Anyone who has ever attempted to analyse horse racing will know it’s all too easy to get buried beneath a cavalcade of statistical data; and not all of it is useful! Serious punters always look to have an edge on the bookie by creating various systems based on dozens of different racing criteria.
As well as the Timeform Ratings, Racing Post Ratings, Bayer Speed Figures and other commonly sought after figures, there are lesser known and some would say little understood ratings that warrant further investigation. Today, we look at the Racing Post Topspeed ratings which appear as TS on a Racing Post racecard.
What Are Topspeed Ratings?
In simple terms, Topspeed Ratings are figures designed to show insight into the ability, potential and preferred conditions of a horse. As is the case with the Racing Post Ratings (RPR), TS ratings are measured in pounds and attempt to eliminate the impact of different track and weather conditions on race times.
In other words, TS ratings should theoretically enable you to avoid quite a bit of research as they can provide you with analysis of the rating achieved by each runner in conditions relevant to the event it is set to compete in.
How Does The Racing Post Calculate Its Topspeed Ratings?
The official line from the Racing Post is that a mature horse carrying 9 stone with a rating of 100 has the ability to meet Racing Post Standard Time in a race that is ‘truly run’ on ‘good’ going. All distances are proportioned back to 5 furlongs and the ratings are calculated on the basis that 1 second at 5 furlongs = 22 pounds.
For National Hunt races, the scenario changes to a mature horse carrying 12 stone with a rating of 135. In this case, 1 second at 2 miles = 6 pounds. Let’s make this a bit clearer with an example using data taken directly from the Racing Post for a Flat race (I’m making up a name of a horse here):
Gentle Ben is a 3 year-old running in a 7 furlong race at Lingfield’s All-Weather track. His weight is 9st 7 pounds and he finishes second 2.5 lengths behind the winner who clocked a winning time of 1 minute 25.99 seconds (85.99 seconds). The Racing Post Standard time for this particular race is 1m 24.20 (84.20 seconds) so the winner was 1.79 seconds slower than standard time.
In this instance, the RP makes a ‘Going Correction’ of 0.24 seconds per furlong. Using the above information, the Racing Post comes up with a Topspeed rating of 59 for Gentle Ben which is achieved with the following set of data:
- Race Distance (D): 7 furlongs which is 1,540 yards (1 furlong is 220 yards; 7 x 220 = 1540).
- Time of Winner (T): 99 seconds.
- Beaten Distance (B): 5. One length is defined as being 3 yards; as Gentle Ben was defeated by 2.5 lengths, he was beaten by 7.5 yards in total (2.5 x 3 = 7.5).
- Total Distance Covered (D2)[For Gentle Ben]: 5 yards. This is because he was beaten by 7.5 yards so he covered 1532.5 yards in 85.99 seconds.
- Speed (S): 82 seconds. This is calculated by dividing the Distance Covered and Time (1532.5/85.99).
- Time Behind Winner (T2): 0.42 seconds. This is calculated by dividing the Beaten Distance and the Speed (7.5/17.82).
- Comparison to Standard Time [C]: +2.21 seconds. This is calculated by adding the Time Behind Winner to the amount of time the winner was outside the RP Standard time mentioned above (0.42 + 1.79).
- Adjusted Comparison (A): +1.58 seconds. Remember, Topspeed ratings are adjusted to 5 furlongs or 1,100 yards. This means we multiply 1,100 by the Comparison to Standard Time and divide it by the actual Race Distance of 7 furlongs or 1,540 yards so (1,100 x 2.21 = 2,431/1,540 = 1.58 when rounded up).
- Going Correction (G): 24 seconds per furlong as mentioned above.
- G2: +2.78 seconds. This is the Going Correction adjusted for 5 furlongs which means 5 x Going Correction + the Adjusted Comparison (5 x 0.24 = 1.2 + 1.58 = 2.78).
- Pounds Equivalent of G2 (P): 16. Again, remember that in Flat races, distances get proportioned back to 5 furlongs and 1 second at 5f equals 22 pounds. Therefore, this figure is achieved by multiplying 22 by the G2 figure of 2.78 so it is (22 x 2.78 = 61.16).
- Weight Adjustment (W): -7. We simply subtract Gentle Ben’s race weight (9 stone 7) from the standard 9 stone figure used by the Racing Post so (126-133 = -7).
- Rating [R]: The formula used by the Racing Post here is (100-W-P + 13 wfa). Wfa stands for Weight for Allowance and is normally less generous to immature horses than the official scale. The 100 figure is of course the stock rating used by the RP for Topspeed so 100-W-P = (100 + 7 – 61.16 (because of the double negative) = 45.84 + 13 = 58.84; this is then rounded up to 59 which is Gentle Ben’s Topspeed rating.
