Seven — 004

The 20-Second Version

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

In this issue: The Market · Stock of the Week: Alphabet, Part 2 · The Setup · Options Desk · The Number · Risk on the Table · What I Learned

I want to start this issue the way most weeks on Wall Street actually feel right now: a little dizzy. Not scary dizzy. Just the kind where you glance at the Dow, see a fresh record, glance at the Nasdaq an hour later, and see it giving back ground, and you have to remind yourself both things are true at the same time.

That was basically the whole week. The S&P 500 finished up about 1.2%, the Nasdaq added a bit more than 1%, and the "Magnificent Seven" mega-cap tech names did a lot of the heavy lifting, up roughly 4.9% as a group. The Dow actually slipped around 0.5% on the week even though it touched fresh intraday records along the way. That gap between "the market is at all-time highs" and "the Dow is actually down" is the first thing worth sitting with. Different indexes are telling different stories depending on how much AI-and-chip exposure they're carrying, and this week that exposure cut both ways.

Monday set the tone. The Dow broke 53,000 for the first time ever, and the President rang the opening bell (virtually) to launch the new "Trump Investment Accounts" program. Semiconductor equipment names like Lam Research, Applied Materials, and KLA jumped after Morgan Stanley raised price targets across the group. It felt like a continuation of the prior week's strong rally, nothing more.

Then Tuesday flipped the script. The Dow pulled back slightly, but the real story was underneath the surface: chipmakers sold off hard, with Micron down almost 5%, as traders rotated out of the most crowded AI trades. "Expectations are up, and fundamentals are struggling to meet these high sky-high demands," one analyst put it, and that line stuck with me because it's really the tension running through the entire AI trade right now. Rivian also grabbed headlines, falling more than 10% premarket after announcing a 75-million-share stock offering, just days after a strong run higher. It's a good reminder that a rallying stock can still get hit hard the moment a company decides to raise fresh capital by issuing more shares. Dilution doesn't care how good your week was.

Wednesday brought the Fed into the picture. The FOMC released minutes from its June meeting, the first held under new Fed Chair Kevin Warsh, and traders spent the day parsing every line for clues on where rates go next. Layered on top of that was a genuinely tense geopolitical backdrop: renewed US-Iran hostilities around the Strait of Hormuz, just as ceasefire talks were supposedly underway. Oil markets and equity markets were both trying to price the same uncertainty, and it showed in a choppier session.

Thursday, the mood swung back toward calm. Jobless claims came in at a four-week low, oil prices eased, and chip stocks rallied hard again, Micron up 4.5%, SanDisk up more than 7%. The VIX (the market's so-called fear gauge) dropped over 6% that day alone, a sign that whatever anxiety built up midweek was draining back out fast.

Friday closed things out with two stories running side by side. First, SK Hynix, the South Korean memory chip giant and a key Nvidia supplier, made its US stock market debut, raising $26.5 billion, the largest-ever US listing by a foreign company, and popped more than 13% on day one. Second, Meta had an outstanding week, jumping around 6% on Friday alone and roughly 15% for the week, its best weekly performance since early 2024. That move came on the back of a Bank of America note suggesting Meta may have found a way to build AI data center capacity far more cheaply than Wall Street assumed, alongside a well-received debut of its newest AI model. If that cost story holds up, it matters well beyond Meta. It would suggest some of the "AI is too expensive" fear that's been rattling chip stocks all week might be overstated, at least for the companies that figure out the efficiency piece first.

So where does that leave us heading into next week? Big bank earnings kick off Tuesday with JPMorgan and several others reporting, and semiconductor heavyweights ASML and Taiwan Semiconductor also post results. And there's a bigger theme sitting underneath all of it. The CME FedWatch Tool currently shows a 69.1% overall probability of a rate hike happening at or by the September FOMC meeting, with the September meeting itself pricing in a 64.3% total hike probability, split between 51.2% for a 25 basis point move and 13.1% for a full 50 basis points. Only 35.7% odds sit on rates simply staying put. That is a meaningfully different environment than what most of us got used to over the last couple of years, when every conversation was about when the Fed would cut, not raise. Keep an eye on that theme. It's going to come up again, probably as soon as this week's bank earnings give us a fresh read on how the economy is actually holding up under this pressure.

The previous issue, we spent our time with Alphabet's business and moat. If you skipped that one, here's the one-line version: this is a company with a genuinely dominant, multi-pronged business (Search, Cloud, YouTube) that's now layering a real AI advantage on top through Gemini and its own TPU chips. Good company. The question we're answering this week is the one that actually determines whether it's a good stock to own from here: what are you paying for that quality, and is it worth it?

