AI consensus
Where the models agree
Today's horses scoring in the top tier across more than one of our AI models. Three models score every runner, each weighing the form differently. When they agree, that's a stronger signal than any single score in isolation. Disagreements are interesting too — they're the runners the models read most differently.
How this works
For every runner on today's racecard we have three AI scores — one from each model. Each scores 0–100. A score ≥ 65 is "top tier" (the same threshold used elsewhere on the site).
- Consensus: two or more of the three models put this horse in the top tier (≥ 65) — shown on the card as Majority agreement.
- Strong consensus: all three models agree — shown on the card as Strong agreement.
- Contrarian: spread between the highest and lowest score ≥ 70 points and the top score is ≥ 70 — at least one model rates the horse highly while another does not. Shown on the card as Strong disagreement.
Every card leads with a plain reading of how much the models agree; the grid underneath shows each score behind it. What each model measures →