Formula For Flat Races: 100-W + wfa – 22(1100C/D + 5G)
Formula For National Hunt Races: 135-W – 6(3520C/D + 16G). Note that there is no Weight For Allowance for jump race calculations.
How Useful Are Topspeed Ratings?
Topspeed and speed ratings in general tend to be put to better use on American tracks rather than UK ones. This is because courses in the U.S. often have similar configurations whereas every UK course is unique in its own way. As it turns out, there are a few things to consider:
Distance
In an ideal world, punters would be able to compare the times of races on different tracks but this is very hard when it comes to UK racing. For instance, a ‘5 furlong race’ at Newbury is actually 5f 34 yards (such as the Weatherbys Super Sprint) while at Beverley a 5 furlong race is 5f exactly.
The distance issue gets even more complicated on turf courses if running rails get reconfigured in order to provide fresh turf. If this change happens around a bend, the distance ran by horses can decrease if a running rail is removed or increase if the rail gets brought out at the bends. You won’t see any evidence of these changes in official records so this muddies the waters a bit.
Tactics
In UK sprint races, horses tend to be reined in at the start and are only unleashed for a furious finish in the final couple of furlongs. You would expect the Topspeed rating to be quite different if the horses followed the U.S sprint race model of going all out from the start (and struggling to maintain speed at the finish).
Going Corrections
It seems as if the Going Corrections included in the Topspeed calculations cannot be taken as Gospel. There are plenty of occasions when the Going Corrections published for past meetings are the same for races taking place over round and straight tracks. Clearly, there will be differences in wind speed and ground conditions for example.
Conclusion
Trying to create a system using Topspeed ratings is certainly complex and you need to have a head of numbers. These ratings could prove handy when used in conjunction with other data such as Official Ratings but they should not be relied upon on their own merits as there are too many variables preventing you from achieving an accurate figure.
The potential issues covered above only scratch the surface which suggests that Topspeed ratings, while a useful tool, contain enough flaws to treat them with caution.
Brilliant. One of the best, if not the best explanation I have ever come across.
Ask some one to explain handicapping. Even if they have been betting for 30 years. Not a clue.
You deserve a pint. If your ever in Bangkok, let me know.
Cheers,
Daniel.
I first got into Horse racing when I was at school in the early 60’s. Split Second of The Handicap Book was The Time expert there, his scale Average time=100 at 9stone(flat),12stone(nh). Every year there would be an article
“How Time Ticks”, explaining Going Allowances and lengths=pounds(1lb) 5F-1M 3lbs=1L 2lbs 10F+, 1 Second=
5 lengths all distances. Later Ken Hussey took over and changed the 5-6F to 4lbs per1L. A Housewife used a £2
perm on Ken’s top &some 2nd best on the ITV7 and won over£350k, so Time became more popular with punters.
Going Allowances(GA) from Split second were calculated by taking the 4 fastest times to Average, divide each by the
number of furlongs in the race total divide by 4= GA per furlong.
I found the addresses of US Publications, Morning Telegraph &Daily Racing Form and pestered them to find out “How” they calculated Speed Ratings. All US Tracks are Left handed and flat, unlike Britain& Ireland where they can be LH,RH uphill, downhill, undulating, etc. The US Track Variant was based on 1/5second(0.20)slower than Record Times, this was totalled for all races and divided by number of races=Track Variant.
Nowadays The Going Stick can be used on it’s own or in conjunction with actual times to Estimate GA.
I Compiled my own figures for a few years and there are many ways to do this, example Course Ratings, Horse who equalled standard at Ascot or Newmarket =100 while one doing same at say Catterick would=84.
Going Allowances you can compare the meeting Above or below Standard to previous meeting and corresponding meeting previous year.
American Racing is much more time oriented, American Pharoah made all in The Breeders Cup and averaged 12seconds per furlong. Each Quarter is timed, AP did 23 4/5 for first Q, 23 3/5 2Q. Time was slightly over 2mins for 10furlongs.
best wishes,
Francis
Hi Francis I have always wanted to know how Split Second compiled its ratings and you possibly let me know or how I could find out, I have searched web and asked other racing forums with no joy.
Thank you for this excellent post 🙂
Hello Francis, very nice to hear from someone who remembers Split Second from the Handicap Book. I too remember the article “How Time Ticks” which usually appeared and the start of each new flat season in the Handicap Book and the blue Form Book annual.. I did have a copy of the article for many years and used it as a reference to produce my own time figures when Split Second Retired. I started using Split Seconds time ratings in the 50’s when the Handicap book was book size and cost 9 old pennies.
Unfortunately, perhaps due to my advancing years, my memory may be playing tricks on me but I have a much different derivation for Going Allowance (GA). The formula I use, based on the “How Time Ticks” article at the time, utilises ratings and a Time Factor to make allowances for races run over different distances and hence at a different pace.
