Odi Capital

Been doing this 20 years. I know a misread when i see one. 🚨 The “openai canceled its $71b ram deal with sk hynix” story going viral right now? It was never a binding deal but letters of intent. Former micron engineers said this in december. Nobody listened. What actually happened: one stargate expansion got scrapped over a financing dispute with oracle. The core 4.5 gigawatt program is still live. Dram contract prices are up 90-95% qoq. That’s not a market pricing in demand destruction but that’s the worst memory shortage since 2018. Yes turboquant is real. Compressing kv cache doesn’t kill hbm demand in training. Jevons takes care of the rest. the headline is wrong but thesis isn’t

@barkmeta Bear markets are a wealth transfer, always have been. Question is which side you’re on 👀

Rate cuts are coming. Liquidity is coming. A wall of money that makes 2020 look like peanuts is coming. All time highs beyond belief. Most people will be sidelined when it hits because they let a few scary headlines shake them out of their positions. Don’t be most people.

@JesseCohenInv @Investingcom honestly? none of them. IPO are where retail pays the premium insiders already priced in. i’ll wait 6 months and buy the lockup expiry dip like a normal person. Then I would be interested in Anthropic, databricks and stripe

Which IPO are you looking forward to the most?

@GarciaCap $MSFT below its 200-week is a compelling entry. no debate. but $MSFU in this volatility regime concerns me. daily rebalancing in a headline-driven market creates decay that works directly against a 12-18 month thesis.

If you’re not buying $MSFT while it’s trading below the 200-week moving average, what are you waiting for? I’ll be buying the 2x ETF $MSFU Monday morning

@EhrmantrautCap_ the market is pricing $MU like memory is going back to oversupply hell.it’s not. HBM is structurally constrained, ASPs are up triple digits, and FY2027 consensus EPS sits at $92-$100+. I will take that bet

I recently bought $MU. I believe this will be a massive long-term winner. The bull case for Micron Technology ($MU) is no longer just a "cyclical recovery" story. It has evolved into a structural "AI-utility" thesis. We are witnessing an unprecedented divergence: While $MU's earnings power is exploding due to the AI-driven memory supercycle, its valuation has compressed to levels historically reserved for dying industries. At a current valuation of below 4x expected FY2027 earnings, $MU represents one of the most asymmetric risk-reward profiles in the entire semiconductor sector. Consensus estimates for FY2027 EPS now sit near $92 - $100+, driven by sustained HBM (High Bandwidth Memory) demand and a triple-digit increase in Average Selling Prices (ASPs). At current price levels, the market is pricing $MU as if earnings will collapse after 2027/2028. However, with the AI infrastructure buildout projected to last through 2030, this "cliff" is looking increasingly unlikely. Even a re-rating to a modest 10x P/E (still a discount to the broader semi-index) would imply a stock price of $900 - $1,000 by 2027.

april 9th, 2025. the single largest daily point gain in s&p 500 history. triggered by a tweet. the most powerful market on earth moved more in one session than any day in its 150-year history because one man changed his mind on tariffs. this is the volatility regime we live in now. price accordingly

Throwback to April 9th, 2025 the single largest daily point gain and the 9th largest single day percentage gain in S&P 500 history 🟢 Here's what the 🇺🇸 stock market looked like on that day

@CaptainisHarry @StockMKTNewz I hope not. Here they still have a choice in the decision, in 1970 it was more complicated. 😉

@oderKinvest @StockMKTNewz will 1970s be back again? that was long 🐻 …

The S&P 500 is currently on pace for its worst month since 2022 🔴

@UziCryptoo 5% variable debt in a rising rate environment gets paid first. non-negotiable. then you split the rest: market for compounding, something for the family. a 45 year old with 2 kids optimizing purely for IRR is missing the point entirely.

My friend just got a $20k bonus at work He’s 45, married with 2 kids Has ~$25k in student loans at a variable rate around 5% Here are his options: Invest the bonus Put it toward the student loans Take the family on a much-needed luxury vacation What would you do?

@TedPillows historically, insider selling has near-zero predictive value on forward returns. insiders have been net sellers in 7 of the last 10 bull market years. the s&p still tripled. clustered buying across multiple insiders in the same name, that one actually moves the needle.

