Field notes: the week the short thesis broke before the stock did

There's a pattern that shows up often enough in retail names that it deserves a name of its own: the thesis breaks days or weeks before the print does, and almost nobody notices until the stock has already moved.
Here's the mechanics of it, stripped down. A short seller builds a position on a comps story: traffic is softening, the chain has over-expanded relative to its trade area, management's guide is stale. The thesis gets written up, gets passed around a few desks, maybe shows up in a 13F footnote six months later. Borrow tightens and days-to-cover climbs. And then, for a stretch that can run four to eight weeks, the stock does nothing. It drifts, or even grinds higher on a market beta move that has nothing to do with the company.
That gap between a thesis being right and a stock agreeing with it is where most of the money gets made or lost on a short book, and it's almost entirely a data-timing problem. Short interest reports on a two-week lag through the exchanges. Management commentary comes once a quarter, filtered through whatever the IR team decided was worth saying out loud. App-based footfall panels cover a sample of shoppers, the ones who downloaded that particular app, and the sample skews toward whichever demographic uses it most. None of these are lies. They're just slow, or partial, or both.
What actually moves first, in the cases worth paying attention to, is the physical count. Lots don't lie about whether cars are showing up. A chain losing share to a newer format two exits down the highway shows it in the parking lot before it shows up in same-store sales, because the SSS number is an average across a fleet and the lot-level count is where the bleed is happening. We've watched this play out in slow-moving categories like home goods and off-price apparel, where the store base is large enough that aggregate comps smooth over a real divergence at the store level for a full quarter or more before it surfaces in the headline number.
The interesting part isn't that traffic leads sales. Every retail analyst already knows that, it's half of why channel checks exist. The interesting part is what happens when you can see the traffic trend and the short interest build on the same timeline, instead of one lagging the other by weeks. If short interest is climbing and the lot count is already flat or rolling over, the position has time on its side. If short interest is climbing while the lot count is still firm, the borrow cost is paying for a thesis that hasn't shown up in the physical world yet, and may not until a holiday quarter forces the issue.
That second case is the expensive one. It's also the one that's hardest to see with quarterly filings and a sample-based app panel, because both of those tell you about the past, not about the current week. A daily series that tracks every lot across a chain, run alongside the short interest tape instead of three weeks behind it, is a way to see which case you're in before the carrying cost makes the decision for you. That's the use case behind the service we built, and it's worth a look if the timing gap between data and conviction has cost you a position before.
The week the thesis breaks and the week the stock breaks are rarely the same week. Knowing which one you're in changes how much borrow you're willing to pay for the privilege of being early.
If the lag between what you believe and what the tape confirms has been expensive, go see what Sales Nowcast measures.