~$1bn of convertible notes and reducing total debt from $3.6 bn to $2.4 bn.

At the end of Q2 they still owned 35,577 BTC, which theoretically represented another major source of potential supply.
But something important changed last week:

MARA has since raised $600m of incremental financing against its Bitcoin holdings, pledging 18,750 BTC as collateral rather than selling them to help finance its infrastructure expansion.

That's more than half of its remaining Bitcoin treasury.

This makes the situation look much less like MARA is liquidating its Bitcoin treasury because it lost conviction.

Instead, MARA monetized a significant portion of its treasury while cleaning up the balance sheet, and has now started using a large part of the remaining Bitcoin as collateral to finance its infrastructure expansion.

MARA used to be one of the largest potential corporate sources of Bitcoin supply.

The probability that all 35k remaining BTC simply get dumped into the market just decreased substantially.

[Stream @CryptoStreamHub]

@RunnerXBT This is bullish. Mara is mostly done selling

[Stream @CryptoStreamHub]

@alpha_pls + high likelyhood that MARA finished selling.

They sold $1.6b in BTC and now pledged around half of their remaining treasury as collateral in a BTC backed loan instead of selling it.

[Stream @CryptoStreamHub]

The supply overhang picture is drastically improving:

Saylor increased his cash balance and STRC is closing in on $100 again.

MARA sold $1.6b and just pledged half of their remaining BTC as collateral to finance its infrastructure expansion instead of selling it.

All of this didn't push us lower and the supply side looks a lot better.

[Stream @CryptoStreamHub]

The Bitcoin supply overhang already looks materially better than one might expect.

Especially the two largest visible corporate sellers, Strategy and MARA, seem to be largely through the most aggressive part of their balance-sheet cleanup.

Strategy has rebuilt its USD reserve while STRC (almost) recovered back toward par, and MARA has shifted from selling BTC to borrowing against a large part of its remaining treasury.

So outside of a broader macro risk-off move, I struggle to see what would force informed holders to materially reduce BTC exposure here.

If anything, one of the clearest near-term bear cases is fading while Bitcoin has absorbed the selling surprisingly well.

[Will @WClemente]

If you haven't already sold because of DAT overhang, quantum concerns, odds of clarity act passing reaching <20%, or relative price underperformance, what would cause informed participants to take down their BTC allocation now (outside of a broader macro sell off)?

[Stream @CryptoStreamHub]

@WClemente DAT Supply seems to quickly improve with MARA almost fully out of the Situtation and Saylor also improving.

[Stream @CryptoStreamHub]

@zeroxkyle @aleabitoreddit Intel + Micron 🤝

[Stream @CryptoStreamHub]

I just read this bearish case for memory, but I don’t think it is entirely accurate.

The first argument describes short-term headwinds from the collapse of Korean leveraged retail exposure. That makes sense as a flow argument, but it applies much more directly to SK hynix and Samsung than to Micron.

Micron is effectively a two-layer trade.

One part is the memory trade, while the other is its strategic role as the only major US-based DRAM and HBM manufacturer in what increasingly looks like an AI arms race between the US and China.

I don't claim that this does make the stock immune to positioning or sector-wide selling, but it gives Micron a different investment profile and a different strategic value proposition.

I also disagree with the second argument.

Reducing the amount of HBM per GPU or per rack does not automatically result in lower HBM demand at the cluster level. Optics enables faster communication between GPUs, racks and datacenters, but it does not replace local HBM.

HBM and optics solve two entirely different problems.

A larger optically connected cluster may use less HBM per individual accelerator while still containing significantly more HBM in aggregate, because the number of accelerators and the size of the compute domain are increasing.

The development can therefore be simultaneously bullish for optics and memory.

The third argument is unfortunately much stronger.

Memory stocks may struggle before the absolute pricing peak if the pace of price increases begins to slow and earnings revisions stop accelerating.

We already saw a version of this after Micron’s last earnings report. The company beat expectations, but the stock sold off because the rate of fundamental acceleration was less dramatic than before.

A similar dynamic could be observed with SanDisk last week. Strong reported results were not enough once the market started focusing on slower sequential growth and the possibility that margins were approaching a short-term peak.

So I agree that memory stocks do not necessarily need fundamentals to deteriorate in order to underperform. It can be enough for the rate of improvement to slow.

But this scenario cannot be viewed in isolation.

Micron has already fallen roughly 40% from peak to trough. Positioning has been reduced, expectations have reset, and at least part of the anticipated slowdown in the rate of change may already be reflected in the price.

That materially changes the risk/reward.

Before the selloff, the market was pricing continued acceleration and leaving very little room for disappointment.

After a 40% drawdown, the bearish case increasingly requires more than just slower growth. It requires either meaningful earnings downgrades, an earlier-than-expected pricing peak, or evidence that the structural demand outlook is weakening.

So the third argument is valid, but it may also be backward-looking.

[Jukan @jukan05]

I think the market ultimately has no choice but to go sell memory, long optical in the "short term." Actually, some hedge funds already seem to have this position on.

There are three main reasons.

1. With Korean leveraged ETFs effectively dead, LPs are in a redemption rush, which could bring out additional sell on flow.

2. Nvidia is nerfing Rubin Ultra's HBM and responding with optics, tying multiple racks together, so that even if Rubin Ultra's per rack performance is not superior to Rubin, at the cluster level optics let the Rubin Ultra cluster hold an edge over the Rubin cluster. This holds even if Rubin Ultra's HBM nerf is a supply problem rather than a demand problem.

