Speed Ratings – 2

Numbers are fine...

But sometimes it's better just to look at graphs.

So, having done the analysis of the effect of Handicap/non-Handicap on the All Weather Average Seconds per Furlong (details of which can been seen at Speed Ratings - 1),

I'm now about to look at how the Average Speeds vary according to the month of the year.

Now, I've been advised that AW racing should be faster in the winter as the tracks are more compacted; this makes sense to me.

However, of the five AW tracks (guess that it'll be six from now on), only one follows this profile.

Oh dear, I feel lots of calculations coming up!

Month of the Year...

Average Seconds per Furlong for each Month.

Please click on the image to see it full size. It should open in a new tab.

Rather than display lists of numbers, I think that it is probably better to just show you a chart.

The Chart shows Average Seconds per Furlong on the vertical axis and Month of the Year on the horizontal axis.

Now, Lingfield (the line in Yellow) follows the expected profile.

The yellow line goes up (in other words, the races are slower) from June to September.

But, for the other tracks, there doesn't seem to be any rhyme or reason to the peaks and troughs.

Kempton (the red line) seems to be the most consistent track all year round.

Although why Wolverhampton (the purple line) is so much slower than the other tracks is a bit of a question that I can't answer.

Perhaps I've got this totally wrong but I thought that Wolves had the same Surface/Surface Treatment as Kempton.

Conclusion...

Anyway, without labouring the point...

If all the tracks had profiles like Kempton ( a pretty straight line) then I could just work out some formulae to adjust the Standard Times according to the Month of the Year.

Or, if all the tracks had a similar profile as Lingfield then I could do a Winter/Summer adjustment.

However (and this is where charts really do tell a story better than lots of numbers), it's pretty obvious that I'm going to have to include Month of the Year as an individual metric for the Standard Times.

As I mentioned elsewhere (probably in a Morning News email), once I've got a working set of Standard Times for the All Weather tracks, then I can take the code and apply that to the Flat Races.

The next thing that I'm going to look at is the effect of Class on the Average Seconds per Furlong on the AW.

My assumption (and I'm aware that assumptions are the mother of all cock ups) is the the higher the class, the faster the race.

But that's just an assumption and I need to crunch the numbers to prove it, one way or another.

I'm going to continue publishing these "partial" analysis posts (each detailing one aspect of my research) as, if I put all of the data into one big post, it might be a bit overwhelming.

In any case, it sort of helps me to break the analysis up into chunks.

I suppose that the one thought that is at the front of my mind is that...

You can't take anything for granted - everything has to be analysed.

As always, your thoughts and comments are welcome.

My kindest regards