Market movers

These are not tips — here's what the data actually shows

Every tracked runner with a recorded move, 19 days from Saturday 11 July to Tuesday 4 August. “Expected” is the market's own price with the bookmaker's margin stripped out.

Read the first two rows against the third, not against zero. Every runner we track beats its de-vigged price by about half a point — most likely an artefact of how the margin is stripped out rather than a real edge — so a steamer beating its price by roughly a point is only a fraction of a point of genuine signal. Measured against an average margin of about two points per runner, that is comfortably inside what the bookmaker charges to take the bet, which is why backing steamers as a rule has not been profitable here. The steamer effect is well inside the noise.

The drifter row is the more interesting one: consistent in direction — negative on 9 of the 11 days — but small, and at the edge of statistical significance.Two caveats we'd rather state than bury: the effect switches on at about 10% and does not deepen beyond it, so smaller drifts of 5–10% are not a negative signal at all; and it lives at shorter prices. Drifters beyond roughly 15/1 show nothing, because there is barely any win probability left to lose.

We publish both because knowing where the money has gone is useful context when reading a card — not as selections. Our actual picks are on the tips page, with the full record on results.