Speed Ratings – 1

You know...

There have been a few times, over the last few weeks, that I have considered giving up on getting any meaningful Speed Ratings compiled.

Indeed, some of you have emailed me to suggest that I do just that.

However, I'm not going to be beaten and I will get the Speed Ratings done.

Even if they don't end up being the Holy Grail, the questions that they ask (and, hopefully, the answers to those questions that we come up with), may help the "thoughtful" bettors.

I now need to do some analysis on the core data in order to help me decide what data to include in creating Standard Times for each Going for each distance at the All Weather tracks.

And, I'm going to start by looking at whether the race is a Handicap Race or not makes any difference.

Handicaps

I have started off by creating a list of Standard Times for each of the AW courses regardless of the distance run.

Now, I know that a shorter race will be run faster than a longer one (apart, interestingly enough, from 5 furlong races which run slower than 6 furlong races.)

However, all AW races are run over between 5 and 16 furlongs, I feel that there is value in generating an overall Standard Time (regardless of distance), at least,for this part of the analysis.

Speeds per Fulong by Course and Going and Handicap Status

Course	Going	HC	SPF
Dundalk	Standard	No	12.52264176
Dundalk	Standard	Yes	12.55314535
Kempton	Standard	No	12.45238409
Kempton	Standard	Yes	12.47294985
Lingfield	Standard	No	12.28364415
Lingfield	Standard	Yes	12.32947623
Southwell	Standard	No	12.91937008
Southwell	Standard	Yes	13.00062423
Wolverhampton	Standard	No	13.64968878
Wolverhampton	Standard	Yes	13.65724468
Wolverhampton	Standard to Slow	No	13.94923434
Wolverhampton	Standard to Slow	Yes	13.78264903

SPD stands for Seconds per Furlong.

We can see that, in general, handicap races run slower than non-handicaps.

The exception is where the Going is Standard to Slow at Wolves. I'll look into this in a minute.

Difference in Speeds per Fulong by Course and Going and Handicap Status

Course	Going	% Difference
Dundalk	Standard	0.24% slower
Kempton	Standard	0.17% slower
Lingfield	Standard	0.37% slower
Southwell	Standard	0.63% slower
Wolverhampton	Standard	0.06% slower
Wolverhampton	Standard to Slow	1.19% faster

The table shows the difference between that Average Speed per Furlong for Handicap races compared to Non-Handicap races.

Now, I know that there is a blip with the racing over Standard to Slow going at Wolves.

But, is it statistically significant?

Well, those races only made up 63 out of the 3370 races used for the initial data sample.

In other words, 1.87%.

I need to test over a larger data sample but, like all things, it's really not that simple as...

When you test the same data over different distances, you see that the Handicap races are faster over the shorter distances (apart from 5 furlong sprints) but slower over the longer ones.

Difference in Speeds per Fulong by Distance over Standard Going

Distance (furlongs)	% Difference
5	2.14% slower
6	0.34% faster
7	0.10% faster
8	0.51% slower
9	0.39% faster
10	1.72% faster
11	1.19% faster
12	0.34% slower
13	4.38% slower
15	3.76% slower
16	3.06% slower

The table shows the difference between that Average Speed per Furlong for Handicap races compared to Non-Handicap races.

Now, I understand that on the 5 furlong races it might be the draw that matters more than anything else.

And I also understand that handicap races are more competitive (and run by horses with more experience) and would, therefore, expect them to run faster.

However, I am at a total and complete loss to understand why the longer handicap races run slower than the non-handicaps over the same distance.

Conclusion

There is enough variance in the Average Speed per Furlong between the handicap and non-handicap races (over different distances) that I feel that I have to use this factor in determining the Standard Times for the different courses (over different distances and goings.)

Next, I have to take a look at the time of the year. Generally, the races run faster in the Winter than in the summer. I need to test this to see if it is statistically significant.

And, if so, I'll have to see if I can just group the races into Summer or Winter races or do I have to split them up by month.

To help me, I'll move my main analysis spreadsheet from the programming machine to my main machine so that I can add graphs to the data.

As always, your thoughts and comments are welcome.

My kindest regards