Predictive runner rankings
Each declared runner is assessed through a machine-learning pipeline, producing a consistent hierarchy of contenders rather than a subjective shortlist.
Independent UK & Irish racing analysis
See the race
beneath the market.
Daily horse racing intelligence built from predictive models, runner-level data and a relentless testing process. Clear enough to use. Deep enough to trust.
Reports cover every declared runner—not just a list of tips.
01Model-led Predictive ranking at runner level
02Race-wide Every contender put in context
03Decision-ready Signal, confidence and risk made clear
More than a racecard
A sharper way to
read a race.
Hoofs turns a large field of raw racing variables into a structured view of the contest—who rates strongly, where confidence holds up, and where the race carries hidden risk.
Confidence, separated
Win, place and contender strength are treated as different questions—because a horse can be reliable without being the likeliest winner.
Context beside the numbers
The companion report adds race shape, evidence and runner-level interpretation to help you understand what the model sees.
“Good analysis should make uncertainty visible—not hide it.”
Warnings, weak races and conflicting signals remain part of the picture.
Audited historical evidence
Measured as analysis.
Reported with context.
Hoofs is assessed on genuine out-of-sample races. These figures describe how well the ranking identified winners and official places—they are not profit claims or a betting strategy.
Explore the historical results →51,605Hoofs Ranking races audited2022–2025
26.80%Rank-1 win strikePLACE-led first rank
55.12%Rank-1 place strikeOfficial paid-place result
60.14%Winner in top threeAll audited races
New to the V2 report? Learn what the rankings, medals, confidence ratings and companion context actually mean.
How to read the report →Published daily
One race day.
Two useful lenses.
Start with the model report for rapid comparison, then move into the companion when you need the reasoning and race-by-race context.
V2
Daily Model Analysis
Hoofs Ranking, signal strength, leading contender groups and confidence controls across every race on the card.
- 9 UK & Irish meetings
- 62 races analysed
- Interactive race navigation
C
Runner Metric Companion
Evidence-led interpretation of the key runners, their strongest metrics and the context behind each race.
- Runner-level evidence
- Race shape context
- Designed to read alongside V2
Looking for an earlier race day?
Browse the report archive →Behind the model
Built by an analyst,
not a content machine.
I built Hoofs to answer the questions a standard racecard can’t: how strong is the model’s view, what evidence is driving it, and where should we be cautious?
V2 is the result of continuous model development, historical testing and daily review across UK and Irish racing. The goal is not to make racing look certain. It is to make better-informed decisions possible.
Open project · Since early 2025
Hoofs reports have been published openly since the beginning of 2025—first through the Telegram group, and later through a public GitHub Pages archive. V2 continues that record with a stronger model and a clearer reporting framework.
Explore the original V1 archive ↗Gavin CampbellFounder & racing analyst, Hoofs
The Hoofs standard
- 01Evidence before opinionThe data earns the conclusion.
- 02Transparency over theatreWeak signals stay visible.
- 03Useful at race timeDepth without unnecessary noise.
Join the conversation
Get the reports
when they land.
Join the Hoofs Telegram group for daily publication alerts, model observations and racing discussion around the cards.