DC Market Intelligence Report –Sep 7, 2026

01 The Market Verdict

Yesterday’s strong-steamer cohort delivered a small but clear one-day edge versus the market’s own pricing. Five wins from 25 qualifiers converted an expected 4.60 winners into 5 actual, producing a 108.7 performance index. That is not an explosion of predictive power, but it is meaningful evidence that the late contraction (10am to the off) contained information the market did not fully price in by SP.

The more important judgement sits against the 30-day tape. Over the last month, the same “strong steamer” behaviour has materially underperformed what final implied prices said should happen: 296 wins versus 343.26 expected, a 86.2 performance index. The coexistence of a positive day and a negative month matters: it suggests yesterday was a “market got it slightly wrong in the right direction” day inside a broader regime where strong late support has, on average, not been enough to beat the closing line.

The returns profile underlines the same point. Theoretical level-stakes P/L was positive at both reference points yesterday (10am and SP), but the 30-day record shows a very large divergence: strongly positive at 10am and heavily negative at SP. That combination is classic evidence of timing sensitivity: when you measure at the close, the edge has been competed away and then some; when you measure earlier, the prices were materially different. Yesterday’s average 10am→final move of 4.92pp confirms how aggressive the late compression was, so any edge is not simply “being on the right horse”, but being right early enough and still right after the market finishes moving.

02 Market Beaters vs Money That Misfired

03 Where the Money Went

04 The Practical Read

1. Keep the timeframes separate. Yesterday’s PI (108.7) says the cohort beat final-price expectation on the day; the 30-day PI (86.2) says the closing line has generally been a ceiling strong steamers have not cleared. Treat any single-day uplift as noise until the rolling index repairs.
2. Respect the closing market, but don’t over-read percentage plunges at huge odds. Moves like King Casper (101.00 to 41.00) can look dramatic while representing only a small absolute probability shift; use the implied-probability change (pp) as the reality check.
3. Use “how it ran” as part of the evidence. Close losers such as Double Rush (2nd) and placed runners like Atreides and Scenic Route indicate the money often located competitive chances even when the win column doesn’t fully comply.
4. Demand confirmation across moves, not just a single late lurch. A profile of continued support (Open→10am plus 10am→final) is a different quality of information to a late-only snap, and yesterday included both types.
5. Price and sample size discipline matters more than narrative. One-day connection clusters (e.g., William Haggas 2/2; Cieren Fallon 2/3) are notes for monitoring, not signals to follow without further volume.
6. No filter-supplied qualifiers were available at generation time, so today’s Live Market work should be framed around real-time confirmation of price contraction/expansion and the size of the implied-probability move rather than pre-listed names.