The Draw on the All Weather
In this article I am going to share an important part of my latest draw research. This is in preparation for the all weather season that will start once the turf season has drawn to a close, near the beginning of November. It will effectively run from then until the end of March, so close to five full months of action.
Draw research was where it all started for me over 25 years ago. I still update my draw stats on a regular basis, usually at the end of each season, but occasionally I use a shorter time frame, especially if I think I noticed a change in bias. You do need to keep ‘on top of’ draw bias as they can change. Spotting even subtle changes can give you an edge before the bookmakers and other punters catch up.
For this article I am going to split the draw into quarters (quartiles). For those of you interested in this type of ‘draw splitting’ the excellent www.horseracebase.com has the capacity to do this within its system builder.
For this article I am analysing 5 and 6 furlong draw data from UK racing from 2018 to 2023 (up to 16th August, when I started to collate the data for this piece).
When examining the draw and any potential draw bias I always look at handicap races only. This is because handicap races tend to be the most competitive, and in addition to this there are a good number of handicaps year in year out. I believe the number of runners is important too when it comes to the draw, and I am using 8 runners here as my minimum.
5f Handicaps
I want to take a look at 5 furlong handicaps on the all weather courses during this time frame. 5 furlong races are the shortest and as a rule, biases tend to be stronger over 5f than other distances.
I am ignoring Kempton as 5f handicaps are extremely rare these days (only 24 qualifying races since 2018, and just 2 in the last 2 years). Also, I will ignore Southwell as they have changed the surface recently and hence draw data from the old surface should not be considered.
The table below will show the win percentages for each quarter (quartile) of the draw.
As we are splitting into quarters / quartiles the percentages should be around the 25% mark assuming there is no potential draw bias in play. Clearly the further away the percentage is from 25 (above or below that mark), the more likely a draw bias exists.
Quartile 1 stands for the lowest quarter of the draw (e.g., draws 1, 2 and 3 in a 12 runner race), quartile 2 stands for the second lowest quartile (draws 4, 5 and 6 in a 12 runner race, and so on. I have had to make some slight numerical adjustments to the final percentages to allow for the fact that some races do not split exactly into four quartiles:
Going from top to bottom, the strongest 5f bias looks to be at Chelmsford – the lowest quartile (quartile 1) has won 37.8% of the races, whereas the highest quartile (quartile 4) has won just 17.6% of the time. Also, if we split the draw exactly in half, the bottom half of the draw have won 63.9% of races compared with the top half on 36.1%.
Lingfield has even looking figures with all scores between 22 and 28.8% – these stats seem to suggest a pretty level playing field in terms of the draw.
Newcastle is the only one of the four where they race on a straight 5f track, and it seems that the highest draws enjoy an edge here. The highest quartile (quartile 4) has won 35.4% of races, nearly double that of the lowest quartile (quartile 1) which stands at 18.2%.
At Wolverhampton, there seems to be a nagging bias here with lower the draw the better. The bottom half of the draw have won 61.1% of races compared with 38.8% for the top half.
From this initial starting point, it is time to dig a little deeper.
This is a Gold member article, if you are a Gold or Platinum member you can read the whole article here or you can upgrade your account here.