**Speed Ratings.**

Basically you need to work out how fast your horse is likely to travel carrying a set amount of weight over a set distance. In the case of horse racing underfoot conditions the jockey and the mood and fitness of the individual horse on the day are however variables. However for the purpose of the basics the following is the science part and although the equation used below is not perfect it is one we may all possibly remember from school and is a reasonable starting point for the purpose of this example.

Velocity = mass x acceleration (measured in feet per second)

To obtain an average or constant as a comparison we need to relate the science to the racehorse by using the following figures which obviously need to be converted into feet or seconds to make the equation viable.

An average racehorse is ten foot long

A two mile race is approximately 10560 feet in length

An average two mile race will be run in 4 minutes which is 240 seconds

Therefore 10560 feet divided by 240 seconds equals 44 foot per second divided by 10 foot means a horse running over two miles in distance will be travelling at approximately 4.4 lengths per second.

Now we have an average we need to work in some figures for the going conditions which will be as per the chart below which is applicable for National Hunt racing. These are not industry standard figures and are my own work so feel free to use your own.

Good Ground = 100%

Good To Soft Ground = 103.5%

Soft Ground = 107%

Soft to Heavy = 111%

Heavy Ground = 115%

Now we will attempt to put this altogether and include an allowance for the weight carried on the day using genuine form from a horse called Artifice Sivola which has run over a variety of conditions.

04/03/16 @ Newbury 2m on Good To Soft carrying a weight of 11-12.

The race was run in 4 minutes 17.80 seconds (the time the winner crossed the line) which is 257.8 seconds. Artifice Sivola finished the race 6 lengths behind the winner so we add on at 4.4 lengths a second approximately 1.2 seconds giving you a finishing time of 259 seconds for our runner.

Now divide our runners time of 259 seconds by the distance of the race which is 2m (but officially 16.5 furlongs) which leaves you with a finishing speed of 15.70 seconds a furlong. The ground conditions that day were Good To Soft so you now divide the 15.70 by 103.5% as per the above list and you achieve a figure of 15.17 seconds per furlong. You now have to make an allowance for the weight carried which was 11-12 which is 166 pound. Divide the 15.17 by the 166 and then times by 100 (to give you a workable figure) which in this case is 9.14.

Confused yet.. you should be because I still am as there are lots of ways to do your own figures but if you get a calculation that works for you it can be a money spinner when used in the right circumstances as although speed ratings are primarily used on the flat they are a cracking reference point over shorter trips especially in hurdle races and of course when used in NHF (National Hunt Flat) races.

Very interesting.

I have used my own speed ratings for a number of years, albeit I calculate them using a slightly different method, based on the teachings of Nick Mordin, plus some useful tweaks provided by US Speed Guru, Andy Beyer.

The main difference between the above method and my calculations is that I do not take weight carried into consideration, which may seem strange, but in my view unless you know the weight of the horse, which is significantly greater than the weight of the jockey, plus tack and additional lead weights, the weight calculation process seems rather redundant.

In addition, scientific studies have shown that weight carried by a horse makes little difference to the overall speed that it can run and really only impacts the speed of acceleration and even then only marginally.

I must admit that I have never used speed ratings for National Hunt races, but have learnt over the years that my method is best suited to assessing performance on dirt at Meydan and Jebel Ali, plus the synthetic ‘All Weather’ Fibresand surface at Southwell, which is the closest thing to dirt that is used in the UK.

I will gladly share my thoughts and speed rating based selections for the above tracks once this site is fully operational.

All the best.

Rob.

Thanks Rob

Maybe we could trial a race using both methods on a race before Meydan comes around. ? It might be interesting to also compare both versions to the Racing Posts ratings etc.

As you know I am not a flat racing pundit but in theory they should be just as effective if not better on the flat than the jumps.

Regards Mike

Hi Mike,

More than happy to compare speed rating for the runners in a race or two.

The comparison will have to be in a race at Southwell between 5 and 8 furlongs in distance, as these are the only U.K. speed ratings that I now calculate.

