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 sharpest signal today is not from the form book alone — it is the market screaming at us from two directions at once, and one horse answers both calls. Today's scan covered all 38 declared races and 412 runners. The filter pipeline ran in four stages: odds band (2.0–50.0 decimal, evens to 49/1, no odds-on) trimmed to 355; field size (6–14 runners) brought that to 257; lay-of-day collision protection dropped a further two, leaving 255 runners to score. Feature weights prioritise market movement most heavily at 1.6, then speed at 1.5, form at 1.2, with class change, trainer strike rate, and going trailing behind. The composite is the weighted sum across those dimensions.
The day opened with 412 runners across 38 races. Odds filtering removed short-priced favourites and long-shot chaos, cutting to 355. Field-size filtering — keeping races of 6–14 runners where the model has most traction — reduced that to 257. Two runners were then dropped because the lay-of-day engine is opposing them, leaving a clean working universe of 255 runners to rank.
Five horses pushed clear of the field. Javelot topped at 7.63/10, driven by a maximum market-move score, a speed figure well clear of its field, and a solid recent form return. Sea La Roma came second at 7.39/10 — the model edge there is enormous (+62.4%) but form and class-change data are both unavailable, which is a meaningful hole in a novice. Martymill scored 7.15/10 with the strongest raw speed z-score on the shortlist, but a cold trainer (6% strike rate) dragged it back. Francillon and Toastmaster both landed at 7.00/10 — Francillon's going record (0-from-2 on good) is a quiet drag, and Toastmaster's SR of 87 against a mean of 68.4 is solid but narrower than Javelot's gap. The tipping point for Javelot was the convergence of maximum market-move, a usable form string, and Marco Botti running hot.
The read I keep coming back to with Javelot is the market-move score: +2.00 — the ceiling, meaning a strict majority of bookmakers shortened and none drifted. At 11/8 (decimal 2.375) that kind of market compression on a nursery handicapper tells you someone knows something, and that sits at the top of the feature hierarchy with a weight of 1.6. The speed figure backs it up: SR 77 against a field mean of 65.7, a z-score of +1.77, weighted up to a +0.72 composite contribution. In a Class 5 nursery at Yarmouth on soft, that kind of relative speed edge matters.
The form feature adds another layer. The string reads '41' — the latest run was a win, and the decay-weighted average comes in at +1.26, scored up to a +0.62 composite contribution. It is a short string, which limits certainty, but the direction of travel is right. Marco Botti's trainer strike rate over the last 30 days is 21% (8 from 38), which the model flags as hot and contributes +0.75 to the composite. That combination — steamering price, pace edge, a winner last time, a trainer in form — is what the 7.63/10 composite is expressing.
The one genuine structural gap is going. Javelot has no recorded career runs on soft or adjacent going, so the going feature returns unavailable rather than a score. That is a blank, not a negative — but it means we are not getting a tailwind from that feature either. The engine's conviction is coded MEDIUM, and the market-implied win probability of 37.6% is almost dead level with the model's estimate of 38.6% (edge -8.3%). That is not an edge bet in the price-hunting sense — the model is backing the market's read, not fading it.
The going blank is the headline risk. Javelot arrives at Yarmouth on soft ground with zero career evidence on that surface or anything adjacent. If the horse dislikes cut underfoot the model has no data to flag it, so we are flying partially blind on conditions. The form string '41' is only two runs deep — one win followed by a fourth — which is a thin evidence base for a nursery handicapper, and the decay weighting cannot do much with two data points. Sea La Roma runs in the same venue's earlier race and a strong performance there from the Haggas yard could shift market sentiment on the card; that is context rather than a direct collision, but worth noting. Finally, the composite lead over Sea La Roma and Martymill is real (7.63 vs 7.39 and 7.15) but not commanding — this is not a runaway score-off, and the MEDIUM conviction label from the engine is the honest read.
Javelot · 11/8 · 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.
