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 biggest market move of the day lands in a Listed race at Ripon, and when that signal sits on top of a horse who has won its last two, the read becomes hard to ignore. Today's scan covered all 51 declared races and 502 runners, run through a four-stage filter pipeline: first stripping abandoned or already-resulted races (no change), then cutting to runners priced between 2.0 and 50.0 decimal — no odds-on, nothing above 49/1 — which removed 107 runners and left 395. Field-size filter then kept only races with 6–14 declared runners, trimming to 301, before a lay-of-day collision check dropped a further 4 runners, leaving a working pool of 297. Feature scoring applied weighted factors in this order of influence: market_move (1.6), speed (1.5), form (1.2), class_change (1.0), trainer_strike (0.8), going (0.8), and draw — which remains unavailable, reserved for v2 of the engine.
The day opened with 51 races and 502 runners. After the odds-band, field-size, and lay-of-day-collision filters were applied in sequence, 297 runners survived to be scored. Draw-bias data was unavailable for every runner in today's pool, so that feature contributed nothing to any composite — a known gap across the board.
The top of the board was tight at the summit. Rogue Jewel came out on 7.18/10, driven by a form string of '11', a confirmed steamer, and an SR that clears the field mean by nearly a full z-score. Caughtinyourtrance was right behind on 7.16/10 — fractionally separated — and actually posted a stronger speed z-score, but a cold trainer (James Moffatt at 5% over 30 days) dragged the composite back. Star Start scored 7.07/10 in the same Ripon amateur handicap as Woodleigh (6.65/10); both steamers, but the class-6 context and Ruth Carr's pedestrian 12% strike rate kept them below the top two. My Boo Boo showed a striking market edge — +106.4% implied-vs-estimated — but the form reading of '29' produced a near-flat score and the going record is blank on good ground. Goal Exceeded at 6.61/10 was penalised by an unrecognised class field, which left that feature unavailable rather than scored. What tipped the balance for Rogue Jewel was the combination of the highest composite, the cleanest form profile, and a trainer who is genuinely hot right now.
The read I keep coming back to with Rogue Jewel is that almost nothing in the feature breakdown is working against it. Form scores +0.87 on a decay-weighted basis — a '11' string at this stage of a two-year-old's campaign, with the most recent win carrying full weight, is about as clean as it gets in the juvenile division. Speed sits at an SR of 101 against a field mean of 90.8, a z-score of +0.87 — it is not dominant, but it is meaningfully clear of the average in a Listed heat, which matters. The going score is +1.00, derived from 2-from-2 on good or adjacent going; there is no lurking uncertainty about ground conditions today.
The market signal is the sharpest part of the case. Edward Bethell's runner is a confirmed steamer — a strict majority of bookmakers shortened, none drifted — which generates the maximum +2.00 on the market_move feature, the highest-weighted factor in the engine at 1.6. Bethell himself is running at 23% over the last 30 days from 15/66 — that is unambiguously hot, and the trainer_strike feature scores +0.89 as a result. The only feature working against the horse is class_change at -0.60: it is rising from rank 3 to rank 2, stepping up into Listed company. That is a real question, and I am not dismissing it.
The model's estimated win probability is 36.5% against a market-implied 22.3%, giving a positive edge of +45.9%. At 3/1 decimal 4.0, that edge is meaningful. The composite of 7.18/10 is a solid score — not the sort of number that demands maximum conviction, but enough to commit to.
The class step is the honest concern. Rising from rank 3 to rank 2 — into Listed company — is not trivial for a juvenile, and the -0.60 class_change score reflects that directly. If the field contains rivals who have already proven themselves at this level, Rogue Jewel's two wins may have come in softer company than today demands. Draw data is unavailable across the board — at Ripon over 6f, the draw can matter, and the engine simply cannot account for it at this stage. The composite gap between Rogue Jewel at 7.18/10 and Caughtinyourtrance at 7.16/10 is marginal — two hundredths of a point — so if anything about the Ripon race reads worse on closer inspection, the Cartmel chaser is breathing down its neck. Finally, the engine conviction is set at MEDIUM, which I am not overriding; this is a solid case, not an overwhelming one.
Rogue Jewel · 3/1 · 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.
