DC Market Intelligence Report –Sep 29, 2026

01 The Market Verdict

Yesterday’s STRONG STEAMER cohort finished marginally below what the closing market priced in. The group returned 17 winners from 17.96 expected at final implied prices, a 94.7 performance index. That is not a collapse in signal; it is the market landing close to par while slightly overrating the most aggressively-backed runners on the day.

The more revealing evidence is the split between early and late pricing. The theoretical level-stakes profile shows a sharp divergence: +20.41 pts at 10am prices versus -18.65 pts at SP. With an average 10am→final contraction of 5.62pp across qualifiers, yesterday again displayed the familiar pattern: when you consume the move, you consume most of the edge. The market did not “miss” the information; it largely priced it in by the off.

Against the rolling 30-day context, yesterday is best read as a relatively normal day inside a weaker month. Over 30 days, the cohort shows 310 wins from 390.19 expected (performance index 79.4), with +384.35 pts at 10am but -475.25 pts at SP. One day’s near-par index does not repair that broader underperformance; it does, however, reinforce the key behavioural point: the market is consistently harsher on late prices than it is generous at 10am, and the gap is persistent enough to matter more than any single-day win count.

02 Market Beaters vs Money That Misfired

03 Where the Money Went

04 The Practical Read

1. Keep the benchmark honest: yesterday’s performance index (94.7) is only slightly below expectation, but the 30-day index (79.4) is materially below. Treat “one day near par” as noise unless it persists; the wider sample still says the cohort has underperformed closing expectation.
2. Separate “finding contenders” from “overpaying for them”. The place rate yesterday (44.6%) alongside a near-expected win count suggests the market often identifies live runners, but the win probability implied by heavily-compressed prices can be too optimistic.
3. Respect price compression as the critical variable. The P/L split (+20.41 pts at 10am vs -18.65 pts at SP) mirrors the 30-day pattern (+384.35 vs -475.25). When reviewing live moves, focus less on “did it shorten?” and more on “how much probability is being purchased by the move?”.
4. Watch for race-level crowding. The Roscommon 04:30 cluster shows how the market can over-allocate confidence within a single race. Multiple strong steamers in the same contest can be a red flag for price efficiency rather than a green light for certainty.
5. Use the drift tape as context, not an automatic veto. Yesterday produced winners that drifted hard (notably Quintana (GB) and Gore Point (GB)) and close losers that drifted. Late weakness is information about confidence and conditions, not definitive about ability.
6. No filter-supplied qualifiers were available at generation time, so today’s Live Market work should be grounded in process rather than a pre-built list: confirmation via sustained contraction, sanity-check via implied probability, and discipline about how much of the move has already gone.