Seven — 002

The 20-Second Version

Full breakdowns below. Not financial advice — see the disclaimer.

A brutal week for tech, and a strange one underneath. The S&P 500 fell nearly 2% to close around 7,354, the Nasdaq dropped 4.6% to 25,298 — its fifth straight losing session — yet the Dow actually rose about 0.6%. That split is the whole story: money didn't leave the market this week, it rotated out of expensive AI names and into healthcare, industrials, and financials. When the index falls but most stocks don't, you're watching a few mega-caps do the damage.

And here's the irony that defined the week. The best earnings report in the entire AI complex landed during the selloff. Micron (MU) reported Wednesday: revenue of $41.46 billion against roughly $35.7 billion expected, EPS of $25.11 against ~$20.5 — revenue up 345% from a year ago. Then management guided next quarter to a record $50 billion in revenue and $31 EPS at ~86% gross margins, with their high-bandwidth memory sold out through 2027 and the shortage expected to run into 2028. The stock jumped ~15% after hours toward $1,200, pushing Micron past a $1.16 trillion market cap. A memory company. That is how violent the AI buildout has become.

So why did the market fall with a report like that? Because the same memory shortage minting money for Micron is now a cost line for everyone who buys memory. The day before, Apple fell ~6% and Microsoft ~3% after both announced price increases — MacBook, iPad, Xbox — citing rising memory costs. Read those two events together and you get the real signal: the AI boom's biggest winner (memory) is becoming its customers' biggest squeeze. Micron's windfall is Apple's margin problem. That's what dragged the index down even on great news.

The economic print was PCE, the Fed's preferred inflation gauge, out Thursday. Headline hit 4.1% year-over-year — the highest since 2023 — with core near 3.4%. But both came in roughly in line with forecasts, and the month-over-month read was actually a touch soft, so it confirmed "higher for longer" rather than shocking anyone. The more important shift: the market is no longer debating rate cuts. It's pricing rate hikes — about three this year, with roughly 60% odds the first lands in September. With the funds rate already at 3.50–3.75%, that's a meaningful hawkish turn.

Which brings us to the war, and the part that actually matters for your portfolio. The Middle East conflict matters to markets mostly through one channel: oil → inflation → yields. When Iran shut the Strait of Hormuz earlier this year, oil carried a war-risk premium that fed straight into headline inflation. This week that premium unwound — tankers are flowing again under a U.S.–Iran agreement, and crude erased its entire wartime gain, with WTI back near $69 and Brent under $73, right where they sat before the war began. Falling oil is disinflationary, and that's exactly why the 10-year Treasury yield fell to about 4.38% this week even as inflation printed a 3-year high. The short end (the 2-year) rose on hike bets while the long end fell on cheaper oil — the curve flattening in real time. (It's fragile, though: a drone attack on shipping near Hormuz on Friday is a reminder this can re-price fast.)

The setup into July: with Apple, ASML, TXN, GOOG, INTC, MSFT, and TSM all reporting this month, Micron may have been the opening act. The question the whole tape is now asking — how much are the Mag 7 actually paying the chip and memory makers, and can those numbers keep impressing a market that's started punishing the buyers of all that hardware?

This week I wanted to sit with a question I keep coming back to: what is a reasonable price to pay for an exceptional company? Amazon is the perfect place to ask it, because it is genuinely one of the highest-quality businesses in the world — and it's far stranger under the hood than most people realize.

Most people know Amazon as an online store. That's the part you can see, and it's the least interesting part of the investment. Amazon is really five independent businesses driving one company — like a car with five engines instead of one. Very few companies have even one engine of this magnitude; Amazon has five:

Here's the thing the storefront hides: the online store is massive but low-margin. The businesses Amazon investors actually value are the high-margin ones bolted onto it.

Financial quality. Amazon is a revenue machine — roughly $742.8 billion over the trailing twelve months, and $181.5 billion in Q1 2026 alone (up 17% year-over-year). But the single number means nothing until you see where it comes from. Growth is now spread across cloud, ads, subscriptions, and marketplace — not just retail:

Market cap is about $2.56 trillion, with an enterprise value near $2.90 trillion — a ~$340 billion gap. That sounds alarming until you understand what it's made of. Amazon carries over $100 billion in cash against roughly $210 billion in debt, so net debt is only about $109 billion. Set that against a cash-generation engine this size and it's trivial: a severe rough patch would slow Amazon's AI ambitions, not threaten its survival. Wall Street treats Amazon as one of the safest companies in existence, and the balance sheet is why.

