kin m

Racing Model Edge Feed One row per runner, per race: a model's own win probability alongside the current market price and the de-vigged market edge (bookmaker margin stripped out via Shin's method). Computed numbers only — this is a data feed, not a tip sheet, and it never claims a win rate.

What you get For every runner on every race card currently tracked, for the date you ask for:

Field Meaning track, race_number, race_name, date Which race horse_name, tab_number Which runner win_probability The model's own win probability (0–1) decimal_odds Current fixed market price, if one is live right now market_prob_devigged Market-implied win probability with the bookmaker's margin removed — only set once every runner in the race has a recorded price edge_devigged win_probability minus market_prob_devigged — the model's edge over the de-vigged market tote_odds Current pari-mutuel pool price, if one is live right now code Thoroughbred or Harness as_of UTC timestamp this snapshot was computed at Any field that has no live price yet comes back null rather than a guessed or stale value.

Input Field Required Description date No YYYY-MM-DD. Defaults to today. Pricing Pay-per-result: $1.40 per 1,000 rows, billed on exactly what's delivered — a quiet day costs next to nothing, a full day of wide coverage costs proportionally more. No subscription, no minimum.

What this is not Not betting advice. Not a recommendation to back anything. Not a claimed win rate or guaranteed return — win_probability and edge_devigged are the model's own computed numbers, nothing more. Not affiliated with any racing body, bookmaker, or tote operator. If you're using this to inform real wagering decisions, that's your own judgment to make — gamble responsibly, and only with money you can afford to lose.