The Wrong Newsletter

The Wrong Newsletter: Chasing the Wrong Wind

Week of May 17–23, 2026

Every once in a while, a week comes along that's a perfect storm of wrongness. Not a little wrong. All the way wrong.


100% of us predicted Friday's Leading Economic Indicators report would show another negative reading. That's not a statistic — that's unanimity.

The LEI came in positive: up 0.1% in April, breaking a months-long streak of contraction.

Recency bias is stubborn like that. Six months of declining data stops feeling like a trend and starts feeling like gravity — a permanent condition. It isn't. The signal that broke the streak was quiet: a rebound in stock prices, a small uptick in building permits. Not dramatic. But enough. We were so locked onto the drumbeat of bad news that we couldn't imagine it skipping a beat.


Ninety-four percent of us had the Yankees taking the subway series.

The Mets won it 2-1, on a walk-off in the 10th.

I've watched enough baseball to know that momentum is real but not magic. The Yankees came in with the confidence of recent form, and ninety-four percent of us just agreed with that narrative. Tyrone Taylor hit a three-run homer in the ninth to tie it, two Yankees infielders collided trying to make the final play, and Marcus Semien jogged home. Collisions between fielders always happen at the worst possible moment — for someone.


Eighty-five percent of us expected Google to drop Gemini 4 at I/O on May 19.

They announced Gemini 3.5 Flash and Gemini 3.5 Pro instead.

The AI community has a reflex: big keynote equals biggest possible announcement. It's understandable — Google I/O is the most public stage Google has, and Gemini 4 felt inevitable. But inevitability has its own timeline, not ours. The actual keynote leaned hard into agent frameworks and Android. Real progress, real announcements. Just not the headline we were writing before the curtain went up.


Three misses this week. Economic trend extrapolation. Momentum bias dressed as sports analysis. AI conference hype mistaken for a product roadmap. The pattern running through all three — certainty that hardened before the evidence was fully in.

The more unanimous the prediction, the more interesting it is when it breaks. That's the whole game, really.

Next week's questions are live. Come be wrong with us.


This week's resolutions

Culture

Markets

Politics

Sports

Technology

World