One important framing: value Amazon on free cash flow, not earnings. Amazon deliberately reinvests enormous sums into AI chips, data centers, warehouses, robotics, and cloud — $43 billion of capex in Q1 alone. That spending depresses today's earnings on purpose. If those investments earn good returns, they show up as future cash flow. That choice is the entire investment debate, and we'll come back to it in The Number.

The moat. Amazon's competitive advantages are deep and reinforce each other:

The AI catalyst. This is the biggest long-term swing factor. Amazon is investing across AWS AI services, custom Trainium and Inferentia chips, foundation models, and enterprise AI infrastructure. Its chips business is already running above a $20 billion annual rate with triple-digit growth, and management has pointed to a very large book of AI-related revenue commitments. The question isn't whether Amazon is in the AI race — it's whether all that spending earns its keep.

Price action right now tells a patient story. Over the past year, Amazon has consolidated in a range roughly between $197 and $250, repeatedly pressing against resistance near $250 and getting rejected each time. That's not weakness — repeated tests of a ceiling often precede a breakout, and a clean move through $250 would be a meaningful signal.

For now, the stock at ~$232.69 sits in the lower half of that range and below its key shorter-term moving averages. Momentum reads as washed-out rather than broken: RSI around 39 (closer to oversold than overbought), CCI near -118, and MACD suggesting the stock is more undervalued than stretched. ATR is roughly $8, so a typical day moves the stock about that much — useful for sizing any position and for setting expectations into a catalyst.

And there is a catalyst: Amazon reports Q2 2026 earnings on July 30, after the close. Management already guided to net sales of $194–199 billion and operating income of $20–24 billion for the quarter, so the bar is set. With a stock coiled near the bottom of a year-long range and earnings two days into August's lineup of mega-cap reports, this is a genuine inflection point — the kind of moment where the next leg of the trend often gets decided.

So — is ~$232.69 a reasonable price for this business? I ran Amazon through my 8-pillar framework. Here's the honest scorecard:

On the surface, Amazon fails this checklist. But look at which pillars fail — and why. Every one of the failing pillars (P/E, Debt/FCF, Price/FCF) is built on near-term earnings or free cash flow. And we already said Amazon deliberately suppresses both by reinvesting tens of billions a quarter. The 8-pillar framework was designed for stable, cash-generative businesses; run it on a company in the middle of a historic capex cycle and the price-to-cash-flow pillars will always look insane. That's not the framework being wrong — it's the framework telling you exactly where the risk lives: you're not paying for the business as it earns today, you're paying for the business its capex is trying to build.

That reframes the valuation question. Compared with its own history, Amazon's earnings multiple is actually below long-term averages while earnings expectations have risen — which is why some analysts argue it's more reasonably priced today than in past years. And the Street agrees the upside is real: 12-month price targets cluster around $305–350 (well above today's $232.69), with the bullish names citing AWS acceleration and AI.

So we have a three-way disagreement, and that disagreement is the answer:

That last line is the whole thesis — and it's exactly what the next section is about.

Here's how I could be wrong about Amazon:

And one broader point worth holding onto, not just for Amazon: not every company wins the AI race. There will be winners and there will be losers — there always are. That's the entire case for diversifying across sectors and industries rather than betting the farm on a single AI narrative, however good the story sounds.

No trade idea this week. Instead, the model underneath every option you'll ever price.

What it is. Black-Scholes is a mathematical formula for estimating what an option should be worth. Start with the question it's really answering: what would I need to know to estimate an option's value? The model says you need five inputs (for a call or a put):

Walk through each one:

Stock price. A call gives you the right to buy at the strike. The higher the stock climbs above that strike, the more that right is worth. Strike $180, stock $200 → you can buy something worth $200 for $180, so ~$20 of intrinsic value. Higher stock → calls worth more, puts worth less.