Brian
BRIAN,
Going Allowance, was based on the 4 best times at a meeting as compared to Average. Example
5F(slow by 0.60s), 7F(Slow by 1.40), 8F(Slow by 1.60) &10F(Slow by2s)…. 0.60 divided by 5= 0.12 slow per furlong: 1.40 divided by 7=0.20slow PF: 1.60divided by 8=0.20slowPF: & 2s divided by 10=0.20 slow PF. Total these 4 figures and divide by 4 to find GA. Total 0.72 divided by 4= 0.18 per furlong is going allowance. I have used simple figures, so not to cloud or complicate. To find the ratings, again to simplify lets say WINNER of each race weights were as follows…..5F(9st6), 7F(8st8),8F(9stone) &10F(10stone). 5F Race GA 0.90 Race was 0.30Fast, 7F GA 1.26 Race was 0.14 slow, 8F GA 1.44 Race 0.16 slow, 10F GA 1.80 Race slow 0.20.
Ratings 5F 100+6(for weight carried)+4 for GA adjusted time0.30 Fast. Race Rating =110.
7F. 100-6(Weight)-2 for GA adjusted time 0.14 slow. Rating=92.
8F. 100 +0(Weight)-2 for GA adjusted time. Rating=98
10F. 100+14(Weight)-2 for GA adjusted time. Rating=112.
100 was basic rating to equal Standard/Average time carrying 9stone. Each 1lb above 9st. add1pt,Each 1lb below minus 1 point.
Standard Times, were compiled from 6 best times over C&D. An American Speed expert used 3 best C&D times each year for 3 years. If using RECORD Times then Standard times/Averages not used.
Split Second also made some adjustments according to class of Racing, example he would give a better GA for a top class track such as Ascot, and would give consideration to previous figures of up to first four finishers in fastest race, and maybe first two in other races. Basically the GA is an “Educated Guess” at best. I used as a guide, FIRM Minus 0.30PF, G-F Minus 0.20PF. GOOD. NIL.
G-S(YIELDING) Plus0.60PF. SOFT Plus 1secondPF & HEAVY Plus 1.40pf+.
Brian hope this clarifies things and answers Thomas’s questions too.
Merry Christmas!
Francis, my apologies for not replying sooner but I am having problems in both receiving and sending emails.
Thank you for your clarification of how time ticks. An excellent breakdown of a difficult subject.
Best wishes for the New Year, Brian.
BRIAN,
Thank you for your reply, only happy to help.
The BHA have website and it gives, Going stick readings.
There is a lot of things about US speed ratings, one is called RAGOZIN figures.
Very best to you and yours for a Happy New Year,
Francis
Francis, I remember Split Second, and I use Topspeed adjusted figures as a part of a guide in a successsful system that I use. However I am puzzled on a couple of things about his adjusted figures, but no point in continuing in a long winded post if you are not still on the circuit( I see your post is dated 2015. If I get a reply I would like to discuss Top speed with you.
This is a first class explanation – thank you very much.
Just one small error the ‘Time Behind Winner (T2)’ figure should read 0.42 seconds – even some of the horses I back cover 2.5 lengths faster than 42 seconds!
Cheers
Pete H
Good spot Pete, I’ve edited that now.
Hi
Very good article. Its Also nice to Know that there are more old timers who
remembers the handicap book and split second. Unfortunately i never found
any time ratings quite so good so stopped using them as my first port of call
in picking winners, but dont do so bad and still get an adrenalin rush watching the horses
Regards Tony
Hello all,
Reading above about the Handicap Book and Split Second brings back great memories.
Is there any reader on here who can advise on how best to compare the old SS Ratings to today’s Racing Post Topspeed Ratings.
Or
is there other commercial ratings comparable to the old SS on the market?
Kind regards
Ramsey
The best speed figures that I ever found was Dataform which disclosed unadulterated figures.
This has now become Raceform Interactive but it is too expensive for my taste and it does not disclose the full history of all the horses runs. Sometimes the beasts do run back into form after a year or so out of form and at a good price!
I am forced now to use the Racing Post as a member to get the TS figures.
I base my betting on TS figures which I reconvert in my head.
The Raceform Update is much missed in this household, especially the wonderful systems page.
Thank you for an excellent article.
few years later but a very clear explanation thanks. One further question. When they publish the TS rating in the Racing Post for future races have they adjusted a previous rating for the likely going, distance and weight carried as they would when determining the rating after the race
As far as I am aware, yes.
I just had to google my way into this article because I couldn’t understand why some favorites have very low TS figures like 9 and beat horses with 100+ rating. As I understand it, the TS rating is largely worthless, except in sprints.
Personally, I don’t use it.
