One horse. One bet. Every day. Sly Man’s read on the day’s card — the shortlist, the price, and the case for the pick.
Browse the logbookFive horses came through the filters steaming — and the engine couldn't split the top of the board cleanly, so today the market had the casting vote. The scan covered all declared runners for Wednesday 28 October 2026, applying an odds band of 2.0–50.0 decimal — no odds-on, nothing longer than 49/1. Field-size filter kept races with 6–14 declared runners, cutting out the noise at both ends. Feature weights ran at 1.6 for market_move, 1.5 for speed, 1.2 for form, 1.0 for class_change, and 0.8 each for trainer_strike and going. Draw bias is reserved for engine v2 and unavailable throughout.
The day opened with 38 races and 372 runners. The race-state check dropped none. The odds band trimmed that to 322. Field-size cut a further 66, leaving 256. The Lay of the Day collision filter removed 2 runners (from 6 excluded horse IDs), delivering a clean pool of 254 runners into scoring.
Six horses cleared the full pipeline. Mask of Zorro topped the board at 7.74/10 — an SR of 123 against a field mean of 109.7, a hot trainer in Joe Tizzard at 23% strike rate, and the strongest market steam on the list. Leo Phelps sat second at 7.40/10, with Owen Burrows running at a scorching 30% over the last 30 days and a sharp price cut to back it up, though form and going data are both unavailable. Scotto came in third at 7.13/10 — the only horse on the list with a going score attached (2-from-2 on good) and a pristine recent form string of '11', but it sits odds-on territory at 13/8 and carries a small class-rise penalty. Bridgetown Boy posted 7.07/10 almost entirely on the back of a market-move signal at 11/1, which flags it as the value standout; it is published separately as the value pick for today, not this bet. Centigrade scored 6.86/10 and Playhouse 6.83/10. The gap between first and sixth is 0.91 points — and critically, the gap between first (Mask of Zorro) and the selection (Centigrade) is only 0.88, with the top lead over second place sitting at 0.34 — below the 0.4 threshold the engine requires before calling a clear winner on score alone.
When composite scores cannot separate the shortlist — and today they cannot, the top lead is 0.34, under the 0.4 clear-lead threshold — the engine defaults to the shortest price on the board as a tiebreaker. The logic is straightforward: the market has done its own aggregation, and when my model is uncertain, I follow the sharpest collective read. Centigrade at 5/4 (decimal 2.25) is that horse. The speed score is the strongest signal here: an SR of 117 against a field mean of 94.6, a z-score of +1.78, which inside the weighted model contributes meaningfully to a composite of 6.86/10 — solid, not spectacular, but it doesn't need to be spectacular in a tied race. The form string '1/1681' is a touch uneven, but the latest position is a win, and the decay-weighted average comes in at +0.96, contributing a form score of +1.44 once the weight is applied.
The going score is the other piece I keep coming back to. Going is unavailable for most of this shortlist — Mask of Zorro, Leo Phelps, Bridgetown Boy, and Playhouse all have no career runs on their declared surface or adjacent ground. Centigrade has 2-from-2 on heavy or adjacent going — a full +1.00 going score. Nottingham is declared heavy today. That's a real edge when the rest of the field is untested in the mud. The market-move signal is there too — a steamer, strict majority of bookies shortened, none drifted — though at 5/4 the model's estimated win probability of 34.9% sits marginally below the market-implied 39.7%, a small negative edge of -21.4%. That negative edge is why this sits at MEDIUM conviction rather than high, and I'm not dressing it up.
The class-rise penalty is noted: today is a Class 1 Listed race, up from a Class 3 last time, and the model docks -0.60 for that. Ralph Beckett's trainer strike rate of 11% over the last 30 days is on par rather than hot, contributing a fractional -0.12. Neither of those readings troubles me greatly — the class rise is priced in at 5/4, and Beckett has run these Listed mile races in autumn before. What matters is the going advantage is real and documented, and the speed figure is the best in the field.
The negative market edge of -21.4% is the headline risk — the model thinks the market has this priced a shade tight, which is precisely why conviction stays at MEDIUM. The class rise from rank 3 to rank 2 (delta -1) adds a genuine question mark: a step into Listed company on heavy ground is a proper test, and the form string shows a 6, 8, and 1 in recent outings, meaning this horse has been beaten in between its wins. Going data is the one area of genuine comfort, but 2-from-2 is a thin sample. If the ground races faster than declared, the going advantage evaporates. Mask of Zorro, the highest-scoring horse on the shortlist at 7.74/10, runs at Newton Abbot today and is worth monitoring as a benchmark for whether the engine called the day right — it has the better composite but a longer price and different discipline entirely. And for those who want the long-odds read, Bridgetown Boy at 11/1 carries the most dramatic market-move edge on the board (+331.2%) and is the value pick published separately.
Centigrade · 5/4 · 1 unit · MEDIUM
The Fox’s Wire · Bet of the Day
The algorithm pulled five contenders from today’s card. One earned the ticket — the rest are what it beat. When each race settles, the finishes fill in so you can judge the shortlist, not just the selection.
Position only — payout-neutral. A process check, not a P/L. Click any horse to read its profile.
Value pick 4 of 24 won · +6.83 units at advised prices · since 11 October 2026. Kept separate from the bet’s record above.