Insiders continue to dump stocks like there's no tomorrow. Not a good sign.

@StockSavvyShay @fiscal_ai i’ve watched markets long enough to know what a sentiment-driven dislocation looks like. $NVDA at a decade-low valuation while every major data center on earth runs on its silicon is one of them. the infrastructure buildout doesn’t care about your macro fears. 😮

$NVDA is now the cheapest it has been in a decade. And this is still the core compute engine of the new AI economy powering the largest infrastructure buildout in history.

@aakashgupta 19.7x forward P/E. sub-20x triggered 13 times since 2020. positive forward returns all 13. I’ve seen two full cycles. the data never changes. only the headlines do

If you're under 40, this is one of the best buying opportunities you'll get this year. Run the math on what happens when you buy at these levels historically. The forward P/E just fell to 19.7x. That's below the 5-year average of 20.1x and the cheapest the index has traded since Liberation Day in April 2025. Citadel's Scott Rubner flagged it: every time the S&P forward P/E has dropped below 20x since 2020 (13 occurrences), forward returns have been positive. Over the last 50 years, the S&P has had a negative Q1 18 times. Last year it dropped 4.6% in Q1 and finished up 16.4% for the year. In 2003 it fell 3.6% in Q1 and posted 26.4% for the full year. The pattern repeats: after 10% corrections, investors who bought the dip averaged 11% returns within a year and 37% within three years. The panic math is even more telling. Miss just the 10 best trading days and your returns get cut roughly in half. Miss the top 50 and they shrink by nearly 5x. The best days almost always cluster inside the worst months. March 2026 has had 1%+ intraday swings on 14 of 18 trading days. The snapback days are hiding inside this exact volatility. Everyone sharing this chart is seeing a 7.6% decline. The people who build wealth from these moments are seeing a forward P/E in the 6th percentile of its one-year range, Wall Street consensus calling for 10-20% upside, and 50 years of data confirming that selling here is almost always the wrong trade. The worst time to look at your portfolio is the best time to add to it.

@Barchart extreme fear at 10. historically that’s been a buying signal. but history didn’t have this kind of geopolitical uncertainty baked in. i’m not gonna pretend i know where this goes. nobody does right now 👀

JUST IN 🚨: Stock Market plunges to an Extreme Fear reading of 10, the most amount of Fear since before Thanksgiving 👻😱🫂

@Gubloinvestor doubt it goes sub $50, volatility is their best friend. more fear = more retail trading = more revenue for $HOOD. the setup works both ways

$HOOD as cheaper as it could get.. Do you think below $50?

@NoLimitGains Gating client funds for 3 years is a bold customer retention strategy ngl 👀

🚨 UBS halted withdrawals on a $469 million real estate fund for up to 3 years. Translation: “Sorry, we don’t have your money.” This is how it starts.

The most important metrics to watch next quarter : $NVDA : Not revenue but gross margin. Drops below 70% and the AI story could starts cracking $NBIS : Watch revenue backlog and contract wins. no point looking at earnings yet. the question is are enterprise clients signing. that's the only signal that matters right now $META : Revenue per user in Asia. that's where the next billion dollars comes from $PLTR : not government contracts. US commercial customer count. cross 1,000 and the whole narrative flips $SOFI : Not loan volume. how many products the average member uses. the whole bet is becoming your financial everything app. cross-sell or bust $HOOD : Not trading volume. assets under custody. if real money is staying and growing they're building a wealth platform. if not it's still just a meme stock casino drop the ticker you're watching closest below 👇

@StockSavvyShay @fiscal_ai wild part is hood makes money when people panic trade. market’s been a bloodbath for weeks.

$HOOD has fallen nearly 60% in the past six months 😳

@PeteRichey Thanks Peter! Inverted MSCI and SMCI

@oderKinvest You have SMCI in two categories

I have been watching the market all week. Here's where i stand 👇 Stocks i'd buy with my eyes closed : $META $MELI $MSFT $NBIS $AMZN $MU $UBER $NU watching but not touching yet : $LULU $AMD $COIN $HIMS $IREN $SMCI $FICO $CRM stocks i wouldn't touch even if you paid me : $TSLA $ORCL $DJT $SMCI $AMC $MSTR where do you disagree 👇