The mechanics of Speed Figures [cog 2]

[cog 2] Class Pars and Going allowances.

Pars.

As discussed in the previous blog, we now know how to calculate a RAW speed figure dependent on the time taken for the winner to win a given race.

However, this figure is not the one we award the horse. Before we assign a figure for the race we need to calculate how big a part the racing surface played in the time of the race.

The way we do this is to ‘estimate’ how fast each race should have been run. This is similar in a way to assigning a ‘Par’ to a Golf course, the par being the amount of shots it should take to complete the 9 or 18 holes.

It is fairly easy to assume that Group Class horses, i.e. the best horses in the World, will run a lot faster than horses competing for a Class 7 event at Wolverhampton on a Friday night.

But how much faster?

In his excellent book, Mordin On Time (available via Aesculus Press) Nick Mordin details how over the years he has calculated that a race for older (3year old +) Listed Class animals will be run approximately 3.6 seconds slower than the standard time for every mile of the race.

For example: the previous blog states that the standard time for 1 mile (1760yds) at Kempton is 95 seconds.

A listed class race therefore is estimated to be take 98.6 seconds to complete.
(95 seconds + 3.6 seconds for every mile) = 98.6 seconds.

A similar calculation is performed for a race over 2 miles. The standard time for 2 miles at Kempton is 201.2 seconds. Therefore the estimate for a listed class race is (201.2 + (3.6 x 2 miles)) 208.4 seconds.

Using the formula detailed previously to work out a RAW speed figure awards the above two performances a rating of 112 and 114 respectively.

Nick also states that races confined to 2 year olds, 3 year olds or Filly/Mare only races will be run slower than the 3.6 seconds per mile.

The following table details the estimated seconds per mile slower each race should be run for all class of races in the UK.

Note: The prediction for a class 1 race over 6 furlongs is worked out as follows:

6 furlongs = 0.75 of a mile. Therefore 3.6*0.75 = 2.7.
A Class 1 race over 6 furlongs is estimated at 2.7 seconds above standard.

A RAW speed calculation much like the previous section can be performed using the above table to indicate a figure for any race.

Performing the calculation using the standard time for every racecourse in the UK, gives a PAR time figure for every distance at any course for any type of race.

These ensuing figures averaged out, result in the following PAR figures.

Following a few years of using the figures I estimate that Group Class races (above Class 1 listed) are run a few points quicker for each Group Class, i.e. Group 3s are generally quicker than Class 1, Group 2s quicker than Group 3 etc.

Therefore, an updated PAR comparison table looks like the one below:

Going Allowances.

We can now estimate using the table above what Speed Figure each race run in the UK should achieve based on the horses that are running in it.

For example, yesterday (Saturday 28th Nov 2009) the 2.15 “Jacksbridge Handicap” at Lingfield Park was a Class 2, 10f race for older horses (3 YO+).

According to the table above, the Par figure for a race of this nature is 101. For the winner to achieve a figure of 101, the race should be won in 127.3 seconds.

However, the winner ‘Bound By Honour’ won the race in a time of 125.27 seconds, earning a RAW speed figure of 118, some 17 points above the Par time.

This +17 points may indicate that Bound By Honour is a Group 3 horse in a Class 2 race, or it may be due to the racing surface speeding up the horse enabling a quicker time to be attained.

The way in which this argument is settled is to compare the estimated PAR times and the achieved RAW times for the rest of the days results at the same course, as follows.

The above table shows that the surface at Lingfield yesterday was very fast and speeding up horses by an average of 36 points each race.
The going allowance is therefore said to be +36.

The original RAW figure for Bound By Honours performance was 118, however this is now adjusted by subtracting the Going Allowance of 36, resulting in final speed figure of 82, some 19 points below the Par of 101 - hardly the group 3-class performance it looked originally. In actual fact the PAR table shows this performance to be more akin to an average class 5 event.

Obviously other important factors such as pace etc may have played a part in the fairly poor time, (Pace will be explored in another blog) but from my calculations there was no real pace bias yesterday at Lingfield, all of the races were won by horses up with the pace throughout. I would be wary of the future worth of the form of this race.

It can be seen from the comparison table that the best performance of the day was in Race No.2 - a Class 5 event for 2 year olds over 6 furlongs.
This race was awarded a speed figure of 83, some 13 points above the PAR of 70. The winner ‘Tenacestream’ performed well on his debut to earn a figure that would put him around Class 3 level in his age group.

He could be worth watching when turned out again, as he should improve for his first run.

This concludes the basic mechanics behind the calculation of speed figures.

The calculations shown can be performed for every race worldwide, as long as you have an accurate set of standard times and Par Times from which to work with.

There are other aspects that could influence the figures; such as the pace of a race or changes in weather conditions part way through a meeting. These variables may lead to minor adjustments, and will be explored in later blogs.