What is the difference between a speed figure and a rating?

A speed figure is built mainly from finishing time, adjusted for how fast the surface was running that day. A performance rating like Timeform or RPR also folds in weight, distance and the strength of the race, so it rates the horse rather than just the clock.

Are Timeform and Beyer figures on the same scale?

No. Timeform and official ratings are in pounds, while Beyer Speed Figures use their own North American scale. You can compare horses within one system, but you cannot read a Beyer as if it were a Timeform without converting.

Why does the fastest time not always win the rating?

Because track speed varies day to day with ground, weather and rail position. A time that looks quick on a fast surface can be a worse performance than a slower time recorded when the going was holding.

Can you make money just backing the top-rated horse?

Rarely. Published ratings are available to everyone, so they are already in the price. The edge lives in correcting for what the figure missed, such as trouble in running, a wide trip or a track bias, and comparing that to the market.

Horse Racing Ratings and Speed Figures Explained

Timeform, Beyer, RPR, official marks — what horse racing ratings actually measure, why the bare clock lies, and where the real edge begins.

A rating is an attempt to put a single number on a horse's ability, comparable across races run at different tracks, distances and weights. Get past the jargon and that is all any of them are doing: Timeform, Racing Post Ratings, Beyer Speed Figures, official handicap marks. This guide explains what they measure, how they differ, why the bare finishing time misleads, and the part most punters miss: where the number stops helping.

Times Lie. Horses Don't.

The temptation is to read the clock straight: faster time, better horse. It's wrong, and it costs casual punters money every week. A racecourse on Monday and the same course on Wednesday can be seconds apart: rain, wind, temperature, where the rail is set, how the ground was prepared. A horse running 1:10.2 for six furlongs on a fast Monday might have run a 115-rated race; the identical time on a slow Wednesday might be worth 125.

Think about your own running. You jog your usual loop in 22 minutes through deep sand at high tide, then 18 minutes over packed sand at low tide. You didn't get fitter in four hours; the ground changed. A racehorse's clock works the same way. That's why a serious figure rates the horse, not the clock.

What a Rating Actually Does

A performance rating normalises a run so it means the same thing everywhere. Done properly it adjusts for three things at once:

Track speed — the whole race day is read together, so a slow surface isn't mistaken for a slow horse.
Weight carried — a horse that matches a rival's time under more weight has run the better race, and the rating reflects it.
Distance — weight and trip are converted to a common currency before the number is set.

The single most useful property is portability. On a properly built scale, 120 means the same thing at Sha Tin, Ascot and Meydan — which is what lets you compare a Hong Kong handicapper with a British Listed horse without guessing.

The Main Systems

Official ratings (OR). Assigned by the handicapper to set the weights in handicaps. Their job is to make horses meet on level terms, not to predict winners, which is exactly why a horse "well treated" relative to its true ability is where handicap value often lives.

Timeform. A long-established independent rating in pounds, with small-letter annotations flagging temperament, doubts and improvement. Widely respected and subscribed to.

Racing Post Ratings (RPR) and Topspeed. RPR is the Racing Post's all-round performance rating; Topspeed is its time-based figure. Reading the two together — ability versus raw clock — can beat either alone.

Beyer Speed Figures. The dominant North American number, time-based and track-adjusted, popularised by Andrew Beyer. It is the standard currency of US handicapping.

Different methods, same ambition: one consistent number for performance.

How a Projection Rating Is Built: Horses as Measuring Instruments

The clock isn't useless — it's a starting point and a sanity check. But the more robust approach is the projection method: when a field contains horses with established ratings from past runs, those horses become measuring instruments for today's race.

A simple version. Horse A has run 110, 110, 111, 110. Horse B has run 113, 113, 113, 112. Today B beats A by a margin worth about three points. Without consulting the clock at all, B ran roughly 113, A roughly 110, and a newcomer two points behind A earned about 108. The race calibrates itself. When the projection and the stopwatch disagree, the projection tends to be the more reliable of the two.

How to Use Ratings Without Being Used by Them

Recent and relevant first. A horse's last few runs at today's trip, going and class tell you far more than a career-best set two years ago in different conditions.

Read the trend, not just the level. Three figures climbing — 95, 102, 108 — describe an improving horse the market may still be pricing on the old number. Three falling describe the opposite. Direction often matters more than height.

Beware the single big number. One figure far above the rest is sometimes a genuine peak and sometimes a fluke produced by a freak pace or a measuring quirk. A reliable horse with a tight cluster of figures is a different proposition from a one-run wonder with the same top mark.

Where Ratings Stop, and Edge Begins

This is the most important section, and the one the rating companies cannot put in a number.

Published ratings are commoditised. Anyone can subscribe to Timeform or pull up an RPR. By the logic of any efficient market, information available to everyone is already in the price. If your entire method is "back the top-rated horse," you're betting the same information the whole market already has — and paying the takeout to do it.

The durable edge lives one layer down, in what the rating missed. A figure is a summary of finishing time and beaten lengths. It doesn't know the horse met trouble at a critical furlong, raced wide throughout, was conceding weight all round, or ran into a track bias that flattered the winners and buried the closers. So the question a serious student asks is never just "what is this horse rated?" It's "what would this horse be rated if its last run had gone normally?" The crowd prices the printed figure. You price the corrected one. The gap between them, race after race, is the edge.

The Practical Workflow

Start from a reliable, consistent rating as your baseline ability estimate.
Weight toward recent runs at today's conditions.
Read the trend and the consistency, not just the top number.
Adjust for what the figure could not see — trip, pace, bias, trouble.
Compare your corrected estimate to the market price, and bet only when yours is bigger.

That last step is where ratings turn into profit — the subject of our guide to value betting.

See it in practice. MWP shows a normalised performance rating and the model's expected rating on every runner across the jurisdictions we cover. Try a real racecard, or learn the method free in Do Your Homework, chapter 2.