You can read a chaser’s last six runs and still not know how fast it has ever gone over fences. There is hurdle form, a bumper win as a four-year-old, a point-to-point that flattered it. The number that matters most is the fastest figure the horse has ever recorded on the kind of race it is running today, and it is easy to overlook.
Race Advisor’s version of that number is called SHorBES. The name is a mouthful, but the idea is clean: take the best speed figure a horse has ever produced on the current race type, then use it to rank the field. A horse with a top SHorBES has shown that it can run quickly under conditions like today’s. The question this piece answers, using a large dataset, is where that information is most useful.
Does the SHorBES speed rating find winners? Yes, it sorts runners in a clean order, and it is most useful in jumps racing on turf. In an earlier Race Advisor analysis of more than 1.18 million UK runners from July 2013 to March 2026, the top SHorBES horse won 18.1% of the time with an A/E of 1.06. A/E means actual winners divided by the winners the odds implied, so anything above 1.00 says the group won more often than its prices suggested. In turf chases that A/E rose to 1.22.
What does the SHorBES rating measure?
SHorBES stands for Speed Horse Best Ever Same-race-type. In plain English, it is the highest speed figure the horse has ever recorded on the kind of race it is running today. A horse running over hurdles is judged by the fastest hurdle figure on its record. A chaser is judged by its best ever chase figure. The same horse carries a different SHorBES over hurdles on a Tuesday than over fences on a Saturday, because the rating is filtered to the race type.
The logic is simple. The rating shows the fastest comparable performance already on a horse’s record, which is a useful marker of what it has shown it can do, although a horse can always improve on it. The rank tells you whose best is highest in the field. The raw figure tells you how high that best is in absolute terms. A rank 1 horse with a SHorBES of 145 is a different proposition from a rank 1 horse with a SHorBES of 60, even though both top their fields.
There are three ways to read it:
- The rank, which sorts the field today.
- The raw value, which shows how strong the signal is.
- The gap to the next horse, which shows how clearly the rating separates the front of the field from the back.
Reading them together is the whole point.
How does SHorBES rank perform across UK runners?
The first test of any rating is whether the rank sorts horses in the right order. In the earlier analysis, 1,188,886 UK runners between July 2013 and March 2026 were settled at Betfair SP, which is Betfair Starting Price, the exchange’s closing price, after 2% commission.
The rank 1 horse won 18.1% of the time across 163,097 runners. Rank 2 won 15.2% across 157,743. Rank 3 won 13.2%, rank 4 won 11.9%, and everything from rank 5 downwards won 8.4% across 578,248 runners. Each step down wins a little less than the one above, which is exactly what a sensible rating should produce.
The A/E for rank 1 was 1.06. Race Advisor treats 1.05 as the minimum bar for a factor worth following on its own, so the top rank clears it. It is a modest edge on its own, and it tells you the market has not fully absorbed this rating. The more rewarding picture comes when you look at where the rating works best.
For context, our latest database check, across Great Britain and Ireland from 1 July 2013 to 14 October 2026, puts the favourite at 34.4% of 167,257 races, with an A/E of 1.00 at Betfair SP, level stakes and 2% commission. The favourite wins more often than SHorBES rank 1, as you would expect from the shortest price in the race. The interesting case is the races where SHorBES rank 1 is not the favourite, which was 74% of races in the earlier analysis. There the strike rate was 11.4% with an A/E of 1.06, at bigger prices. Our guide to A/E ratios explains how to read that measure in full.
Why does the raw SHorBES number add so much?
Rank tells you who is top today. The raw figure tells you whether that top signal is loud or quiet. Splitting every runner in the dataset by absolute SHorBES value, whatever the rank, makes the point quickly.
- Horses with a SHorBES of 125 or higher won 12.6% of the time across 767,913 runners, with an A/E of 1.06.
- Ratings between 75 and 124 sat at an A/E of 1.00 across 254,732 runners.
- Ratings between 0 and 74 sat at 0.99 across 133,041 runners.
- The few runners with a negative SHorBES, 32,737 in all, sit well below the others, so you can set them aside quickly.
