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 logbookThe market and the model are pointing the same direction today, and when that happens I pay close attention. The engine scanned all 21 races declared for Saturday and applied a four-stage filter pipeline: first a race-state check leaving 186 runners in play; then an odds band of 2.0–50.0 decimal (no odds-on, nothing beyond 49/1) trimming the field to 158; then a field-size gate of 6–14 runners — small fields are noisy, big fields are chaos — bringing us to 116; and finally a lay-of-day collision check that dropped 1 runner, leaving 115 qualified runners to score. Feature weights running through the composite are market_move at 1.6, speed at 1.5, form at 1.2, class_change and going both at 0.8, trainer_strike at 0.8, and draw at 0.5 — though draw remains unavailable pending a v2 data layer.
The day opened with 21 races and 186 runners declared. The odds band alone removed 28 runners — mostly odds-on favourites and long-shots beyond the model's confidence range. The field-size filter cut a further 42, eliminating the very small fields that produce volatile outcomes and the large-field scrambles where model signal degrades sharply. One runner fell at the lay-of-day collision stage. The 115 survivors across the qualifying races form the pool from which today's shortlist was drawn.
The shortlist shakes out as follows. Minella Buoy at the top on a composite of 7.69/10 — a meaningful gap to second place. Prosperity Angel scores 7.01/10, driven by a big class drop and a strong market move, but the going feature is entirely unavailable on soft, which leaves a hole in the case. Eternal Angel comes in at 6.70/10 — solid form string and a steamer in the market, but rising in class (delta -1) is a penalty I can't ignore at this price. Sir Vinny posts 6.62/10 and has the strongest raw speed z-score of the whole shortlist, but the market is flat — no movement either way — and that absence of money signal in the highest-weighted feature costs him dearly. Motawaared at 6.51/10 and Bowdens at 6.44/10 both carry too many unavailable features and a going concern (literally, in Motawaared's case — 0-from-4 on soft) to get past the top two. What tipped the balance for Minella Buoy is the alignment of the two highest-weighted features firing together at meaningful levels.
The read I keep coming back to is this: the two features the engine weights most heavily — market_move at 1.6 and speed at 1.5 — are both firing hard on Minella Buoy, and they're saying the same thing. On market_move, the horse scores +2.00 — a strict majority of bookies shortened overnight, none drifted. That is the ceiling for this feature. On speed, the SR of 105 sits +2.17 standard deviations above the field mean of 72.6 — a z-score that only Sir Vinny beats on today's card, and Sir Vinny has no market signal to go with it. When the pace figure and the money are pointing the same direction, I'm interested. When they're the two highest-weighted features in the model, I'm very interested. Trainer John J Nallen brings a 27% strike rate over the last 30 days — the engine flags that as hot, scoring +1.27 on the trainer_strike feature. The form string '222' is not flashy — Minella Buoy has been placing without winning — but the decay-weighted average on the form feature scores +1.50, which is solid, and the most recent effort in the string is what the model leans on most. The composite of 7.69/10 is a strong score by any reasonable reading of this model's output range. The gap to Prosperity Angel at 7.01 is not enormous, but it is clear, and the quality of the signal underneath Minella Buoy's score is better — fewer unavailable features, and the two that contribute most are both maxed out.
The going feature is a quiet concern. Minella Buoy is 0-from-3 on good or adjacent going, though the engine notes three places from those three runs — so it's not a horse that falls apart on this surface, it just hasn't converted. Class is entirely unavailable for this race at Listowel — the race class is unrecognised, so I have no class-change signal either way. The odds of 11/8 are at the very bottom of the qualifying band; there is slim margin for error at this price. The engine's estimated win probability is 38.9% against a market-implied 37.6%, which gives a modelled edge of just -7.6% — that is a negative edge flag, and I don't dismiss it. The model selected Minella Buoy on composite strength despite the edge reading, and I'm backing that judgement, but the honest position is that the price is tight and the market is not obviously wrong. If Prosperity Angel had a going score available on soft, the gap between first and second on this shortlist could narrow considerably.
Minella Buoy · 11/8 · 1 unit · HIGH
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.
