Will Thrower

Enjoyed chatting with @longriver_hk and Brian Mushonga. The Caledon Global Energy Transition Fund is positioned at the intersection of the AI infrastructure buildout and the clean energy transition. Off to a great start, but we still have a long way to go! open.spotify.com/episode/5K83z3…

@MikeZaccardi Taiwan returns will likely be multiples of Indonesia returns over the next ten years

Across Asia, Indonesia is cheap... Taiwan is expensive Korea $EWY jsut 7.3x f12m EPS

$SPGI All I know is that I get more utility from less than $100/mo of various AI-powered financial data subscriptions than I ever did from a $20k/yr CapIQ seat

@akramsrazor @rubicon59 The market has been pricing in that scenario for years, growth falling into the teens two years out is still the consensus

People calling this stuff fud clearly don’t trade/risk manage in new paradigms. Mkts betting on just one thing on 95% of trading days will tend to get crowded as more and more people forgo diversification for one paradigm to rule them all. Thus many investors will often find themselves in a poor position risk mgmt wise. It is a given that one day a fundamental disruption of some meaningfulness will have developed. Maybe its way better efficiency or maybe its demand digestion due to llm limitations or maybe it’s just share shift/competitive landscape driven. Whatever it is one day, every shortage driven commodity boom has eventually hit that transition moment. Saying it’s too soon is only relevant if everyone has not crowded into a theme like too soon is five years away. And nobody can truly measure that w any degree of certainty. But u sure as shit can see it in an imbalanced tape as perf gaps widen. And if that happening bankrupts u because ur all in on that not happening anytime soon, you simply need people reacting to developments that require discounting some of this risk again to bankrupt u. Thats reflexivity. So by saying it’s all been fud after the fact ur actually missing the point of all of this. And nvda back and forth isn’t a debate about its business contracting/collapsing for the stock to reset. If it just starts doing a teens cagr on eps because of share shifts, it will see a msft/zm level multiple contraction period. And then yes if that remains stable, like an apple it will multiple expand again. It’s the folks downplaying the sensitivity to that well established phenomena after a 25x move in a couple of years that are the real mkt and tech investing fudders.

I predict the Google TPU FUD will end same as the Deepseek FUD, the Microsoft cutting-datacenter-spend FUD, and the countless smaller $NVDA FUDs in between (who remembers?)

Burry crashing out

What you get for $100k/yr

@akramsrazor fire that analyst 😂

$CLS I'm short this...i'm over stack investing.. My supreme analyst sums it up nicely. "Owning the narrative around the AI chain is not the same as owning the profit pool of the AI chain. Celestica has the former; Micron has a real shot at the latter.”

@SolutionOrganic @PythiaR nobody invested $1m at inception. he outperformed for the first 3-4 years and then badly underperformed for the next 25. net destroyer of value

@PythiaR I don't get this critique. He's beating the S&P over the life of his fund while taking less risk than what you're suggesting here, which is what his investors expect from him. Seems like he's doing what he says he's going to do

Sometimes I think about the future, like when I’m 65 and I’ve been doing this a long time what kind of investor will I be? One thing I’ve learned from watching old boomers get clowned by FAANG and now MAG7 for basically 20 years without ever bothering to do the introspection required to admit they were wrong and update how they look at the world is that I don’t want to be that kind of investor.

@value_invest12 nvda is deep value

I own 0/7 at this point. Best days are over.

Eos Energy [Full Investment Thesis]: Everything You Need to Know About $EOSE $EOSE 🔋has been my biggest position since 2023 - studying it has been a part/full-time job in itself.🔋 Here’s my 119-page, ~two-hour presentation on Eos Energy. Many have reached out wanting to learn more about the company reshaping America’s grid. I did this video and tried including everything I know in here, hopefully helping speed up the learning curve for anyone interested. This isn’t meant to be flawless; it’s meant to be done. I hope it helps you grasp the company’s story, the team driving it and the potential they have (at least how I see it…). During the video I’ll cover: 00:00 Introduction 3:49 Global Energy Market 8:15 Why Energy Storage is Key 25:21 Choosing the Right Battery Company 27:15 EOS’ Background and Timeline 38:35 The Product 1:01:56 Production at Scale 1:11:56 Unit of Economics and Financials 1:18:26 Team 1:26:59 Risks 1:29:03 What Comes Next - “Best-In-Class” US Battery and It’s Demand 1:45:52 Boots On the Ground $EOSE is executing toward an unprecedented hyperscaler energy demand: U.S.-made, non-lithium storage, ramping for 24/7 data centers, military bases, supply-chain resilient, and addressing key national energy security issues... I believe significant orders should be coming very soon. Please feel free to share your thoughts, feedback, questions, things I left out, etc. 🔋🔋

is this guy for real? lol

@patientinvestor opportunity cost

✅️ 14 PE ✅️ 11% EPS growth ✅️ 97% Recurring Revenues ✅️ 4% buyback yield ✅️ little debt ✅️ bad sentiment Adobe is one of the best risk/reward setup in this market. $ADBE