Strike price. The strike sets how hard it is to profit. A $180 call is worth more than a $240 call on a $200 stock — the lower strike already has value baked in. Higher strike → calls worth less, puts worth more.

Time. Time is opportunity. A stock that needs to rise $10 has almost no chance by tomorrow but a real chance over six months. More time → generally more value, for both calls and puts, because more can happen.

Interest rates. Money today is worth more than money later. Owning a call lets you control the stock while delaying the cash to buy it, and you can earn interest on that cash in the meantime — so the delay is worth more when rates are high. Higher rates → calls worth a bit more, puts a bit less. (This is the smallest lever of the five — but notice it ties straight back to the Treasury-yield chain in The Market.)

Volatility — the one that matters most. Volatility is the expected size of future moves, not the direction. A stock that moves 8% a day carries pricier options than one that moves 1%, because bigger moves mean bigger potential payoffs. The key trap to avoid: volatility does not mean bullish — it means uncertainty. A big move up or down makes both calls and puts more valuable. Higher volatility → calls and puts both worth more.

For a trader, that's the ranking that matters: volatility first, then time, then stock price, then strike, then rates. It's why traders obsess over IV Rank, IV Percentile, and earnings volatility — and why options get expensive into an earnings date and then deflate the moment the result is out and the uncertainty collapses. (You watched this happen live this week with Micron.)

The thing to internalize: Black-Scholes does not predict direction. It's asking, given these five inputs, what's the probability this option ends up profitable — and converting that probability into a fair value for the opportunity. Not a forecast. A price tag on uncertainty.

This week I dug into the relationship between interest rates, bonds, and yields — and then watched it play out in real time.

Start with the mechanic that trips everyone up: bond prices and yields move in opposite directions. A bond paying $50 a year yields 5% at a $1,000 price. If everyone wants it and bids the price to $1,250, it still pays $50 — so the yield falls to 4%. If nobody wants it and the price drops to $800, the yield rises to 6.25%. Same coupon, opposite yield, driven entirely by price.

Why should a stock investor care? Because the Treasury yield is the risk-free benchmark — the return you can earn with virtually no risk. If a government bond pays 5% guaranteed, every stock now has to justify the extra risk of owning it. Higher yields make stocks relatively less attractive, and they bite growth stocks hardest: when yields rise, investors discount far-off future earnings more heavily, which shrinks the present value of companies whose profits are mostly years away. That's why high-growth tech is more rate-sensitive than a mature cash-generator — and, circling back, it's the same discounting that sits inside the interest-rate input of Black-Scholes.

Here's the chain I now keep in my head:

Inflation → Fed → Interest Rates → Treasury Yields → Bond Prices → Stock Valuations → Option Prices

When a headline hits, I try to locate where in that chain the news landed.

And this week handed me a perfect, slightly counterintuitive example. Inflation printed a 3-year high and the Fed turned hawkish — both forces that should push yields up. Yet the 10-year yield fell. Why? Because oil collapsed back to pre-war levels as the Strait of Hormuz reopened, and cheaper oil is disinflationary. The long end of the curve cared more about that than about the next Fed meeting, so it dropped — even as the 2-year rose on hike bets. The same news, hitting two different points in the chain, pulling yields in opposite directions. That's the lesson: the chain isn't a straight line you memorize once — it's a live system, and the skill is reading which link just moved.

Seven has always meant completeness. In the oldest stories, the world was finished in seven days — and the seventh was the day to step back and take in the whole of it. That's the promise of this letter: seven sections, one complete and careful look at a company, every single week. No shortcuts, no half-finished thinking — the whole picture, with precision, laid out plainly enough that anyone can follow it.

This week, completeness meant holding two true things at once — the best earnings report of the year and a market that fell anyway — and understanding why both belong in the same picture.

See you next Sunday.

— 7even Seven — 002

Disclaimer: SVN Research is educational commentary and reflects personal opinions only. It is not financial, investment, legal, or tax advice, and nothing here is a recommendation to buy or sell any security. Markets are risky and you can lose money. Always do your own research and consider consulting a licensed financial professional before making any investment decision.