Running Alphabet through the 8-pillar framework tells a clean story, maybe the cleanest I've seen us score. Six of the eight pillars come back green. Shares outstanding have shrunk 8.34% over five years thanks to steady buybacks, a real tailwind for per-share value. Five-year ROIC sits at 18.74%, more than double our 9% threshold, meaning Alphabet isn't just growing, it's growing profitably on the capital it reinvests. Revenue, net income, and cash flow have all grown at a healthy clip over the trailing five years. And the balance sheet pillar, debt relative to five-year free cash flow, comes in at a conservative 1.14x, well under the 3-4x danger zone. Alphabet is carrying about $77.5 billion in total debt against roughly $126.8 billion in cash and short-term investments. That's not a company managing debt, that's a company sitting on a net cash position of around $49 billion, or somewhere in the neighborhood of $4 per share. Worth remembering, because it's exactly the kind of thing the analyzer tool won't account for on its own. It only looks at operating cash flows, so any balance sheet cushion like this has to get added back manually to whatever fair value number the tool spits out.

The two reds are both valuation pillars, and they're the ones that matter most for a decision on price. The five-year P/E comes in at 44.78, roughly double our 22.5 threshold, and five-year Price-to-Free-Cash-Flow lands at 64.23, nearly triple its own 22.5 ceiling. Worth being precise about what that "5-year" framing actually means here: it's not an average of the stock's historical P/E readings over time, it's today's price measured against the company's average earnings and free cash flow over the last five years. That distinction matters, because Alphabet's TTM P/E on today's actual trailing earnings is closer to 27, and its forward P/E (using next year's expected earnings) sits around 24. The stock isn't wildly expensive on where earnings are today. It's expensive relative to where earnings have averaged over the last five years, which is a very different statement. Alphabet's earnings power has grown so much, so fast, particularly the last two years, that measuring today's price against a five-year average earnings base makes the stock look more stretched than a simple look at current earnings would suggest.

That's really the tension in owning Alphabet right now. This is pillar-1-and-8 red for a reason connected directly to the AI story, not a sign the business is broken. Free cash flow itself has actually come under pressure lately, TTM free cash flow of $64.43 billion is below the five-year average of $67.86 billion, because Alphabet is plowing an enormous amount of capital into AI infrastructure, reportedly $180 to $190 billion in capex guided for 2026 alone. You're being asked to pay a premium multiple on cash flow that the company is voluntarily suppressing right now by reinvesting aggressively into the thing most people believe is going to drive the next decade of growth. If that bet pays off the way Search did twenty years ago, today's multiple will look cheap in hindsight. If it doesn't, or if it takes longer than the market's patience allows, that 64x Price-to-FCF pillar is exactly where the pain shows up first.

So where does that leave us? A great business, by the numbers, that also happens to be a genuinely expensive stock by its own five-year history, for a reason that's at least defensible: heavy, deliberate reinvestment into AI infrastructure rather than a business losing its footing. That is a very different kind of "expensive" than a company propped up on hype with nothing underneath it. Whether that premium is one you're comfortable paying is the real question Part 2 of this deep-dive is meant to leave you with, and it's one worth sitting with rather than rushing past.

Naming the tension is one thing. Actually putting a number on it is more useful, so let's build one, using a few different lenses side by side rather than leaning on any single formula.

Start simple. If the market priced Alphabet exactly at our 22.5x pillar threshold against today's actual trailing earnings ($13.24 EPS), that's about $298 a share. Add back the roughly $4 per share of net cash sitting on the balance sheet (remember, this is a net cash position, not net debt, and the analyzer tool won't do this step for you), and you land around $302. Do the same math against next year's expected earnings instead of today's, and that climbs to roughly $334.

But a flat 22.5x is a generic ceiling, and Alphabet isn't a generic business. A company compounding at an 18.74% five-year ROIC, double our pass threshold, with a fortress balance sheet, has earned some right to a premium above the baseline. Give it a 26x-28x multiple on forward earnings instead, and the range moves up to roughly $385-$415 with the cash adjustment included.

Then there's the check that actually matters most here: free cash flow. Even generously assuming Alphabet's FCF normalizes back toward an 85% conversion of net income once this AI capex supercycle eases, rather than the roughly 40% conversion it's running today, that only works out to about $11.26 per share in "normalized" free cash flow. Running that same 22.5x-to-28x range against that number puts fair value closer to $262-$323, cash included, meaningfully below every earnings-based estimate above.

