Satish Bathini

Engineering Manager/Investor Disc:Ideas shared are not for recommendation. Consult a Reg SEC/SEBI adviser

[ishmohit @ishmohit1]

Must read..

What a touching letter

Sometimes we forget the relationships that we have. Always count your blessings- v well written

[Vala Afshar @ValaAfshar]

Do not use your energy to worry. Use your energy to believe, to create, to learn, to think and to grow.

—Professor Richard Feynman

[Investment Wisdom @InvestingCanons]

Charlie Munger:

“The secret to life is easy because it's so simple:

You don’t have a lot of envy or resentment. You don’t overspend your income. You stay cheerful in spite of your troubles. You deal with reliable people and you do what you’re supposed to do.”

[Dr. A. Velumani.PhD. @velumania]

Grow or Go!!

[Satish Bathini @satishbathini]

@BeatTheStreet10 Study Arham, similar pattern can be oberved.

[Natural Philosophy @Naturalphilosy]

Søren Kierkegaard on the Importance of Walking

“Above all, do not lose your desire to walk. Everyday, I walk myself into a state of well-being & walk away from every illness. I have walked myself into my best thoughts, and I know of no thought so burdensome that one cannot walk away from it. But by sitting still, & the more one sits still, the closer one comes to feeling ill. Thus if one just keeps on walking, everything will be all right.”

[Aman Godhasara @Aman81407075]

The Art of Patience: Why Growing With a Business is Exactly Like Raising a Child

They say raising a child is the ultimate test of patience. I’d argue that holding a conviction stock through market chaos comes a very close second.

Let me take you through a journey—my own story with a company called Sky Gold—to show you exactly what I mean.

The Infancy Stage: Spotting the Potential

Every parent remembers the moment they brought their kid home. For me, that was March 2024. Sky Gold was just finding its footing, making a respectable 10-15 crore profit a quarter.

Then, management unveiled their "Sky Gold 2.0" vision. It was like hearing a child articulate their first big dream for the future. I watched them for one more quarter, saw the spark, and built my full position.

The Growth Spurt: Fast, Furious, and Misunderstood

What happened next was a blur. Quarter over quarter, this kid didn't just meet expectations; they shattered them. The market took notice, and the stock hit a massive growth spurt, rocketing from 1100 to 4880 in under nine months. It happened in the blink of an eye.

But rapid growth always attracts critics. The whispers started. People said the PE was too stretched, the PAT margin was dangerously thin, and the CFO was negative. The consensus? This kid isn't going to survive the real world.

Teenage Rebellion: The First Big Test

Then, the market sentiment went sour. After a bonus price adjustment, the stock tumbled down to the 300-350 range. Here is the kicker: the company was still performing brilliantly. They were doing everything right. This drop was purely a PE re-rating—the market judging my kid unfairly.

I held on. But many of my friends, the ones I had excitedly told to invest, panicked and sold. They didn't understand the business model. They lost faith at the very first sign of a tantrum.

Outside Bullies: The Macro Shocks

If that wasn't enough, the outside world started throwing punches. Trump came to power, tariff fears dominated the headlines, and the noise grew deafening. The stock was bullied all the way down to 260.

I went from sitting on a 4x return to just 1x from my buying price. Did I panic? Honestly, yes.

Every parent worries when their kid gets knocked down. But I had a serious conversation with myself. I looked at the fundamentals. The business was still executing every single promise they made. I took a deep breath and stayed in the game.

Maturation: Quiet Conviction

Slowly, the dust settled. The stock became range-bound between 300 and 360, staying quiet until December 2025. Then, this January, they announced Sky Gold 3.0.

Because of the earlier market beating, the PE was sitting at a very reasonable 25. I looked at this setup and thought, "My kid has grown up, and the world is underestimating them." I loaded up and bought more.

The market tested us a few more times. War fears pushed it back to 300. I just shrugged and said, "Fine." After March, the true value finally began to show, pushing the stock to an All-Time High of 552.

Then came the new import duty regulations. The stock threw a two-day fit, dropping to 450. I stayed perfectly calm. Almost immediately, Sky Gold management released a one-pager stating, "No impact on us." It was the corporate equivalent of a kid coming home and saying, "Don't worry, I handled the bullies."

Graduation: The Magic of Compounding

The market realized its mistake. The stock touched 700 before the Q1 results, and today, it stands at 840, trading at a 38 PE.

Just look at the resume they've built:
A crystal-clear roadmap straight through to FY30.
CFO is firmly positive.
The absolute best PAT margin in the B2B segment.

Does this look like a juicy deal for a new investor? Of course it does. The market will likely assign a 45-50 PE before the Q2 results, which could push the price into the 960 to 1100 range. And if they increase guidance after Diwali? The upside from here is anyone's guess.

