Start here · Lesson 7 of 10

Why we rejected momentum

It is tempting to believe a stock's trailing trend predicts its next month. We tested that — and price/macro patterns, earnings surprises, insider activity, an earnings-in-window effect, and momentum itself — eight candidate predictors in all.

Each was measured out of sample against the base rate, with an overlap-aware confidence interval. Not one cleared the bar. The full table, with numbers and dates, is on the methodology page: see 'What we tested — and refused to ship.'

That is why we publish a base-rate probability with calibrated ranges rather than a stock-specific call. A track record of honest refusals is the case for everything else we show.

See also: climatologyThe base rate from history — how often this kind of situation resolved up versus down across many past years, before any stock-specific adjustment., calibrationWhether our probabilities match reality — of all the times we said an outcome was, say, three-in-five likely, did about three in five actually happen?