DC Market Intelligence Report –Sep 6, 2026

01 The Market Verdict

Yesterday’s STRONG STEAMERS cohort underperformed what the final prices said it “should” have done. The group delivered 11 wins from 15.52 expected winners, producing a 70.9 performance index. That is a meaningful miss versus market expectation on the day, and it matters because these are not marginal movers: qualification required at least a 15% contraction from 10am to the off.

The shape of the day explains the negative SP outcome. Even though the final-price market identified a sizeable set of runners worth shortening (111 qualifiers), the strike-rate lagged the implied probability, and the theoretical level-stakes figure at SP finished at -32.79 pts. The place rate (37.8%) held up far better than the win rate (9.9%), which is consistent with money often landing on the right horses but not consistently on the winner.

Zooming out, the one-day miss fits the broader 30-day picture rather than contradicting it. Over the last 30 days, strong steamers show a 84.8 performance index (299 wins vs 352.48 expected): still below expectation at final prices, and paired with a very poor theoretical SP P/L (-463.76 pts) despite a strongly positive figure at 10am (+273.55 pts). That divergence is the key forensic signal: the edge, when it appears, has been earlier rather than at SP, and late contraction alone has not been translating into winners at the rate the closing market implies.

02 Market Beaters vs Money That Misfired

03 Where the Money Went

04 The Practical Read

1. Respect the closing market, but don’t overpay for it. The 30-day split is the forensic takeaway: strong steamers show a positive theoretical return at 10am but a large negative at SP. Yesterday replicated the “expensive late” problem with -32.79 pts at SP, despite the market clearly identifying many competitive runners (42 places).
2. Prioritise confirmation over magnitude. Huge collapses such as The Love Machine (9.50 → 3.25) can still fail to win; magnitude alone is not sufficient. Use the move as a prompt to demand additional confirmation (field context, whether the price stabilises, and whether the late move is echoed rather than one-way).
3. Think in probabilities, not stories. Yesterday’s 70.9 performance index means the cohort won around 29% fewer races than the final odds implied. A day like that can happen without invalidating the signal; the question is whether it persists. The 30-day index of 84.8 suggests the under-delivery has been structural over a meaningful sample, even though a single day can be noisier.
4. Use pp movement as a diagnostic, not a target. The average Open→10am move was modest (-0.73pp), but the average 10am→final move was large (3.88pp), indicating the day’s information arrived late. That timing has not been converting to SP returns in the recent window, so treat late moves as “market agreement” rather than automatic value.
5. Be disciplined about sample size on connections. Repeated trainer/jockey support is worth logging (e.g., Bethell and Murphy converting well yesterday), but one day is not enough to label a yard or rider as “in” or “out”. Watch whether the same names keep attracting contraction and whether results regress toward expectation.
6. No filter-supplied qualifiers were available at generation time, so today’s Live Market work should start from price action and confirmation rather than a pre-built shortlist.