@HedgieMarkets garbage in, garbage out

🦔 I've been digging into the AI data center economics, and a hedge fund manager just discovered something that confirms my worst fears about this bubble. Harris Kupperman initially thought data centers were financially questionable, but after talking to industry insiders, he realized the situation is catastrophically worse. What I Found Most Alarming Kupperman originally assumed data center components would depreciate over 10 years, but learned from two dozen senior professionals that the actual lifespan is just 3-10 years due to rapid technology advances. His revised calculations show the industry needs $320-480 billion in revenue just to break even on 2025 data center spending alone. Current AI revenue sits around $20 billion annually. The Industry Panic I'm Seeing What strikes me most is that none of the senior data center professionals Kupperman spoke with understand how the financial math works either. These aren't outside critics but industry veterans who "shoulder a heavy burden" because they know the economics don't add up. When the people building the infrastructure can't explain the business model, I see that as the biggest red flag possible. The Depreciation Treadmill The 3-10 year component lifespan creates a constant capital expenditure cycle where companies must replace entire facilities before they've generated returns on initial investment. This isn't just chips becoming obsolete, it's physical infrastructure wearing out under high-powered usage while technology advances make older systems worthless. My Take This validates everything I've been tracking about the AI bubble being built on impossible economics. When you add 2026 projections with hundreds of new data centers, break-even revenue approaches $1 trillion. Kupperman concludes that "doing it at massive scale doesn't make the economics work any better, it just takes an industry crisis and makes it into a national economic crisis." When industry professionals privately admit the math doesn't work while publicly promoting expansion plans, I'm witnessing financial engineering disguised as technological progress. Hedgie🤗

@buccocapital maybe you're missing something

The worst businesses in the world have no sellers The best businesses in the world are being scrutinized and sold because the CEO gave a weird answer during the last earnings call

@N04121804 @kaisynthbuilds @unusual_whales $LODE can recycle them with zero waste

@kaisynthbuilds @unusual_whales By 2030, the US will need to manage around 1M tons of solar panel waste. This number is expected to grow to 10 million tons by 2050, making the US the second-largest producer of solar panel waste globally. Currently, only about 10% of decommissioned panels are properly recycled

US cancels the world's largest solar farm, with the Trump administration cancelling approval for the 6.2 gigawatt Nevada project, per FT

@jupiters_string @orrdavid people always forget the GARP chapter in that book

@orrdavid You need a really speculative market to get away with that The irony is that those type of value investors need speculation to realise profits (unless they can effect corporate change or cash out) GARP guys and activists are the real antifragile investors

The true backbone of Graham's strategy: Buy less liquid value traps that don't have big downside risk. When they spike someday, sell to some bag holder. The market got more competitive. Unsurprisingly, that no longer works.

It can be tricky for people outside of AI to understand how the massive build out of AI infrastructure makes sense right now. Most of the use of AI at this point is basically asking a question and getting an answer back. It’s hard to perceive how this could turn into 10 or 100X more infra needed to scale.  But everyone inside AI is seeing a very different reality of where AI is going with agents. Just in the past year, in coding for example, we’ve gone from a world where AI was doing basically type ahead tasks or being able to generate a few hundred lines of code, to now agents being able to do hours of coding on their own without interruptions. The economic value difference in these two usage patterns is massive. And the token usage is 100X’s more.  It’s very clear that as we can have agents that can do long running tasks across every area of knowledge work, we’re going to gobble up all the AI infra we can get our hands on.

@Compoundtitans @BourbonCap @theTIKR full scale autonomy drops crash rate to zero

@BourbonCap @theTIKR I am long on Copart business. I believe with self driving cars, the safety will improve, less accidents but cost per repair will go up. Ultimately it will benefit Copart. Further, current reasons for drop are temporary and occurred many times in the past.

You rarely see a company experience a +30% correction even though its operating income is 5X higher than its operating expenses, and the company has zero debt.

@MooreRoger_10 know what you own, scared money don't make money. well done

$ASTS I'm a delivery driver in Seattle WA that makes 65k per year and yesterday I became a millionaire thanks to all the great people on this board and Reddit. Still not selling anytime soon.

Scaling laws can hit a wall tomorrow and the ROIC on the entire cohort of capex could be deeply negative. Not a single person is debating that. Not even Ilya or Dario. But every investment should be guided by probability x expected value. The vast majority of wall st is moderately mis-calibrated on the probabilities. Thats fine. But by far the biggest error is on the expected value. Because if scaling laws hold for another year or two we get material science breakthroughs. If it holds for another three years we solve robotics. If it holds for five we probably solve cancer. If it holds for ten we might reverse aging. The internet was a breakthrough in communication. The smartphone amplified that breakthrough by radically increasing the distribution and persistence of communication. This is not that. This is intelligence. There is nothing ethereal about the humans mind. It is hyper-evolved it is hyper-specialized but it doesn’t operate on quantum entanglement. Wall st is failing because they are attempting to use a linear framework to simulate the actions of a class of people who have an order of magnitude more imagination than they do.