The Moral of the Story

Parenting teaches you that the days are long, but the years are short. It took me 2.5 years to get a 4x return from my first buying price. But after that? It took just 25 days to hit 5x, and a mere 3 days to hit 6x.

That is the magic of compounding.

All you need is patience, conviction, and absolute faith in the businesses you pick. This isn't blind faith—if management had ever come out and admitted they couldn't hit their targets, I would have sold without blinking an eye.

Just like raising a child, if they fail an exam, you don't abandon them, but you do evaluate what happened and make a rational decision.

You don't need a chaotic portfolio of 15-20 stocks to build wealth. You just need 3-4 excellent ones, given good capital allocation, that have the potential to go 5x to 10x. That is where your entire alpha is generated.

[Satish Bathini @satishbathini]

@IndianTechGuide @grok for Nuclear power capacity, what was the govt budget allocation in FY26-27, how much time does it take for approvals and to complete fully operational 1GW plant.

[Satish Bathini @satishbathini]

@gregoryblotnick @grok does the incremental returns on capital invested apply while analysing cyclical businesses like Chips, energy, copper etc. What are the other parameters to check during it’s peak?

[Krynex @kryneeex]

Howard Marks, co-founder of Oaktree Capital, who paid $160 million a year to not invest $11 billion:

"We waited a year before activating that fund. 1.6 percent of 11 billion is 160 million a year that we didn't take."

This is Kelly sizing from an angle almost nobody talks about. The cost is not always being wrong. Sometimes the cost is being right too early and paying full price to wait.

By early 2008, Marks and his partner Bruce Karsh had raised the largest distressed debt fund in history, $11 billion, betting the credit markets were headed for trouble. The fund charged management fees on committed capital whether it was deployed or not. For a full year, Oaktree sat on that capital while paying 1.6 percent annually just to hold the position, watching other managers deploy immediately and start collecting performance fees. $160 million evaporated during the wait, real money, gone, for doing nothing.

Then September 2008 arrived. Lehman collapsed. Oaktree deployed $500 million a week into distressed debt while the rest of the market ran for the exits. The fund that cost $160 million a year to leave uninvested became, in Marks's own words, the proudest moment of his career.

Same point as my article: correct sizing is not free even when it is right. The formula tells you how much to bet. It does not refund the carrying cost of waiting for the bet to make sense. Every dollar Oaktree paid in fees during that patient year was the price of not confusing being early with being wrong, and not confusing being wrong with needing to act anyway.

Here is the catch. Almost every fund manager who raised capital in 2007 felt the same pressure to deploy it fast, because sitting on committed capital looks like failure to clients who are paying for it. Marks did the opposite. He treated the fee drag as a cost of discipline, not a sign the thesis had failed, and waited for a signal instead of a deadline.

The lecture is free, fifty minutes, where Marks walks through exactly why the $160 million was the correct price to pay. Almost nobody watching it changes how much waiting they are willing to fund out of their own pocket before the market proves them right.

[Sammy @sammy_capital]

A 24 year old graduate student wrote 14 pages arguing that Wall Street was measuring risk incorrectly.

His dissertation adviser, Milton Friedman, told him the work was not economics.

The global asset-management industry now holds $147 trillion inside the framework.

His name was Harry Markowitz. The paper was Portfolio Selection.

Before Markowitz, investors mostly searched for the best individual security. He asked a stranger question:

What if two risky assets could become safer simply by being owned together?

The answer depended on covariance.

A security could look dangerous alone and still reduce the risk of the total portfolio if its losses arrived at different times from everything else.

Thirty-eight years after the paper appeared, Markowitz received the Nobel Prize.

In 2009, Yale preserved the cleanest classroom demonstration of the idea.

Professor John Geanakoplos draws two investments with identical expected returns and identical standard deviations.

Holding either one alone produces the same result.

Splitting the capital between both looks pointless.

Then he calculates the combined variance.

It collapses while the expected return stays fixed.

With (N) independent investments, risk falls by (1/\sqrt{N}). Ten positions reduce volatility to roughly one-third. One hundred reduce it to one-tenth.

That is the mathematical engine underneath index funds, portfolio construction, insurance pools, and institutional risk management.

But Geanakoplos refuses to end with the comfortable version.

He turns to the bell curve and explains that classical models make extreme outcomes look nearly impossible.

The real world keeps producing them.

The Yale lecture is sixty-nine minutes long. Free on YouTube.

It begins with two childish-looking bets and ends at the flaw capable of hiding a financial catastrophe inside an elegant model.

[Satish Bathini @satishbathini]

@sammy_capital @grok so holding two risky assets, the second being the better investment opportunity reduces risk here and how should allocation move from first to second in terms of percentage considering the risk.

[Satish Bathini @satishbathini]

@DealsDhamaka @grok what is that gap which Indians are not able to fill and adapt, which Philippine’s are doing.

