Eight up weeks, ten weeks of war, no trades

Friday close: the S&P 500 — the U.S. large-cap stock index most people picture when they say "the market" — settled at 7,473.47. Another record. Eighth weekly gain in a row, the longest streak since December 2023. Saturday morning my screener ran. I didn't buy.

The screener is the Saturday-morning filter I do every week — a set of quality and value rules I run across thousands of U.S. and Canadian stocks, surfacing a short list of names worth a closer look. This week it gave me more to look at than usual: a U.S. utility I've been tracking for months, a Canadian financial that had been brushing the edge of the list for weeks, and a mid-cap industrials name I hadn't seen before. All three looked closer to honest prices than anything I'd seen since winter.

I didn't buy any of them. Two things kept me watching instead of moving. One is the tape — eight straight winning weeks at valuations near a 140-year high is not the moment to add something new. The other is the bond market. Yields climbed all week. A high risk-free rate compresses what you can justify paying for equities, and I'm not convinced any of the three candidates have priced that ceiling in honestly yet. The watchlist got longer. The book didn't.

Two stories on the same front page

One half of Friday's front page is the streak. The S&P closed at a record, the Dow set a fresh record close, and Goldman Sachs put out a note warning that rising yields and inflation could trigger a correction. Eight winning weeks in a row is the kind of run financial papers write standalone columns about.

The other half is a war in its tenth week. Trump called off imminent strikes on Iran on Monday to make room for another round of nuclear talks. Both sides signalled progress. Both sides also stayed dug in on the Strait of Hormuz and Iran's enriched-uranium stockpile. Oil moved on every headline — Brent down more than 5% on the week, U.S. crude down more than 8% — settling near $103 and $97 a barrel by Friday.

Both stories run above the fold. The index isn't pretending the war is over. It's just decided the war isn't priced in yet.

The number I keep staring at

The Shiller CAPE — a long-window measure of how expensive the index looks against the past ten years of company earnings, named for the Yale economist who built it — printed at roughly 41.6 this week. In 140 years of U.S. market data, only one month has ever been higher: December 1999, at 44.19 — three months before the dot-com bubble peaked and the index went on to lose roughly half its value over the next two and a half years. The 1929 pre-crash high was 32. The 2007 pre-financial-crisis peak was 27.

That's the company I'd be joining if I added to a position into Friday's close. CAPE has been wrong about timing for years now — it's been screaming expensive since well before this rally started — but "expensive for a long time" and "honest entry price" aren't the same sentence. The screener I trust isn't surfacing honest entry prices into this tape. So I'm not adding.

How the portfolio did

The Blue Portfolio — the part of my book I run with intent, names I've picked one at a time and grade against a quality floor every week — finished up +0.56%. A quiet, slightly green five days.

The week's stinker was INTU (Intuit, the tax-and-payroll software giant), down −18.59% after the company reported earnings Wednesday afternoon. The print itself was fine — revenue up 10% year-over-year, full-year guidance raised. The reaction was about the second announcement: 17% of the workforce gone, roughly 3,000 jobs, plus a trimmed TurboTax forecast for next year. The market read it as "the AI restructuring has arrived and the growth story has to come from somewhere else now." I read it as a hold worth re-grading, not a sell.

The week's winner was AMD (Advanced Micro Devices, the chipmaker), up +10.24% on the kind of chip-sector enthusiasm I don't pretend to forecast. The position is small. A 10% move in either direction doesn't change a sizing decision.

The middle of the deck did what a Blue book is supposed to do in a quiet week — moved a percent or two in either direction and didn't make me look. MRK (Merck, the U.S. pharmaceutical company) was the second-best name, up +9.90% on a guidance bump.

What I'm watching

Whether the tape keeps climbing through bad news. Eight weeks of gains while the University of Michigan consumer-confidence print fell again, while a war drags into double-digit weeks, while CAPE sits at 41.6 — that's a market that's decided something I haven't decided. The next time it gives back a week, I want to see whether the screener's candidates start looking honest at lower prices, or whether the same expensive names show up at the same expensive multiples and the conservative read still holds.

Whether INTU's 17% layoff is a thesis crack or a one-print scare. The 10% revenue growth and raised guidance say the franchise is intact. The TurboTax trim and the staff cuts say management sees something softer coming. Both can be true. I'm holding for now and re-grading next Saturday.

One thing I'd do differently

The move I'm second-guessing is one I didn't take. On a quiet week into a record close, there's an argument for trimming a small slice off the richest names and letting the cash sit until the next opportunity. I didn't because nothing in the book is that rich yet. But the cash position is the thinnest it's been in months, and that's the lever I'd be reaching for if a real opportunity showed up Monday.

If you're holding into this

If you've been adding into this tape every week because every week looks like a win, this is the week to ask whether your screener still has anything new to say — or whether you've started overriding it with the chart. Mine didn't this week. That's the diagnostic I'm running. If yours is finding fresh names at honest prices in week eight of a streak that includes a ten-week war, you've got a tool I'd like to see.

Eight up weeks. Ten weeks of war. Zero trades. The discipline was the trade I didn't make.

— Mark

— Mark