Likewise, I also keep up to date draw and pace bias information for these distances on Fibresand, which are used in conjunction with my speed ratings to identify possible selections.

Not sure when the next meeting is on the Fibresand at Southwell, but will monitor and advise when known.

All the best.

Rob.

Hi Rob

No problem it will be an interesting experiment for me as I have not tried these calculations out on the AW. Unfortunately the next flat fixture at Southwell will not be until November so this idea will have to go on the back burner for now. But I will hopefully be launching the Tips and competitions pages this week so I can always trial a few general AW tips involving the ratings to see how they pan out.

Mike

Hi Rob

Just as an after thought my Speed Rating Calculations are done on a home made excel program how do you do yours. ? Is it a pen and paper job or have you got to the stage where you can more or less note a fast time just by looking up the race times. ? I only ask as one direction I would like to head down eventually is to link my tissue odds program and my Speed Rating programs together and make both or either available for any interested members to download and play around with.

Hi Mike,

I currently use pen and paper, as the calculation is quite complex due to the fact that I use a number of look up tables to provide the following data:

Class Pars for races based on distance and class of runners.

Standard Times for each distance at each track.

Standard deductions for distance beaten by the winner for each race distance.

Far too complicated for an IT illiterate like me and I also like to work the figures manually, as there is often a need to place anomalous figures into context, such as low rating due to a massive early pace battle between a number of front runners.

However, I think I have mentioned before that I have a young relative who passed GCSE Computer Science with top grade who now needs a project as part of the A Level course who reckons that the whole process can be automated, including maintenance of the separate tables that are kept for identifying both draw and pace biases.

Will initially require some manual input, but we will look to see if we can find a method of dragging the data from an online results service at some stage and if all goes well we will also consider automating the collation of data for assessing future races.

This would save a lot of time and effort, but I am concerned that I would still need to intervene to deal with the anomalies.

Good luck with your plans.

Rob.

If I were you Rob I would buy this young relative a extra x-mas present as although I still do most of my work on pen and paper when it comes to complex issues like speed ratings a friendly computer literate guru is essential (as I have found out to my cost on this project…lol).

The only thing I would suggest is to check that the standard times information you are using for the tracks is up to date as when I first started the information I used was way out of date somewhat blindsided me and I had to do my own in the end as I could not even obtain standard times for the Irish NH courses.Granted I was working on NH times and there is probably far more information available for the flat tracks however it might be worth checking. ?

Also just as a final thought I know yourself and Tommy have more than a passing interest in the racing at Meydan have you tried your ratings out there yet. ?

Hi Mike,

Don’t worry I am not exploiting child slave labour, as I pay weekly ‘pocket money’ based on school performance, plus bonuses when exams are passed, whilst this year’s Christmas present will include a significant upgrade to her home PC, as the current tech has been outgrown.

As for the standard times for the 3 tracks where I compile speed ratings are regularly reviewed to ensure they remain accurate and are tweaked if necessary.

Finally, the speed ratings have been fully tested and proven profitable to follow when linked to draw and pace analysis for some years now, so I am fairly confident that they will remain so for the coming winter seasons.

All the best.

Rob.

Hi

As we approach the end of the first year of the site I have now started to dabble further into some of the subject matter and will add another thought into this section.

If a horse runs over say two and a half miles in a hurdle race at Roscommon he would most probably record a faster speed rating than running in the same circumstances at Killarney as there is a massive twenty one second difference between the standard times at the two tracks.

Is it therefore worth building this factor into the equation. ??

