DC Market Intelligence Report –Oct 24, 2026

01 The Market Verdict

Yesterday’s strong-steamer cohort delivered 11 winners from 87 qualifiers, a strike-rate that looks serviceable in isolation but sits clearly below what the closing market itself implied. The key evidence is the gap between actual wins (11) and expected winners at final prices (13.38), which pushes the performance index down to 82.2. In plain terms, even after restricting the sample to ≥15% 10am→final contractions, the late market still overestimated how often these runners would win.

The price-shape matters. The cohort’s average Open→10am move was small (a slight early ease), then the defining move arrived late: an average 4.31pp swing in implied probability from 10am to the off. That is consistent with “information arriving late” behaviour, but yesterday it did not translate into a commensurate edge at the win line. The place rate (37.9%) shows the money was not random; it found competitiveness often enough, just not enough conversion into winners to match implied expectation.

Zooming out, this was not a one-day anomaly. The rolling 30-day profile is strongly aligned with yesterday’s result: 80.6 performance index from 2086 qualifiers indicates a persistent pattern of underperformance versus closing prices. The striking divergence between theoretical P/L at 10am (+255.68 pts) and at SP (-490.66 pts) over 30 days reinforces the same message: whatever value existed earlier in the day has tended to be competed away by late money, and the closing price has not been “soft” enough for the subsequent win-rate.

02 Market Beaters vs Money That Misfired

03 Where the Money Went

04 The Practical Read

1. Anchor expectations to the performance index. With yesterday at 82.2 and the 30-day at 80.6, the evidence says late-steam strength has not, on average, delivered winner rates consistent with the closing price. Treat closing-market confidence as informative about chance, not as a standalone edge.
2. Separate “competitive” from “winner-likely”. The cohort’s place rates (yesterday 37.9%, 30-day 37.1%) suggest the market often finds runners that can get involved. That is still usable intelligence, but it is different from assuming the win price is value after a large contraction.
3. Respect the size and timing of the move, but don’t overinterpret it. Yesterday’s average 10am→final move was a material 4.31pp, yet the group underperformed expectation. Big late moves can be correct about direction and still be wrong about magnitude.
4. Be price-aware: yesterday’s theoretical P/L swung from +56.08 pts at 10am to -2.50 pts at SP. The 30-day swing is even more extreme. The market evidence continues to indicate that “getting on earlier” and “taking the final price” have behaved like two different propositions.
5. Use connection clustering as a watchlist input, not a conclusion. Repeated trainer/jockey support appeared, but one day cannot establish a durable pattern; defer judgement until you see repetition across weeks, not meetings.
6. No filter-supplied qualifiers were available at generation time, so today’s Live Market work should lean on process: confirmation across books, disciplined thresholds, and an explicit view on whether the remaining price still compensates for the 30-day underperformance versus expectation.