Put it all together and you get two honest answers, not one tidy number. On earnings, a defensible range runs roughly $300 to $415, depending on how much moat premium you're willing to grant. On free cash flow, even a generous recovery scenario tops out closer to the low $300s. At today's price around $358, you're sitting comfortably inside the earnings-based range, but you're already past what the cash-flow lens can justify unless that capex spend converts into durable earnings power the way Search once did.

If I had to compress this into one usable framework: $300 to $335 is the disciplined, no-benefit-of-the-doubt entry zone. $360 to $415 is what you're implicitly paying for if you believe the AI reinvestment thesis plays out. This is a way of thinking about the price, not a price target, and reasonable people will land in different spots on it depending on how much they trust that thesis. As always, none of this is a recommendation to buy or sell anything, just one way to frame the decision for yourself.

The fundamentals tell you what Alphabet is worth. The chart tells you what the market is currently willing to pay for it, and right now those two things are having a bit of an argument.

Alphabet peaked at an all-time high of $404.47 back on May 18, and has spent the two months since correcting off that high, currently trading in the high $350s, off around 11-12% from the top. That pullback lines up almost perfectly with the story we just walked through in the valuation section. Q1 earnings back in April showed genuinely excellent operating results, Cloud revenue up 63% with a $462 billion backlog, Search hitting record query volume on the back of AI Overviews, but the same report raised 2026 capex guidance toward $190 billion, which dragged free cash flow margin down from roughly 21% to about 9%. The market has spent the last several weeks digesting that trade-off, and the chart is a pretty faithful record of that digestion.

Zoom into the moving averages and you can see the tug of war directly. The stock is sitting above its 100-day ($340) and 200-day ($318) moving averages, both comfortably, which keeps the longer-term trend intact and bullish. But it's currently trading below both its 25-day ($357) and 50-day ($370) averages, the shorter-term signals, which is exactly what a stock in the middle of a corrective pullback inside a bigger uptrend looks like. Momentum indicators tell a similar story: RSI has cooled into the mid-50s, solidly neutral territory rather than the overbought reading it was flashing back near the highs, and the MACD has been running negative recently, meaning short-term momentum has faded even while the longer trend holds. None of that is alarming on its own. It's a stock catching its breath.

The levels worth watching from here: resistance sits in the mid-$370s, roughly where that 50-day average also lives, so a close back above that zone would be a meaningful signal that the correction is over and the longer uptrend is reasserting itself. On the downside, support sits in the low-to-mid $330s, and a break below that opens the door toward the $310 area, which is roughly where the 200-day average and a longer Fibonacci retracement level both converge. That confluence of technical support with the 200-day trend line is exactly the kind of level that tends to hold, or exactly the kind of level that causes real damage if it doesn't.

The single biggest catalyst ahead is Alphabet's Q2 2026 earnings report, scheduled for Wednesday, July 22, after market close. Wall Street is currently modeling around $2.90 in EPS on roughly $116.85 billion in revenue. Worth being precise here, since we've gotten earnings dates wrong before: some secondary sources floated a July 29 date, but two separate primary-adjacent sources (Google's own earnings page and stockanalysis.com) both confirm July 22, so that's the date we're working from. That report is going to be the real referee in the argument this section opened with. If Cloud's growth and backlog conversion keep accelerating the way they did last quarter, the market may decide the capex spend is buying real, durable growth and reward the stock. If the numbers show the AI investment cycle biting into margins with less to show for it, expect that $330 support level to get tested in a hurry.

Zoom out and the setup here is a genuinely uncrowded kind of interesting: a business executing well, correcting on a fundamentals-driven reason rather than a broken chart, sitting right at the pivot point between resuming its uptrend and testing a much more important support level below. That is precisely the kind of setup that tends to produce a real move, in one direction or the other, around a known catalyst date. It's also exactly the kind of setup we'll be building this week's Options Desk trade idea around.

Every options price has two ingredients baked into it: where the market thinks the stock might go, and how confident the market is about that guess. That second ingredient is implied volatility, or IV, and there's no moment in a stock's calendar where IV behaves more predictably than the days around an earnings report.