Hello Mike,

I have just found this article and found it very interesting indeed as I am a speed man. I have fiddled around with all kinds of stuff like using the usual 3.5lbs per length for 5 furlongs, 3lbs for 6 furlongs etc. However I ditched the weight calculations as I was getting some stupidly low figures where it becomes impossible to compare the ratings with an Official Rating which is what I have always tried to do without success. Topspeed ratings seem about 14lbs lower than the OR and calculate evrything to 5 furlong using 22lbs per second. Timeform calculate to 5 furlongs but use 25lbs per furlong! Trying to include weight this way is a little convoluted I think. So. Using your calculations for the lengths per second on the flat at different courses it does seem that using an average 5 lengths per second is a fairly accurate figure. But then using 100 as a base rating doesn’t take accountof the class of the race. Rather than trying to convolute class ratings I use the RPR (Racing Post Rating) as my “Base ” figure for that race. It is an ability rating based on form. The calculation is very simple. Using the RP going allowance and 5 lengths per second I have a raw speed figure and by adding or subtracting the figure from the RPR I get a figure that can be directly compared to the Official Rating . This figure is a Form and Time rating. I dont calculate any weight and use one point for one length for beaten lengths which is probably not accurate but if I am using 5 lengths per second and the official lengths used for for all photo finishes is 5 lengths per second I see no reason not to use it. I only calculate non handicap races as most of these are level weight affairs and small penalties dont make that much difference. I have managed to put together a table in Excel where I copy and “paste special” the time analysis page from the RP. Click theMacro anditcompletes everything leaving th ratings. By using macros for the trim (Control H) and DATA< Text To Columns I can delete the "s" "f" "m" "/" "+" leaving the clean figures for calculating. I only need the Furlongs, Going Allowance and the time above or below standard. This is multipled by 5 and deducted or added to the RPR. It is a relatively simple method but helps in seeing horses that have a good rating on standard form reading and speed. Invincible Army at Doncaster first day of the flat in the Gammidge Trophy ran 1.63 slow. (Topspeed rating is 87.) After going allowance the figure was 1.15slow. Multiply by 5=5.75. The RPR was 116 so less 5.75 meant my rating for Invincible Army is 110. Its OR is 112. Other methods wont produce such a figure as far as I am aware. Anyway I hope this might be of interest and if anyone wants help with the Excel code I can email. There are two smart 2yold's that I have rated at the moment :THREAT (87) R Hannon and CHASING DREAMS (85) Godolphin. Pretty high figures for 2yolds at this time of year. Thank you for your article Mike.

Thanks for the interest Brian and the offer of assistance it is very much appreciated.

I am unfortunately a bit tied up with the football side of the site this weekend but will reply in due course (hopefully within the next few days).

Regards MIke

Hi Brian

Apologies for the late reply but the last day of the football season was always going to be hectic. Obviously the biggest difference between my speed ratings and anyone else’s are that mine are based around NH horses.

The second difference would be that I target handicaps but this is because unlike most people I am always looking for a long shot that can be backed using the optional markets available such as betting without the favourites etc.

However having just noted the obvious differences, we are all (and I include Rob) singing in the same choir just not from the same hymn sheet. I have mentioned to Rob previously on this thread that it might be an idea to all have a punt at a certain set of runners or type of race so that a comparison could be made between out individual findings. ?

As Rob and yourself are using your ratings on the flat I obviously would have to adjust my ratings to join in but if the idea should appeal to either of you let me know.

The bonus for me Brian would be that you obviously have a better handle on excel than most of us and this may be very useful in any future coalition work.

Obviously I do not know if you are already tied into other site’s that might restrict your options Rob is currently writing on a weekly basis for MBR racing which may restrict his input.

Regards Mike

Hi All

I think I may have found some common ground with regard to speed ratings which would enable us all to get together on a project without taking up to much time and effort.

How about trying flat races but only at or over the one mile six trip in class 3 and above this would enable me to use some of my two mile hurdlers who sometimes drop back for a spin on the flat and would lessen the effect of the draw. I think I could cope with the transition here as my lenghts per second etc would be unnafected and obviously anyone who dabble in the occult (flat racing) 🙂 should be able to join in. ?

These races are not exactly as rare as hens teeth but from memory only occur occasionally and making a list of these runners may have the advantage of being a good guide for those who go on to the hurdling scene later in life. I will try a race out and post my findings on the site to see how the idea pans out feel fre to join in at any time.

Hi Brian,

Thanks,

can you please email the excel

12keepgoing@gmail.com