The practical lesson is that rank alone can miss something. Rank 1 horses with a SHorBES of 125 or more returned an A/E of 1.07 across 135,535 runners. Rank 1 horses with a SHorBES below 75 returned 0.98 across 6,237 runners. A top rank in a collectively slow field is not the same as a top rank in a strong one, and the raw figure tells you which you are looking at.
Where does the SHorBES edge really sit?
The headline A/E of 1.06 is the average across every kind of UK racing. Splitting the rank 1 sample by race type changes the picture in a helpful way.
Turf chases are where this rating earns its keep. Rank 1 in turf chases won 20.7% of the time across 22,675 runners, at an A/E of 1.22. Turf hurdles follow at an A/E of 1.09 across 36,750 runners. Combined as jumps turf, SHorBES rank 1 returned an A/E of 1.14 across 59,425 runners.
Flat racing is a quieter picture. Rank 1 on Flat turf returned an A/E of 1.02 across 60,073 runners, and Flat all-weather 0.99 across 37,361. On the Flat, prices have followed this rating closely in the results. National Hunt racing, with longer fields and more varied prices, is where the rating has shown more. In National Hunt bumpers (flat races for jumpers-in-the-making), rank 1 returned an A/E of 1.00 across 5,613 runners, so bumpers are a sensible place to lean on other clues.
The going, which is the state of the ground, sharpens the picture. Within turf hurdles, SHorBES rank 1 on good ground returned an A/E of 1.13 across 13,333 runners, good to soft 1.08 across 5,954, heavy 1.12 across 4,118 and soft 1.06. It is not simply a case of the softer the better, so check the going for the slice you are looking at.
Class helps too. SHorBES rank 1 in Class 3 races returned an A/E of 1.13 across 14,474 runners, with Classes 1, 2 and 4 between 1.06 and 1.07. The rating is a little less sharp at Class 5 (1.03) and Class 6 (1.01). The middle of the class ladder, where form is reliable enough to project forward, is where the rating does its best work.
How do you use SHorBES with confidence?
A good test for any rating is whether it still shows its value when the market disagrees. SHorBES rank 1 in turf chases passes it. Where the rank 1 horse was also the Betfair SP favourite, the strike rate was 39.5% with an A/E of 1.21 across 6,022 runners. Where it was not the favourite, the strike rate was 13.9% with an A/E of 1.22 across 16,653 runners. The same edge shows up at bigger prices, which is a sign that the rating carries real information.
The other good test is whether the pattern holds over time. Splitting the jumps turf sample in half, 2013 to 2019 returned an A/E of 1.13 across 30,244 runners and 2020 to 2026 returned 1.15 across 29,181. That consistency across two non-overlapping windows is what a durable signal looks like.
Four habits make the rating easy to use:
- Use SHorBES rank as a shortlist tool for jumps turf races, especially turf chases.
- Read the raw figure next to the rank, so you know whether the signal is strong.
- Check the going, particularly on hurdles.
- Check the A/E for the slice you are looking at before you rely on the headline number.
Race Advisor’s weight-adjusted speed rating covers the same family of speed ratings from a different angle, and the two together give a fuller read than either alone. Our single-rating profit test shows the same pattern across the form ratings: one figure is a good starting point, and the race conditions complete the picture. If you want to see how a second rating combines with this one, our guide to combining ratings walks through it.
What is the takeaway?
SHorBES does something real. It sorts UK runners into a clean strike-rate order from rank 1 down to rank 5 and below, with an A/E above 1.05 on the top rank across more than 163,000 runners in the earlier analysis. It gets its best results when you give it the right setting: top SHorBES rank, in jumps turf racing, ideally turf chases, with a raw figure that confirms the signal is strong. In that setting the earlier analysis found an A/E of 1.22 across more than 22,000 turf chase runners, and the pattern held across two separate six-year windows. These A/E figures describe how prices compared with results across a very large number of runners. They do not describe the quiet spells you would meet along the way or how any staking plan would have fared, so treat the rating as a way to shortlist and keep your own records. No rating can promise winners, and this is a sensible way to point your attention at the right races.
To see how SHorBES ranks sort this afternoon’s runners, open today’s race cards and compare the SHorBES ranks and raw figures with the prices for yourself.
