The Winners Kept Winning. The Board Kept Punishing Them For It.

10 Aug 2026 · DAILY NOTE

Three sessions ago we published a note about the week's biggest score moves: Palantir up 39.8%, its score falling 80 to 31. Chevron down 5.2%, its score rising 54 to 87. We called it the model doing the one thing it has ever been built to do — mark down what got expensive, mark up what got cheap.

We didn't yet know whether that would hold once the week actually finished playing out. It did, and more sharply than the first three days suggested.

The winners kept winning

By Monday, Palantir had added another 1.9%, taking its 60-session return to +31.0%. Unity added 0.2%, to +58.0%. Shopify added 2.4%, to +59.3%. Their scores barely moved further — they were already near the floor — but their entry plans crossed a line worth naming: all three now sit above their own first target. The trade plan the board would have published on the way in no longer exists for any of them.

The losers kept losing — into higher scores

The sharper story is on the other side. Chevron, the energy sector fund, the oil fund and Datadog all continued the moves that had already re-rated them the week before, and their scores gave back exactly what the extra price bought them.

TickerFri → Mon priceScore Fri → Mon
Chevron+4.5%87 → 74
Energy Sector ETF+4.7%92 → 70
US Oil Fund+6.7%96 → 84
Datadog+11.5%98 → 85

Thirteen points off Chevron. Twenty-two off the energy fund. Twelve off the oil fund. Thirteen off Datadog — given back from the model's own ceiling score, in a single session, because the price that earned it moved on.

Read plainly: the assets that had just become the board's strongest buys got more expensive relative to their own 20-day average, and the score corrected the same session it happened. That isn't the model being slow. It's the model doing exactly what it says it does, at the pace it says it does it.

What thinned underneath

Board-wide, the erosion that started the week before kept going. Median score: 82 on July 31, 77 on August 7, 76 on August 10. The count scoring Avoid: 2, then 13, then 14. Meanwhile SPY closed Monday within 0.03% of its own 52-week high.

Three straight readings, same direction, while the index itself sits essentially at a record. That isn't noise. Fewer names are carrying more of the tape, and the board is one of the only places that number gets tracked and dated rather than just felt.

The honest caveat

None of this says the strong names are wrong or the cheap ones are right. A score is a statement about price relative to its own recent average, not a forecast and not a business judgment. Palantir at 31 could keep running for months; Chevron at 74 could give back the whole re-rate next week. What the board can tell you, and what it told you accurately three sessions in a row, is which side of its own average each asset is standing on, and by how much. That's the whole claim. This week it held.

Scores in this piece come from the ZonedInOut model. How the score works →

Frequently asked

Why does a stock's entry-timing score fall even though the price keeps rising?
An entry-timing score measures how far the current price sits from the asset's own recent average (typically its 20-day moving average), not whether the company or the trend is good. When a stock keeps rising, it moves further above that average, which makes today a worse price relative to where it traded recently — so the score falls even as the price climbs. This is the model measuring price stretch, not making a directional prediction.
What does it mean when a falling stock gets a higher entry-timing score?
It means the pullback has brought the price closer to (or below) its own recent average, which historically has been a lower-risk point to consider an entry than chasing a stock at a fresh high. It is not a claim that the decline is over or that the stock is undervalued — only that the price has moved toward, rather than away from, its own short-term mean.
What is market breadth and why does it matter when a stock index is near a high?
Market breadth measures how many individual stocks are participating in a rally, as opposed to how the headline index (like the S&P 500) is doing. An index can sit at or near a record high while breadth deteriorates if a small number of large or fast-moving names are doing most of the work while the median stock lags or extends. Falling median scores alongside a rising or flat index is a sign of narrowing participation, historically associated with more fragile rallies.
Does a low entry-timing score mean a stock is a bad investment?
No. The score measures entry timing relative to the asset's own recent price history, not the quality of the underlying business. A great company trading well above its own 20-day average will score low, because the immediate entry is expensive relative to where it traded recently — not because the business is bad. The score is one input about timing, not a verdict on the company.

Get the Sunday Brief

One email a week, before the open: what the board did, which calls worked, which didn't, and what changed. Free. No spam, unsubscribe anytime.