Every race day, a horse’s name in the program is followed by a small line in italics — typically sire / dam (damsire). That single line tells you more than most bettors realize. It carries information about where the horse is most likely to run well, how it handles surface changes, and whether the upcoming distance plays to its inherited strengths.
Pedigree alone won’t pick a winner. But it tells you which runners are bred for the race shape in front of them — and that’s a leg up the public usually under-prices.
The horse’s father. The strongest single pedigree signal — sire effects show up in win rates after just 200 starts. We track every active sire’s progeny by surface (dirt / turf / synthetic) and by distance bucket (sprint / mile / route / marathon). When a sire’s progeny win 22% of dirt sprints but only 8% of turf routes, that’s real, persistent signal.
The mother. Less statistically reliable than sire because each dam typically has only 5-15 foals — too small to estimate win rates from. We surface the dam’s name for context but don’t bet off it.
The mother’s father. This is where pedigree gets interesting. The damsire’s influence on stamina and surface aptitude is sneaky-strong — Coolmore-style breeders deliberately cross specific sires with specific damsires to get a known performance signature. “A.P. Indy on the damsire side” traditionally signals route stamina; “Storm Cat as the damsire” signals turf grass-fitness. We track the damsire when the data’s available and surface it as context.
On every race detail page, you’ll see colored tags next to each runner. Each one has a precise definition:
The horse’s sire’s progeny win ≥18% on this race’s surface, across at least 30 starts in the database. The high-end of pedigree fit.
The sire’s progeny win ≤6% on this surface. Not always a fade signal — some sires throw runners that excel even when the average doesn’t — but the surface is working against them.
The sire is materially better on this surface than another (≥8 percentage-point gap). Common with first-time-turf horses by sires like Kitten’s Joy or Galileo.
Sire’s progeny win ≥18% specifically at this race’s distance bucket and surface (≥20 starts). The most precise pedigree signal we publish.
Sire’s progeny win ≤6% at this distance × surface combination. The horse may simply not be bred for the trip.
The sire has logged 50+ progeny starts on turf — meaningful enough turf sample to trust the surface estimate.
We chunk every race into one of four buckets so we can build defensible sample sizes per sire:
This lets us say “Promises Fulfilled’s progeny are 26% in dirt sprints” instead of trying to slice every quarter-furlong (which would make every sample size too small to mean anything).
It does not tell you the horse is fit. A perfectly-bred runner returning from a six-month layoff is a question mark — fitness trumps breeding when the body isn’t ready.
It does not tell you the trainer is sharp. Pedigree is the genetic ceiling; the trainer determines how close the horse gets to it. We separate connection signals from pedigree signals deliberately.
It does not tell you the horse will get a clean trip. Post-position bias, pace, traffic — all of those override breeding on any given afternoon.
We treat pedigree as a baseline expectation. Then we ask: do the day’s connections, pace, and price agree with that expectation, or contradict it? When they agree, conviction is high. When they contradict, the race is harder than it looks on paper.
Open any race detail page (/today → click a race) and scroll to the Pedigree fit section. Each runner shows the sire’s win rate on the day’s surface, distance-bucket fit when sample size allows, damsire when available, and the X-factor tags above.
If you’re reading the day’s picks card, the picks already incorporate pedigree fit into the conviction tier. Pedigree is one of four signals (model edge, connections, price, line drift) used to classify HIGH / MEDIUM / LOW conviction.
Pedigree stats on TBredIQ are descriptive multi-year rollups of settled North-American racing data. Informational only. Past performance is not indicative of future results.