Follow @FocusRatings | ![]() |
I have, for some time now....
Been trying to work out why we do better in some races and...
Less well in others?
I've been doing all sorts of complex calculations but...
This morning I realised that...
All along...
I'd been missing the bleeding obvious!
Now, looking at the raw data, I suddenly realised that you could define a race by the quantity of Good horses running in it compared to the quantity of Less Good horses.
As an example...
Let's take an 8 horse race where it's pretty obvious that the winner is going to be one of four of those horses.
I know that because those four horses are the ones that I've found easier to rate;
They have form,
They have had success (going wins, class places and other stuff like that.)
Because of this, the 8 horse race now becomes a 4 horse race.
And it's far easier to pick a winner from a 4 horse race than from an 8 horse race.
Well, from today...
I shall be showing an indicator of which of these races are considered good..
And which are considered less good.
Eventually I shall be adding this as a column (and perhaps making the name of the Top Rated horse bold) and the data will be included in the Builder.
But, for the moment...
This will have to do.
For today, this will only show up on the normal PDF (number 1 on the email.)
Good race is either 1 or 0.
If it is 1, then it is considered to be a Good race.
And, you know...
I may well have toshed up the code, so...
It might not appear at all!
If that happens, I'll send out an email listing all Good races.
There's confidence for you, isn't it?
Don't you just love statistics?
Sometimes I do and sometimes (when I have too much of them), I don't!
But here goes...
Overall...
Since the 8th of July 2013 there have been 12390 rated races.
Of these, in 763, the Top Rated horse was a non-runner.
This leaves 11627 rated races where the Top Rated horse actually ran.
The Top Rated horse won 3013 of those races.
That gives a Strike Rate of 25.91%
Now, according to my calculations...
Bad Races...
7873 of these races (taking non-runners into account) were bad races.
The Top Rated horse won 1840 of those races.
That gives a Strike Rate of 23.37%
Less than the overall average.
Good Races...
3754 of all races (taking non-runners into account) were good races.
The Top Rated horse won 1173 of those races.
That gives a Strike Rate of 31.25%
Substantially more than the than the overall average.
Average SP
The average SP for the Bad races is 3.59
This gives an ROI of 83.88%
The average SP for the Good races is 2.768
This gives an ROI of 86.51%
The average SP for the ALL races is 3.269
This gives an ROI of 84.73%
So, a drop in average SP for the Good races but...
And this is the important bit...
An increase in ROI for the Good races compared to all races or the Bad races.
Now...
By using this filter, you will be cutting down the number of races quite severely.
Since the 8th of July 2013 only 32.28% of all races (where the Top Rated horse wasn't a non-runner) were considered to be Good Races.
And...
Please bear in mind that this calculation only applies to the Top Rated horse.
However, it may be something to consider.
After all, increasing the strike rate (of the Top Rated horse) from an overall average of 25.91% to 31.25%
Might be something that tickles your fancy?
However, I'd suggest that you just watch this new metric until you are happy with it?
Or, perhaps, play around with it in the Builder once I have added it there?
Please feel free to leave comments here (or email me privately) should you have anything to say.
As always...
My kindest regards
You must be logged in to post a comment.
Sorry but I can not see the good race 1 or 0 may be I could do with new specs. Kind Regards Glyn.
Hi Glyn
The Good Race/Bad race is only on the first PDF at the moment – it’s next to the Race Distance.
All the best
Keith
Found it. Thank you.