DC Market Intelligence Report –Sep 16, 2026

01 The Market Verdict

Yesterday’s Strong Steamer cohort delivered a clear underperformance versus what the closing market said should happen. We recorded 7 winners from 77 qualifiers (9.1% win rate) against 14.21 expected winners at final implied prices, producing a 49.3 performance index. In plain terms: the horses that were most strongly backed from 10am to the off won at roughly half the rate the final market priced in.

That matters because these are not marginal moves: the average 10am→final contraction was 5.72pp, while the average Open→10am move was only 0.27pp. The information, such as it was, arrived late and decisively. When late, strong contractions fail to convert at this scale, the market is effectively “overpaying” for the signal on the day, which is exactly what the level-stakes history illustrates: theoretical P/L was -31.75 pts at 10am and worsened to -46.87 pts at SP.

Zooming out, the 30-day profile is less severe but still below par: 85.1 performance index (300 wins vs 352.32 expected). The key distinction is that the 30-day record shows very different behaviour by timing: theoretical P/L at 10am is strongly positive (392.95 pts) while SP is heavily negative (-410.02 pts). Yesterday fits that broader theme: late steam into the close has not been paying its own implied probability, even when the direction of travel looks compelling.

02 Market Beaters vs Money That Misfired

03 Where the Money Went

04 The Practical Read

1. Keep yesterday’s message in proportion: it was a weak single-day conversion (performance index 49.3), but it also aligns with the 30-day pattern (85.1) that the closing market has, on average, been too optimistic about the win yield of these late contractions.
2. Respect timing. With Open→10am movement small on average (0.27pp) and 10am→final large (5.72pp), most “information” arrived late. The 30-day split between strong 10am P/L and poor SP P/L is consistent with overcompression into the off: the same horse can be a rational contender and still be a poor late price.
3. Use confirmation rather than magnitude alone. The biggest collapses included both a win (The Good Wife) and a decisive miss (Mandolin Rain). Size of contraction is not sufficient; treat it as a prompt to interrogate whether the move is being chased into an increasingly fragile price.
4. Don’t downgrade every loser: places and close defeats still carry signal, but they do not pay the win probability implied by the final odds. Results like Dream Diamond (3rd) and Pure Granite (4th) are consistent with “in the right area, overpriced to win.” That distinction is exactly what the performance index is capturing.
5. Keep sample size discipline on connections. The perfect day for Jonjo & A.J. O’Neill (3/3) and for W. J. Lee and Jonjo O’Neill Jr. (2/2 each) is noteworthy as a descriptor of where the market money was well-directed yesterday, but it is not, by itself, a durable rule.
6. Allow for the drift-win outcome. With multiple late drifters winning (Ice Jet, Hamlet’s Night, Bad Dreams), yesterday underlined that negative late price moves can still coincide with a win; treat drifts as information about price, not a definitive statement about ability.
7. No filter-supplied qualifiers were available at generation time today, so the Live Market read should lean on process: assess contraction size in percentage points, compare 10am to final/near-off behaviour, and avoid over-weighting one day when the 30-day evidence already warns about late price compression.