Methodology — Edge Card v0
How these numbers are made (v0). Edge Card v0 is a transparent heuristic, not a trained model: we convert Brisnet's published power ratings into within-race probabilities, blend them 50/50 with the de-vigged morning line, and apply a published rules-based pace adjustment computed from early-pace figures. Every constant is on this page, every pick is timestamped before post time, and every result — wins and losses — is archived and scored against closing odds. A trained, walk-forward-validated model (v1) replaces this in August; until then, treat these as structured analysis, not proven edge.
The formula
Per betting interest, within each race: p_pp = softmax(power / tau), p_ml = devig(morning line), p_base = w·p_pp + (1−w)·p_ml, p_v0 = renormalize(p_base · pace_adj). Fair odds = 1 / p_v0.
Published constants
| Blend weight on power rating (w) | 0.5 |
|---|---|
| Softmax temperature (tau) | 10.0 |
| Missing-figure fallback | ml |
| Pace front-runner rule | field_relative |
| Front-runner: early-pace within N of race top | 5.0 |
| Contested pace when ≥ | 3 early types |
| Contested: closer boost / early penalty | ×1.1 / ×0.9 |
| Lone-speed gap / boost | 8.0 / ×1.15 |
| Per-interest pace clamp | [0.85, 1.15] |
| Flag: minimum edge / minimum p | 5 pts / 10% |
| Flag: power-rating rank required | top 2 on Prime Power |
| Flag: minimum odds multiplier | ×1.1 of fair |
| Max flags per card | 4 |
Honesty notes
We report market drift vs. morning line — not CLV. At a mid-morning publish there is no bettable price; the morning line is a handicapper's estimate. Drift asks whether the money moved toward our flag after we published. A change to any constant above is a new model version, disclosed here before it affects a card.
How we grade ourselves
Three surfaces on the record hold the model accountable, and all three run off the public ledger — nothing is hand-picked.
Closing-line scoreboard. In a pari-mutuel pool your payout is set by the final odds no matter when you place the bet, so the close is the only price that matters. For every flagged play we check whether the closing tote odds actually reached the “play at” price we published before post. If they did, the value we flagged was real and takeable; if the horse closed shorter, it was a pass — and a run of them is a sign our number was measured against a generous morning line rather than the real market. This is not a fixed-odds CLV claim; it is the honest test of whether our “value” is value at the window.
Calibration. Across every runner we price — not just the flags — we bin the model's win probability and compare it to how often those runners actually won. When we say 30%, do they win 30%? We publish the reliability curve and our Brier score beside the morning line's, so you can see whether the model beats the board or just matches it. Because every probability is hash-committed before post, this is a live reliability check, not a backtest you have to trust.
Provenance. The ledger is one continuous SHA-256 chain: each card's lock entry carries the hash of the entry before it, so any back-edit, deletion, or reorder breaks the walk. The chain is public and re-verifiable by anyone.
Model versions
The current model is v1.2, effective for cards locked on or after July 17, 2026. Every version change is disclosed here before it affects a card, and the archive is never re-scored — a card keeps the model that locked it.
- v1.2 (July 17, 2026) — flag-quality tightening. A flag now also requires the horse to rank in the top two on our power-rating blend, so a generous morning line alone can no longer produce a flag; the minimum edge rose to 5 points, the morning-line weight came down (blend 50/50), and the “play at” price was re-set to a reachable level. Prompted by live evidence on July 16 that looser flags were bet through our fair price by post.
- v1.1 (July 12, 2026, pre-launch) — field-relative front-runner detection in the pace rules.
Computed from licensed Brisnet data — derived figures only; raw vendor ratings are never republished.