I remember years ago when the Sporting Life was still going, the Speed Ratings for the Sporting Life were as far as i can remember compiled by Topspeed ratings. When the Sporting Life became defunct and the Racing Post was the sole racing paper, the Sporting life Weekender continued to produce and still does Topspeed Ratings. They were compiled by Dave Edwards, but i do not know who compiles them now.
Thank you for the info Andrew 🙂
Why do you not use it.?. What’s your reason
I prefer our own ratings which are less widely used and have a bigger edge.
i am a south african who is trying to use topspeed ratings in a south african racing context.i have the problem of converting furlongs to metres.somebody please help.
Hey there, 1 furlong is 201.168 meters.
Hi michael just for a matter of interest do you use raceform interactive and also when compiling your speed rating how do you work out the All weather surfaces against turf or do you keep a complete set of different ratings for the A/W, i find myself that because in the winter i do not get involved with jumps racing but i stick with A/W though the winter and with newmarket more than likely getting an A/W track as well there is reason to believe these types of tracks offer a more stable and reliable edge over turf tracks. The reason i mention this is because we do now seem to find the weather in this country is not very reliable and with the A/W champions day at lingfield and the offer of good prize money on the A/W A great number of bigger owners and stables are willing to keep good horses in training over the winter campaign and i do now find myself winning more money on the A/W then on the turf even though i do win a lot on the turf i personally stick with horse that run between 7f and 1m 4f this gives me a great advantage and how do you bring a draw bias into your ratings because surly you would have to allow a degree of tolerance if the draw has an affect on any given race and also jockeyship as to affect a horses performance whats your insight on this.
We create our own standard times and then base everything off those. Adjustments are then made for different race types when comparing across race types, however we also have speed ratings for ground conditions so that you can see best performance over the same ground as the current race. I have always found AW racing to be the least variable of racing in the UK and therefore the easiest to profit from. I no longer use draw bias as I’ve found the edge it yields to already be included in the market, and it’s far easier to use the market to adjust my figures before betting. Generally I don’t use jockey or trainer information, although I know I am rare here, the reason being is that I find it easier to assess the horse than the humans.
hi michael i have read your article about speed ratings for ground conditions, iwould be happy if you will share that info. with me.
i find race advisor very interesting and educating since RP has change a lot of the info. that uses to be free
A vintage Ken Hussey fan here. I will resist writing at great length on the subject of Speed Figures but if it were available I would be a candidate for a PhD on the topic. There appears to be one vital understanding missing from this thread. A key value of SF analysis is that it confirms (or queries) the value of a rating derived from orthodox linear or non time based assessments. In essence a form assessment that is based on a truly run race (as confirmed by a good time figure) is reliable and the rating can carry confidence that a figure derived from a slowly run race cannot. From this give special attention to performances that attract a good S/F with and accompanying race comment that a winner performed with some ease (mainly in NH races). This barely scratches the surface of my understanding of the practical application of S\Fs. Hope this makes sense – happy to entertain any queries but not on decimal point calculations etc. Sophisticated judgement is what is required – not a slide rule.
Let’s be serious time is irrelevant as each race has one important factor ommited the weight of the horse 1200 Pounds this changes every race every distance whatever the going 2 year olds 3 year olds. A jockeys weight 126 lbs or 101 lbs which is 1/10 the horses weight
Absolutely, weight of horse is important as well. Unfortunately the horse’s weight isn’t published in the UK. In our speed ratings we use a standard estimate (as you mentioned in your comment) and use that along with an average jockey weight to adjust our speed ratings on, I couldn’t say whether the Racing Post do this on TopSpeed ratings or not.
Hi Micheal, Once becoming a member, is it possible to have access to how you work out your own ratings?
Hey Conor, we share elements of how they’re calculated for some ratings, but not the actual formulas.
So the topspeed rating is calculated with reference to 9 stones so why is it -7 and not +7 in the example?
(oh yeah in the formula two minuses make a plus) … The lengths beaten attracts a variable value, dependent on the raceDistance (5F, 6F etc… is that right?) But the rating is eventually adjust to 10 stone on the racecard, is that right? The big question is there any software available to auto-generate TS ratings. Please :O)
i do my own handicapping using just times and have developed my own set of speed figures but i am only interested in five and six furlong races where the early pace is most likely to run at a true fast pace and then only on good or faster going on turf and the AW racing (my favourite surfaces), if anyone wants to develop their own figures all i can say is do not base your work around Median Times as these are flawed and then when the going allowance is worked out at the end of the days racing the way the figures are compiled is flawed again, and to be honest if using figures that have been worked out with a going allowance for good to soft ground at one course isnt of any use at any other course as it is way too much of a generic term
Great message, thank you Eric. Would you be interested in sharing some more insight into how you go about calculating your ratings in our forum at https://www.raceadvisor.co.uk/community/, I’d love to hear what you’re doing.