[Porter Stansberry @porterstansb]

There are two things that are likely to prick the A.I. bubble.

The first is financial: spending trillions in capex that's funded by debt starts the clock ticking towards a crisis. I didn't say a potential crisis -- there's never been a capex cycle of this size that didn't erase about half the debt that funded it. There will be a crisis here too, but not for several years. The crisis clock is starting now, as the hyperscalers have moved from funding the build out with cash to debt. Nvidia's new massive $500 billion debt fund is a clear sign of the funding sources now moving to debt. There will be more such announcements.

The second hurdle isn't financial; it's physical.

The hardest step in making an AI wafer is linking the chips and the memory together. Only a few factories in the world can do it. In 2024, those factories made about 370,000 wafers. A wafer of AI chips holds only a few dozen chips on the silicon. In 2026 the industry will make ~ 1 million A.I. wafers, so nearly 2.7x annual growth. And next year? Morgan Stanley now expects about 19 million AI chips to be built in 2027.

Each chip uses about 2,000 watts of electricity when it runs -- that's about 38 gigawatts.

A gigawatt is about the size of one large nuclear reactor.

Keep in mind, data centers also need fans, pumps, chillers, and network gear. Those roughly double the power a chip needs. So one year of new chips can fill about 70 gigawatts of data center demand -- or 70 new nuclear plants.

In 2024, American data centers used ~ 180 terawatt-hours of electricity. By 2030 they are expected to use ~500 terawatt-hours. That takes them from 4% of all electricity in the country to more than 15%. In only three years.

Utilities have to ramp up power gen fast. Natural gas plants, solar farms, and batteries can be built in one to three years, so those sources will see the most increase in demand through about 2030. Nuclear is the ultimate solution, but those builds take longer. A reactor needs a license, a site, heavy steel parts, and fuel. That process runs 6 to 10 years and, historically, has always come in way over budget and with ~ 20 year timelines.

Vogtle Units 3 and 4 in Georgia are the only new reactors America has built in three decades. They cost more than $35 billion (!) against a $14 billion estimate. And they were delivered 7 years late.

They are the most expensive power plants ever built. Georgia Power raised customer rates 6% to cover the last $7.56 billion of the overrun. Today, zero large reactors are under construction in the United States.

The only commercial construction permit issued in the last decade went to TerraPower in March 2026. TerraPower is building a "Natrium" plant in Kemmerer, Wyoming. It's a 345 megawatt, sodium-cooled fast reactor paired with a molten salt heat storage system. Instead of sending heat straight to a turbine, the reactor heats molten salt in tanks. That lets the plant hold output steady at 345 megawatts or dump stored heat to push more than 500 megawatts for 5 or more hours. It sits next to PacifiCorp's retiring Naughton coal plant, using the existing transmission lines and workforce.

TerraPower plans to pour nuclear concrete by 2027, load fuel by 2030, and reach commercial operation in 2031. But here's the big problem. Cost was estimated at roughly $4 billion in 2021, with the Department of Energy covering up to $2 billion through its Advanced Reactor Demonstration Program. At $4 billion for 345 megawatts, that is about $11,600 per kilowatt — in the same neighborhood as Vogtle. And that is, of course, if it can be built on time and on budget.

Despite these hurdles, technology companies have already signed contracts for close to 10 gigawatts of nuclear power. But the capital investments needed to build these nuclear plants has to begin now. The steel parts come first. A reactor's pressure vessel, its heat exchangers, and its fuel get ordered years before the plant makes any electricity. Few companies in North America can forge and weld parts thick enough to prove they meet nuclear standards. BWXT is one of them. It has built reactors and reactor fuel for United States Navy submarines and aircraft carriers for decades, so it already holds the licenses, employs the experienced welders, and is familiar with the inspection systems.

The Department of Energy has committed $17.5 billion in loans just to buy long lead reactor parts for Westinghouse's AP1000 plants. Canada is likewise planning for 10 large new reactors. BWXT's order book is up 40% this year. And its commercial order book grew 85% in 2025.

I'm skeptical that America will be able to produce enough power to supply the A.I. data centers that have been planned through 2030 -- at least, not with the current nuclear power regulations and not at current electricity prices.

The real bottleneck is going to be electricity. And nuclear is no panacea.

[Satish Bathini @satishbathini]

One of the most important lessons from Rakesh Jhunjhunwala Sir:
he bought Titan after it had already gone up 8X, and again at 16X—showing immense conviction and the courage to buy. He held on even when the stock fell 50%.

Don’t get fixated on price. If the business continues to grow, have the conviction to buy—even after a 50% drawdown. #RJ #stockmarkets

[Vikas Vij @TheClubJunto]

S. Naren's Big Lessons for Small Investors

“Global Central Banks are running the stock markets. Pumping unlimited liquidity, they have distorted asset prices. They’re the kings; we’re all pawns. At some point, they will make a mistake.”