Here's the mechanic. Alphabet's options right now are trading with real, elevated implied volatility, an IV Rank of 68% and an IV Percentile of 88.89%, meaning current premiums are richer than they've been on nearly 89% of trading days over the past year. As July 22 approaches, that IV isn't going to sit still. Options expiring right after earnings will get progressively more expensive relative to options further out, purely because the market knows a big, binary, one-day repricing event is coming and wants to be compensated for the uncertainty. This is called volatility term structure going into backwardation: near-dated options carrying a temporary premium over longer-dated ones. Alphabet has historically moved an average of roughly 4.8% in the single trading session after reporting, with individual quarters ranging anywhere from a slight drift to a double-digit swing in either direction.

Then, the instant the earnings numbers hit the tape, that uncertainty resolves. Whether the stock moves 1% or 8%, the specific reason options were priced richly (not knowing what would happen) disappears in a single moment. IV collapses, often within minutes of the print. That collapse is what traders call IV crush, and it is one of the most important, least intuitive lessons in options trading: you can be completely right about the direction the stock moves and still lose money on a position you bought right before earnings, because the drop in IV can outweigh the gain from being correct. This is exactly why "buy a call right before earnings because you're bullish" is one of the most common ways beginners lose money on options that otherwise had the right thesis.

Given everything in The Setup, here's the trade worth walking through. Alphabet is sitting on top of a well-defined support zone in the low $330s, right where a longer Fibonacci retracement level and the 200-day moving average both converge, while the longer-term trend above that level remains intact. The thesis isn't "Alphabet moons after earnings." It's narrower and more disciplined than that: "the stock holds above its major long-term support level through the report, regardless of exactly how big the earnings pop or drop turns out to be."

That thesis is a textbook fit for a put credit spread. Sell the $335 put, currently showing a delta of -0.24, and buy the $325 put, currently showing a delta of -0.16, both expiring the week of July 31, a little over a week after the report. Selling premium here means we're on the right side of the IV crush we just talked about instead of the wrong side of it. At current mid-market pricing, the $335 put is worth about $5.28 and the $325 put about $3.25, for a net credit of roughly $2.03 per share, or $203 per contract. Worth noting for real-world execution: the bid-ask spreads on these strikes are running about $0.60-0.70 wide right now, which is typical this close to an elevated-volatility event, so an actual fill may land a touch below that pure mid-price credit. As long as Alphabet closes above $335 at expiration, the full credit is ours, crush or no crush.

Using our POP shortcut (POP ≈ 1 minus short delta), a short put at -0.24 delta implies a probability of profit of roughly 76%. Max loss is capped at the $10 width of the spread minus the credit collected, or about $7.97 per share ($797 per contract), against a max profit of that same $203 credit, a roughly 25.5% return on risk. Breakeven sits at $332.97, just below the technical support zone we identified in The Setup, which is exactly where you'd want a defined-risk thesis like this one to have some room to breathe.

The other trade that comes up naturally when a stock has a scheduled binary event and a history of double-digit moves is a long strangle, buying an out-of-the-money call and an out-of-the-money put simultaneously, betting on a big move without needing to guess the direction. It's tempting, especially with Alphabet's own earnings history showing quarters with 7%, 9%, even 10% single-day swings.

Here's why it doesn't make this issue's cut. A long strangle is a bet that the realized move will be bigger than what's priced into the options, and right now, with IV Rank at 68% and IV Percentile at 88.89%, options premiums are richer than they've been on nearly 89% of trading days this past year. You can see exactly how rich in the actual numbers: the at-the-money straddle for this same July 31 expiration, the $355 call plus the $355 put, currently costs about $28.48 combined at mid-market. That prices in breakevens of roughly $383.48 on the upside and $326.52 on the downside, meaning the stock would need to move about 7.1% higher or 8.8% lower just for this position to break even by expiration. Alphabet's historical average post-earnings move is only around 4.8%. The options market isn't just pricing in a real event here, it's pricing in a move meaningfully larger than what this stock has typically delivered, even across some genuinely surprising quarters. Buying that premium means needing an outsized move just to get back to even, before IV crush even enters the picture. It's a structurally harder trade to win than it looks, precisely because of the IV mechanics we just walked through, and it's a good example of a trade that sounds intuitive but fights the math of the moment it's entered in. The put credit spread above is built to collect that same rich IV premium instead of paying it.

$358 vs. $428 vs. $300-335.

That's Alphabet's current price, against Wall Street's average 12-month analyst target of roughly $428-432, against the disciplined, no-benefit-of-the-doubt fair value range we built earlier in this issue using the 8-pillar framework and a strict earnings multiple.

