Micron just posted the best quarter in company history.

The stock fell 32% anyway.

If you've ever watched a stock you follow crash the week after a record earnings print, you know the specific confusion that creates. You re-read the numbers. They're good. You check the price. It's down. Something isn't lining up, and most people just close the tab.

Don't close the tab. This is the whole game, and it's worth five minutes to see clearly.

What actually happened

On June 24, Micron reported $41.46 billion in quarterly revenue. Up 346% year over year.

Non-GAAP earnings: $25.11 a share. Gross margin: near 85%. Every one of those numbers is a company record.

The stock popped 15% on the print and hit a new high near $1,255. Then it round-tripped the entire move. Today it sits near $859.

Two weeks later, Samsung reported its own record profit. Its stock fell too. So did SK Hynix. So did SanDisk. The entire memory sector got marked down together, on good news, because the market asked a different question than "were the earnings good." It asked: is this margin sustainable, or is this 2021 semiconductor euphoria wearing an AI costume. Michael Burry disclosed a put position near $1,052 on July 1. The loudest sound in this stock all quarter wasn't the earnings call. It was that one skeptical headline.

Call this the euphoria tax. A stock doesn't fall because the business got worse. It falls because the price already contained a bet that everything would go perfectly. There was something in the air in June: nobody left in the room asking what could go wrong. That's usually the exact moment something does. That's not a verdict on Micron. It's a verdict on the crowd that bought it in June.

The numbers that actually matter

DRAM revenue: $31.3 billion. A record. 76% of the business.

NAND revenue: $9.9 billion. Also a record.

Data center revenue crossed $25 billion for the quarter, a $100 billion annualized run rate. HBM4, the memory chip every AI accelerator needs, shipped over $1 billion in the quarter and is ramping twice as fast as its predecessor did. CEO Sanjay Mehrotra called supply "structurally constrained in its growth and ability to meet industry demand." Guidance for next quarter: $50 billion in revenue, $31 in earnings per share, 86% gross margin. Records again, before they've happened.

Here's the detail that actually changes the risk profile. Micron has signed 16 Strategic Customer Agreements, covering roughly 20% of its DRAM volume and a third of its NAND volume. Those contracts lock in $100 billion of minimum revenue at floor prices through 2030. The last two memory busts, 2019 and 2022, happened because the entire business was spot-market pricing, and spot prices collapse the moment supply catches demand. A third of this business no longer works that way.

Here's the bitter half, so you're not only tasting the sweet part. CFO Mark Murphy already flagged that price increases are slowing. New DRAM and HBM capacity from Samsung and SK Hynix is coming online. If hyperscaler AI spending genuinely pauses instead of just decelerating, the spot-exposed two-thirds of the business will feel it before the contracted third does.

The five scenarios, priced from $859

Weight those honestly and the math still lands positive. Probability-weighted, this comes out near +29% over 12 months. Wall Street's own consensus rating sits at Strong Buy, average target $1,486. The market already half agrees with the bull case. It just hasn't admitted it out loud yet.

One thing I'm not going to hide from you

I run one entry rule for the Compounding Portfolio: weekly RSI has to confirm oversold before daily RSI times the exact day. Right now weekly RSI on MU sits near 67. Not oversold. This stock is still digesting a roughly 3x run from last year's lows, and 14 weeks of losses hasn't dented that yet. Daily RSI is oversold, near 25, but that's not the signal my own rule is built to wait for.

So this isn't a textbook entry. It's a drawdown-plus-fundamentals case: a business posting records, guiding to more records, with a third of its revenue locked at floor prices through 2030, trading 32% below a high it hit five weeks ago. I'm treating the size of the move and the strength of the print as the signal, not the weekly chart. That's a conscious call, not the default play. A gambler needs the chart to agree with them before they'll admit it. A probabilistic investor writes down why they're overriding the rule, and checks the math either way.

You don't need to predict the AI cycle to run this. You need the earnings release, five scenarios, and the discipline to price all five instead of just the one you're hoping for. Run it again next earnings. Then the one after that. The edge isn't the call, it's doing this every single time.

That process is exactly what we build inside Sprint Club, on every position, every week. Start your free 7-day trial: https://www.strategysprints.com

Ready to accelerate now? Book a Discovery Call: https://calendly.com/strategysprint/discovery-call

Happy hunting.
Simon & The Sprinters 🐬⚡️🐆