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 money has been talking all morning at Perth, and the model is saying the same thing — Ras Kassar is where the edge lives today. The engine scanned 28 races and 269 runners declared for Saturday, applying a filter pipeline designed to cut noise before a single composite score is calculated. Odds-on runners were removed immediately — the band runs 2.0 to 50.0 decimal, evens to 49/1 only. Field size was then trimmed to races carrying 6–14 runners, stripping out both the tiny fields where results are coin-flips and the open handicaps where chaos swamps signal. Feature weights reflect what I trust most in this market: market_move carries the heaviest load at 1.6, speed is next at 1.5, then form at 1.2. Class change, trainer strike rate, and going each carry lighter influence. Draw bias is reserved for a future version of the model and plays no role today.
The day opened with 269 runners across 28 races. The odds-band filter cut that to 219, removing anything odds-on or priced above 50/1. The field-size gate then reduced the pool to 159 runners — the working universe against which every composite was calculated. From those 159, six horses reached the shortlist.
The top six by composite, in order: Ras Kassar at 7.94/10, Scots Poet at 7.39/10, Casheldale Lass at 7.25/10, Atlantic City at 7.19/10, Thickthorn Tom at 7.01/10, and Grand Clermont at 6.87/10. The gap between first and second is 0.55 points — not a landslide, but it is a clear separation rather than a photo finish. What tips the balance for Ras Kassar is the combination of the maximum market-move score and the strongest speed figure relative to its field. Scots Poet actually posts a higher raw z-score on speed and a cleaner form string, but its composite trails because the underlying feature scores convert at lower weighted contributions — and crucially, its going record is paper-thin at 2-from-23. Casheldale Lass carries a spectacular implied-edge number of +231.2%, but that is partly an artefact of opening at a big price; no form string on file is a real gap, and I treat that with caution.
The read I keep coming back to with Ras Kassar is the market-move signal. The engine awards a maximum +2.00 on that feature — a steamer where a strict majority of bookmakers shortened the price and none drifted. At the feature weight of 1.6, that is the single largest contribution to the composite, and it arrived cleanly, without ambiguity in the data. When the money moves that decisively on a 2m handicap chase at Perth on a Saturday morning, I pay attention. The market-implied win probability is 25.5%; the model's estimate is 40.1% — an edge of +40.4%. That is not a rounding error. It is a meaningful divergence, and the direction of market movement suggests at least some informed money shares the view. On top of the market signal, the speed figure is the strongest in the shortlist relative to its specific field: SR 123 against a field mean of 105.2, a z-score of +1.43, weighted up to a +1.43 contribution at the 1.5 speed weight. That is a strong reading, not a marginal one. The form contribution of +0.87 is built on a recent string of '11-101' — the decay-weighted average of +1.25 reflects a horse that has been running into the places with genuine recency. The engine flags three features as unavailable — class change, trainer strike rate, and going — and I will not pretend otherwise. Those are real gaps. But the three features that do score are all pointing the same direction, and the one that matters most to this model fired at maximum.
The honest risk here is that three features are absent. Going is the one that nags at me most: there are no career runs on good or adjacent going on file, which means the model cannot confirm that Ras Kassar handles today's conditions — and Perth in July on good ground is a specific ask. If the ground has dried out further since declaration, that unknown becomes more uncomfortable. Trainer strike rate is below the noise threshold with only one runner in the 30-day window, so there is no quantified stable form to lean on. Scots Poet is the closest challenger at 7.39/10 and it is not far away — if anything shifts in the market between now and the off, I want to know about it. The odds of 5/2 are fair rather than generous; the edge is real but this is not a wide-open price, and a single bad jump on a 2m chase can end the conversation early.
Ras Kassar · 5/2 · 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.
