Anyone who has used or even looked at our Dynamic WebRatings will know that we love data and statistics and in this blog post we are going to discuss the A/E Statistic. If we were looking at just one statistic we can present an additional 5 statistics associated with that statistic. Yep we love it.
So let’s get back to it, what is the A/E statistic? How is it calculated? And what does it tell us?
A/E stands for Actual over Expected.
Now anyone from a statistics background will be familiar with the term Expected but not quite in the same way we use it here. In this process we take the expected chance of winning implied by the odds, so if a horse is 2/1 or 3.0 (depending on your preference) it would have an expected chance of winning a given race of 1/3.0 = 0.33. This figure is then multiplied by 100 to give us the % chance of winning so in this example it is 33%.
So in order to make the A/E value we take the Total Number of wins (Actual) and divide it by the sum of the Expected values for those runs (Expected).
So if a horse went off at 2/1 or 3.0 five times on the trot and won two of them, the math would be: 2/(5*0.33) = 1.2
Therefore the A/E is 1.2 which is better than expected.
The key points to take away here:
1. An A/E value greater than 1.0 means those runners outperformed what we expected of them based on their betting odds.
2. Likewise an A/E value less than 1.0 means those runner underperformed what we expected of them based on their betting odds.
Looking back at that example
So using the above example of Paul Nicholls performance in the last 7 days we see that his horses are being over backed compared to their performance as we have Win and Place A/E values less than 1.0 which is quite interesting considering the high strike rate ;)
Just to give Mr Nicholls fair credit here are his all-time A/E values too:
So in summary an A/E value is an indicator of relative performance, it is a powerful little stat that few punters use and provides an edge over the masses reading too deeply into their formbooks.