Open any race card and you will find numbers. Official Rating, Racing Post Rating, Topspeed, plus Race Advisor’s own PR and VDW ratings (our two in-house measures of form and class) and a dozen tipster columns, each offering the same thing in slightly different handwriting. It is tempting to treat them as verdicts: the biggest number is the answer, the smallest is the throwaway.
Ratings are more useful than that, and also more interesting. They are estimates, often very good ones. Once you see what a rating is for, you can use it to narrow a big field quickly and then judge the price, which makes a card much easier to read.
What do horse racing ratings actually tell you? A rating estimates a horse’s relative ability so you can compare runners on a single number. In an earlier Race Advisor analysis of more than 1.5 million UK runners since 2013, the top-ranked horse by our PR rating won 32.3% of the time, while the fifth-ranked won 7.8%. The ranking is a reliable guide to who is likely to win. The price already reflects that, so the real skill lies in comparing the rating with the odds.
Why is every rating an estimate?
A rating squeezes a lot of information into one number: preferred ground, course suitability, race type, distance, the quality of opposition and more. That compression is what makes ratings easy to read. It is also why they sit best as a guide rather than a final answer. Nobody can know in advance how a race will be run or how the ground will ride by the third race.
A sensible habit is to allow a rating some give. As an illustration, a speed rating of 100 might be better read as a band of 90 to 110 until you have a reason to narrow it. The width of that band is a rule of thumb, not a measured figure. If you built the rating yourself, you can be more precise. If you did not, leave the band wide.
How do the main ratings perform?
Race Advisor’s database covers more than 1.5 million UK runners between July 2013 and March 2026. That is enough to look at each commonly used rating, rank by rank. Prices are Betfair SP, the exchange’s closing price. The A/E ratio is actual winners divided by the winners the odds implied, so 1.00 means a group wins exactly as often as its prices said it should.
- PR rating. The top-ranked horse won 32.3% of the time and the second-ranked 19.6%, easing to 7.8% by fifth. The A/E sat between 0.98 and 1.02 at every rank from one to ten. The rating sorts the field well, and the market agrees with it.
- VDW. For Race Advisor’s VDW rating, our latest database check (Great Britain and Ireland, 1 July 2013 to 14 October 2026, Betfair SP, level stakes, 2% commission) shows Rank 1 winning 21.6% of 153,920 races, with an A/E of 0.99. Another method, the same shape.
- Official Rating. In handicaps the OR sets the weight, so a good handicapper makes the race close. Across more than 950,000 UK handicap runners, the top-OR horse won 14.3% and the fifth-top 10.8%, a much flatter line, with the A/E at 0.99 to 1.00 at every rank. That is the handicap system working as designed.
- Dynamic Speed Rating. The top-ranked won 17.8% and the fifth-ranked 10.6%, with the A/E close to 1.00 throughout.
Four ratings, four methods, one pattern: a higher rank means a higher win rate, and the price is already in tune with it. That is useful to know, because it tells you where ratings help most. They narrow a big field to the few runners worth your time.
The methods differ, and the differences are a feature. The Official Rating is the BHA handicapper’s assessment, set to decide weight. The PR rating weighs recent form in the context of class and distance. VDW emphasises class and ability comparisons. Speed ratings focus on raw times. When they line up, you have a consensus pick. When they part company, the race has an open question, and that is often where the most interesting decisions are made.
What happens when two ratings agree?
Finding a horse that is top of two ratings feels like a green light, and in one sense it is. In the earlier analysis, when the same horse was PR Rank 1 and VDW Rank 1 (68,300 runners), the win rate rose to 35.6%, a clear lift on PR Rank 1 alone at 32.3%. When PR, VDW and Dynamic Speed all agreed (16,239 runners), the strike rate climbed to 41.0%. That is a strong filter for finding likely winners.
The A/E stayed at 0.99 each time. That makes sense: the market can see those ratings too, so agreement is priced in. You are identifying winners more often, and the odds move to match.
The practical point is that a rating consensus is good evidence about probability, and the price answers a separate question. Keeping the two apart is a helpful habit.
How do you use ratings well?
Treat ratings as ordering tools. They tell you, with reasonable accuracy, who is most likely to win. Then you ask the second question: is the price fair, generous or short?
That changes how you start. Instead of “who is the top-rated horse?”, ask “what does this horse’s rating profile look like next to its price?” A horse that is PR Rank 1 and VDW Rank 1 at 5/2 is priced about right. The same horse at 7/1 is a prompt to find out why the price differs from the rating picture, for example whether one poor recent run has been given too much weight. Two habits make this easier:
- Hold ratings as ranges. A PR of 90 is really “around 85 to 95, leaning to 90”. If two horses sit within each other’s range, the rating cannot separate them, so another factor must.
- Look at how the rating was earned. A 95 built on three consistent runs at today’s class and trip is a different story from a 95 built on one fluky win and two flat efforts. Same number, different picture. Our guide to what RPR means is a useful read on this, and the Official Rating works on the same principle.
Context matters as well. A rank-one horse wins at different rates in different types of race, so the same number can mean different things from one card to the next. For the bigger picture, see how combining several ratings works, and our guide to value explains how to compare a horse’s chance with its price.
What is the takeaway?
Ratings are excellent at sorting horses by chance of winning. For the four ratings we looked at over thirteen years, each rank group won close to as often as its Betfair SP prices implied. They give you a structured starting point, cut the work of comparing ten or fifteen runners, and flag the horses worth thinking about.
Use them as guides. Compare them. Trust a consensus when it appears, and let the price tell you how much it is worth. No rating can promise a profit, and used this way they help you spend your time on the right horses.
To see how Race Advisor’s ratings rank the runners today, open today’s race cards and compare the rating picture with the prices before you decide.
