Matching & intelligence
Deterministic. Transparent. Yours to question.
JockeyX recommendations come from a published scoring model — not an opaque AI. AI may later offer a second opinion, never the source of truth.
Step 1 · Gates
Availability and weight
A rider committed elsewhere at post time, or who cannot make the assigned weight, is never recommended — no matter how strong their record.
Step 2 · Weighted factors
Scored 0–100, factor by factor
Each factor is normalized and multiplied by a configurable weight. The top contributors become the plain-English "why this match".
Step 3 · Human decision
Trainers choose. Agents commit.
Scores inform; people decide. Outcomes feed back into statistics, improving the network over time.
Default factor weights
Track experience14%
Trainer / jockey history14%
Horse / jockey history12%
Surface record10%
Distance fit6%
Race class level8%
Recent form14%
Overall performance10%
Travel / circuit6%
Trainer preference6%
Put your engagement book on the network.
Free for jockeys and agents. Trainers get a 14-day trial.