3. Consensus is forming that memory prices will peak within the next two quarters.

Medium to long term I am still a memory bull, but short term I am somewhat bearish on memory. I currently have no memory position.

[Stream @CryptoStreamHub]

I’m noticing a growing wave of concern around U.S. solvency.

More headlines, posts, and videos are focusing on persistent inflation, rising debt, and uncertainty around U.S. monetary policy.

Some interpret this as bullish for Bitcoin (I don’t). Others sees it as a potential revival of the "metals narrative" i.e. Gold, Silver which is already moving.

I’m not sure what to make of it yet, but it feels like an important data point. A new market narrative may be forming, potentially shifting attention away from AI stocks, which could be a risk to be aware of.

[Stream @CryptoStreamHub]

I believe semiconductor & memory stocks present one of the most attractive trading opportunities we've seen in the market in a long time.

Previously, I talked extensively about the fundamentals of memory stocks and how demand for compute continues to be underestimated. Today, however, the trade offers an attractive setup following a massive positioning unwind.

The positioning reset wasn't isolated to one group of investors. It occurred across multiple parts of the market simultaneously:

Global hedge funds reduced semiconductor exposure from record highs.
Korean retail investors experienced widespread liquidations.
One of the most prominent AI-focused hedge funds was forced to deleverage.

Importantly, this positioning reset wasn't accompanied by a comparable deterioration in fundamentals.

If anything, the opposite happened.

Throughout July, the fundamental outlook for AI infrastructure continued to improve. Hyperscalers reaffirmed aggressive AI investment plans, while memory companies continued to emphasize that demand remains structurally stronger than supply.

Recent commentary from both hyperscalers and memory vendors suggests that these supply-demand imbalances are expected to persist well into 2027, driven by accelerating AI deployments and continued HBM supply constraints.

In other words, positioning changed.

The fundamentals didn't.

Demand for compute continues to outpace supply. AI infrastructure spending continues to rise. Long-term supply agreements remain in place, pricing expectations are constructive, and the industry's long-term earnings power appears stronger than ever.

To me, this combination of improving fundamentals, compressed valuations and a significant positioning reset creates one of the most attractive risk/reward setups currently available in the market.

[Stream @CryptoStreamHub]

500k people using agentic AI.

And we are already in a compute shortage.

[Guy @Credib1eGuy]

500k people using agentic AI

And we are already in a compute shortage

[Guy @Credib1eGuy]

500k people using agentic AI

And we are already in a compute shortage

[Stream @CryptoStreamHub]

I believe semiconductor & memory stocks present one of the most attractive trading opportunities we've seen in the market in a long time.

Previously, I talked extensively about the fundamentals of memory stocks and how demand for compute continues to be underestimated. Today, however, the trade offers an attractive setup following a massive positioning unwind.

The positioning reset wasn't isolated to one group of investors. It occurred across multiple parts of the market simultaneously:

Global hedge funds reduced semiconductor exposure from record highs.
Korean retail investors experienced widespread liquidations.
One of the most prominent AI-focused hedge funds was forced to deleverage.

Importantly, this positioning reset wasn't accompanied by a comparable deterioration in fundamentals.

If anything, the opposite happened.

Throughout July, the fundamental outlook for AI infrastructure continued to improve. Hyperscalers reaffirmed aggressive AI investment plans, while memory companies continued to emphasize that demand remains structurally stronger than supply.

Recent commentary from both hyperscalers and memory vendors suggests that these supply-demand imbalances are expected to persist well into 2027, driven by accelerating AI deployments and continued HBM supply constraints.

In other words, positioning changed.

The fundamentals didn't.

Demand for compute continues to outpace supply. AI infrastructure spending continues to rise. Long-term supply agreements remain in place, pricing expectations are constructive, and the industry's long-term earnings power appears stronger than ever.

To me, this combination of improving fundamentals, compressed valuations and a significant positioning reset creates one of the most attractive risk/reward setups currently available in the market.

[Stream @CryptoStreamHub]

@avotoast all of the ai trade is pumping

[Stream @CryptoStreamHub]

Still bearish on crypto

[Stream @CryptoStreamHub]

I think it’s a good time to get into AI and memory stocks.

It’s the most interesting sector in the market, and the fundamentals continue to improve.

Heavy deleveraging and a positioning unwind have created a high probability that we’re forming a bottom around these levels.

(Korean investors have been wiped out, Aschenbrenner got liquidated, and retail investors panic sold.)

I’ve followed both prices and news around the AI trade very closely, and nothing I’ve seen suggests that the AI thesis is breaking.

Quite the opposite.

The signals are everywhere: structural demand for AI, compute, and the components required to support it will continue to outpace supply.

[Stream @CryptoStreamHub]

I’m currently looking at:

U.S.-based memory stocks and chip foundries: Micron (MU), Intel (INTC)

Smaller neoclouds with NVIDIA partnerships: IREN, SHAZ, and maybe HIVE

NVIDIA-backed optical networking and data center component companies: Coherent (COHR), Marvell (MRVL), and Lumentum (LITE)

[Stream @CryptoStreamHub]

@kai_xbt It only took 2 prompts to understand that is wrong and I didn't even have to leave twitter.