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 logbookOn a soft-ground Saturday at Haydock, the money moved with a conviction that the model couldn't ignore — and when you trace it back to a horse dropping in class with the best speed figure in its field, the read writes itself. Today's scan covered all 50 declared races and 512 runners, filtered through a pipeline that strips out odds-on shots (decimal below 2.0), long-shots beyond 50.0 decimal, and races outside the 6–14 runner field-size window where the signal-to-noise ratio collapses. After the odds band was applied the field fell to 345 runners; the field-size cut brought that to 318; a final check against today's Lay of the Day list removed three further runners — six horse IDs excluded — leaving 315 runners eligible for scoring. Feature weights were applied as configured: market_move carries the heaviest load at 1.6, speed at 1.5, form at 1.2, then class_change, trainer_strike, and going in descending order, with draw reserved for v2 and marked unavailable throughout.
The day opened with 50 races and 512 runners. The odds-band filter (decimal 2.0–50.0) cut that to 345. The field-size gate (6–14 runners) brought it to 318. The Lay of the Day collision check dropped a further 3 runners (from 6 excluded horse IDs), leaving a clean working universe of 315 runners — roughly 62% of the original card — going into composite scoring.
Five horses separated themselves at the top of the scoring table. Wild Clary led at 7.09/10, driven hardest by a maximum market-move signal, a standout speed reading, and a class drop. Homestrait came in second at 6.70/10 — another Haydock steamer with a superior speed z-score but a going record of 0-from-2 on soft that gives me pause. Thickthorn Tom posted 6.69/10 on exceptional recent form ('212111') and a hot trainer in Dan Skelton, but it's rising in class and the market edge is effectively nil at -10.5%. Soldante was fourth at 6.67/10 — a steamer in Kilbeggan, though class_change is unavailable and its going record on good ground reads 0-from-2. Jonny Concrete rounded out the top five at 6.66/10, a Class 3 Ascot runner with solid form but rising in class. Wild Clary pulled clear of the pack on the combination that matters most to me today: the single largest market-move score in the field plus the widest model-vs-market edge on the card.
The read I keep coming back to is the market-move signal on Wild Clary. A strict majority of bookies shortened, none drifted — that earns the full +2.00 weighted contribution at 1.6x, and it's the same maximum score posted by several runners today. What lifts Wild Clary above the rest is what sits behind it. The model estimates a 36.0% win probability against a market-implied 11.9% — an edge of +170.2%, the widest on the shortlist by a meaningful margin. That's not just the model talking loudly; it's the model and the money pointing the same direction simultaneously, which is the combination I trust most. The speed figure backs it up: an SR of 85 against a field mean of 65.8 gives a z-score of +1.47, contributing +0.53 after weighting — strong, not spectacular, but comfortably clear of the competition in this Class 4 sprint. The class drop is meaningful too: today's rank 5 against a last-time-out rank of 3 represents a delta of +2, and the engine scores that at +1.80 — a genuine drop into softer company. The form string '-42314' is imperfect, but the latest win carries a decay-weighted average of +0.80, and the going record — 1-from-4 with 4 placed on soft or adjacent — tells me this horse has been competitive in these conditions before, not just inconvenienced by them. The composite lands at 7.09/10, which I read as strong.
The one score that stands out as a drag is trainer Richard Hannon's current strike rate: 3/123 = 2% over the last 30 days, cold by any measure, contributing -0.80 to the composite. That's a real concern — a yard that isn't firing can blunt even a well-handicapped horse. The draw is unavailable across the entire card (feature reserved for v2), so I have no data on how the stall sits in a 6f Haydock sprint on soft, and that's a genuine blind spot — draw can matter significantly at this track over this trip. Homestrait at 6.70/10 is breathing close enough down Wild Clary's neck that a late market development in favour of the former would make me think twice; the gap between them is 0.39 points, not a chasm. Going can soften further through the day at Haydock in September, and while Wild Clary's soft record is a positive, there's no guarantee the horse handles extremes.
Wild Clary · 13/2 · 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.