Three numbers, three different stories. Wall Street's consensus target implies about 20% upside from here, essentially betting that Alphabet's AI reinvestment cycle converts cleanly into durable earnings growth, the bull case in full. Our own disciplined framework, built off actual trailing and forward earnings rather than a growth narrative, lands meaningfully below today's price, the same tension we spent the whole valuation section walking through: a great business, priced today for a future that hasn't been proven out yet in the free cash flow numbers.

Sitting between those two is the number that actually matters most day to day, the $358 the market is paying right now. It isn't at either extreme. It's a price that only makes sense if you believe the AI capex story resolves the way Search's build-out did twenty years ago, but a price that also hasn't fully priced in Wall Street's more optimistic view either. That gap, between what you'd pay with no benefit of the doubt and what the Street is willing to pay assuming the thesis plays out, is the entire debate on Alphabet right now, compressed into three numbers. Keep it in your pocket heading into the July 22 report. Whichever direction the stock moves that day will tell you which of those three numbers the market is starting to believe.

Risk #1: This trade is deliberately held through earnings, and that's the whole point and the whole danger.

Most options education tells beginners to close positions before an earnings report to avoid the binary risk. We're doing the opposite here on purpose, because a defined-risk credit spread is structurally built to benefit from the very event most traders are told to avoid. But "defined risk" doesn't mean "small risk." If Alphabet's Q2 report disappoints on the two things the market is watching most closely, Cloud backlog conversion and any further increase to that $190 billion 2026 capex number, the stock can absolutely break below our $332.97 breakeven in a single session. Alphabet's own history includes single-day moves as large as 9-10% in either direction. An 8.8% drop from today's price would put the stock right around $326, past our breakeven and eating meaningfully into the spread before it even reaches our $325 long strike. This isn't a trade for someone uncomfortable holding a position through a knowable, scheduled binary event.

Risk #2: the $330-335 support zone is a good technical level, not a guarantee.

We built this trade around a real confluence, the 200-day moving average near $318, a longer Fibonacci level, and recent price memory all clustering in the low $330s. Confluence like that tends to hold more often than a level with no supporting evidence behind it. It is still not a floor with a contract attached. If broader tech sentiment sours going into the print, the kind of AI-cost-efficiency-doubt narrative that hit chip stocks earlier this same week in The Market section, Alphabet can trade through that zone on momentum alone, independent of what its own earnings actually say. Watch how the stock behaves in the days directly before July 22. A break below $335 before earnings even happens would be an early signal to reconsider the trade entirely, not wait for expiration to find out.

Risk #3: sizing, and why the max loss number is the one that actually matters.

The max loss on this spread is $797 per contract, against a max profit of $203. That asymmetry, risking roughly four dollars to make one, is completely normal for a high-probability credit spread, but it means position sizing has to be disciplined rather than a hunch. A common rule among premium sellers caps any single defined-risk trade at 1-5% of account value. At the conservative end of that, a $797 max loss implies an account of at least roughly $16,000 to $80,000 depending on where in that range you sit, before this trade is sized appropriately. The credit collected should never be the number that decides how many contracts to run. The max loss should. And because this is a defined-risk spread held through a known event, there is no "rolling the position to fix it" if Alphabet gaps hard against us on July 23. Defined risk should mean defined, full stop, even when the outcome disappoints.

The moment that actually stuck with me this week wasn't a number, it was realizing two completely different things can share the exact same name.

When I first pulled Alphabet's "5-year P/E" and saw 44.78 against our 22.5 threshold, my instinct was to treat it as confirmation the stock was expensive, full stop. Then I went looking for the same figure elsewhere to double check it, and found other sources putting Alphabet's 5-year average P/E closer to 22-23, seemingly a completely different verdict. My first reaction was that one of the numbers had to be wrong. It took slowing down to realize both were right, they were just answering different questions. One measures today's price against five years of average earnings. The other averages the stock's own historical P/E readings over time. Same three words, "5-year P/E," genuinely different math, and a very different story depending on which one you're holding.

That's a small thing, but it's exactly the kind of small thing that separates actually understanding a metric from just knowing where to find it on a website. Anyone can pull a number off a screen. Knowing what question that number is actually answering, and being suspicious enough of a clean, round-sounding stat to go check it against a second source, is the muscle I'm actually trying to build with this newsletter. I got lucky this week that the discrepancy was big enough to notice. I'm sure there are others sitting quietly in past issues that weren't.

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 sitting with two prices that can't both be right, Wall Street's and my own, and learning that even a number as familiar-looking as "5-year P/E" can quietly mean two different things depending on who's doing the math.

See you next Sunday.

— 7even Seven — 004