Naren’s Goldmine of Wisdom

Key takeaways from ICICI Pru AMC CIO S. Naren’s new interview with Vishal Khandelwal @safalniveshak:

a. Liquidity Game: “Global central banks are the people running the stock market. They control equities. We are all pawns. They have distorted asset prices in every asset class worldwide.

Central bankers don’t face elections. So, they don’t bother about the consequences. They have unlimited capacity to print money. In the name of stability, they have ensured that markets never go down.

As a result, there are so many companies without earnings, which are trading at record prices across the world. At some point of time, global central banks will make a mistake."

b. Priced to Perfection: In the last 10 years, you were just left looking like a bystander in many good stocks in India. If the earnings increased, the PE also increased at the same rate.

(Means: If earnings doubled from ₹10 to ₹20, PE also doubled from 30 to 60. Stocks became more expensive at the same rate. So, no edge for an investor.)

c. Investing in US Stocks: In today’s US market, if you get attracted just by low PE (world-class companies at very low valuations), you are finished. US investors are only attaching value to companies that have a strong moat.

d. Indian Stock Market: It is time to be conservative and diversify partially from equities to debt and offload some of your existing equity positions. But I find people are again becoming aggressive equity buyers as the market becomes overvalued.

e. Be a Contrarian: Avoid super-hot sectors. Invest in neglected sectors that have been underperforming for a long time. Many investors leave those sectors because they get tired of prolonged underperformance. But that is the time to enter.

f. Avoid Value Traps: Avoid stocks that have underperformed for a long time. Ask yourself: Can it continue to underperform for many more years? Wrong stocks can go to zero. But if any entire sector is underperforming, it will come back.

g. Current IPO Mania: IPO market is currently disconnected from the main market. IPOs are coming at extended valuations, and when they open, they open at even more extended valuations. This is the biggest risk investors face right now.

h. Overvaluations in Quality Stocks: High quality businesses at PE multiples of 60x, 80x, 100x etc. will see sideways movement possibly for several more years. Many of these companies have underperformed since 2020.

i. Management Quality: When a business in a cyclical sector is at the top of the market cycle, very few managements have the humility to realize their success is not due to their skill, but due to the cycle.

These are the managements which can take exceptional decisions at the bottom of a cycle, and they are the best companies to invest in.

j. Corporate Governance: In India corporate governance is also cyclical. In a bull market, corporate governance standards also improve. (So PE goes up.) In a bear market, corporate governance standards deteriorate. (So PE goes further down.)

k. Promoter Quality: The real test of a promoter is what kind of corporate governance standards they maintain in a down cycle or a bear market (when the going gets tough.) Three out of four “visionary promoters” are only visionary in bull markets.

l. 'New Promoter' Risk: In new IPOs, there is no way to know promoter quality because the promoter’s standards have not been tested yet. Their true standards will get discovered only over a cycle. So, new IPOs are risky.

m. Temperament: It is easy to say: buy in bad times and sell in good times. The average investor’s temperament is the opposite of it.

It is not rocket science to know that certain hot sectors, hot stocks, or hot IPOs are overvalued. Everyone has this knowledge. But how many have the temperament to go against the herd instinct and walk away when everyone else is rushing in?

n. How to Beat Temperament: Use Munger’s model of a checklist. Any asset class (like equities, gold, or real estate) falls 40%, put double the money. Any asset class falls 80% (maximum pessimism), put 5x the money. Any asset class goes up 100%, take out the money.

Don’t do this at the level of an individual stock (don’t buy falling knives). Stocks can go to zero. Asset classes can’t go to zero.

ENDQUOTE

“If you’ve got 160 IQ, sell 30 points to somebody else because you won’t need it in investing. What you do need is the right temperament.” – Warren Buffett to NDTV (March 2011) on his visit to India

@arabicatrader

[Satish Bathini @satishbathini]

@FinanceLancelot @grok what does this mean and impact on global markets primarily US

[Porter Stansberry @porterstansb]

Here's the real reason the A.I. boom is going to run out of capital. It's the same reason why rates are rising. And it's exactly why there's suddenly a mad scramble for capital in A.I. Equity values will fall as the cost of capital increases 20%-30%. But that's only the beginning👇

[Finding Compounders @F_Compounders]

Li Lu has held a single company’s stock for over 22 years

During this period , the stock has dropped more than 50% up to 6 times !

[Satish Bathini @satishbathini]

As machines are trained with decades of Internet data in text, photos, and videos, companies that operate the machines which listen to what the physical world speaks and provide intelligence will be the leaders for the next decade. #AI #Stockmarkets #Growth #Trends #Industries