|Welcome Guest ( Log In | Register )|
Articles >> horse-racing >>
Draw Bias Revisited
Draw Bias Revisited
This article goes back to my roots – draw bias. With the flat season soon to be in full swing, I have decided to update my draw figures and I will share with you some of my findings. For the recent part of this research I have chosen the period from 1st January 2008 to 20th July 2010 and I have only considered handicap races with 9 runners or more and have split the runners into “thirds”. Hence in a 12-runner race, draws 1 to 4 would lie in the bottom “third”, draws 5 to 8 in the middle “third”, and draws 9 to 12 in the top “third”. On a completely fair course the winning percentages for each third should be around 33% and, while some courses hover around that figure, others clearly do not. For comparison purposes I have also looked at the period from 2005 to 2007 to see what changes have taken place (if any). Let us have a look in more detail at some of my findings:
Thirsk 5f - I have chosen the minimum distance of 5 furlongs at Thirsk where the draw bias has been traditionally at its strongest. High draws have dominated for many years, but the most recent figures see a change, and a big one at that. There have been 26 qualifying races over 5 furlongs during this period with the draw split as follows:
Amazing correlation with the win and placed stats! However, where has the high draw bias gone? If we look at 2005 to 2007 we can see how things have changed:
As the ’05 to ’07 figures indicate, there was a very strong bias to horses drawn high during this period. The ground nearest the stands’ rail (high) has traditionally been the quickest and hence horses drawn closest to the rail had a significant edge.
There have been 8 qualifying races this year with the following results in terms of winner and their draw position/number of runners: 6 of 12; 2 of 12; 2 of 11; 4 of 11; 2 of 11; 8 of 13; 5 of 10 and 16 of 16. Only 8 races, but you still would not expect high draws to be doing much better.
In 2008 and 2009 the top “third” of the draw secured 44% of wins so even that was a bit of a drop. Hence, I am not sure whether this season especially is simply a statistical blip, but nevertheless I cannot be as confident of a high draw bias and I will be monitoring results for the rest of the season very carefully. I have not noticed any rail movements that other courses have used as a draw bias elimination method; maybe they have been watering the ground near the stands’ rail more than the rest of the straight course.
Catterick 5f - There have been 24 qualifying races over 5 furlongs during this period with the draw split as follows:
There is better news here with Catterick’s low draw bias still looking very strong. Indeed the low draw bias seems to have strengthened over the past 3 seasons. Here are the ’05 to ’07 figures:
Going back to the more recent period of ’08 to ’10 we can see that the lowest two draws have provided 9 of 24 winners. Indeed backing both stalls “blind” would have produced a combined profit of £14.13 (ROI +29.4%). For those interested in forecast returns, if you had placed reverse forecasts on the two lowest draws in all 24 races you would have secured a healthy profit of £200.48 (ROI +417.7%). Good returns if you can get them!
Hamilton 5f - There have been 20 qualifying races over 5 furlongs from ’08 to ‘10 with the draw split as follows:
Now draw experts will know that historically Hamilton has almost always favoured higher draws, especially in the 1990s and the first part of the new century. Even from ’05 to ’07 the top third of the draw secured 45% of wins, but even then the writing was already on the wall. Now we seem to have a reverse bias with low draws favoured, and it looks like a bias worth ‘playing’. Indeed if you had permed the lowest two drawn horses in a £1 reverse forecast in the Scottish Stewards Cup on 16th July this year, you would have received £546.39 back; the exacta return was even better at £776.50. If you had permed the lowest 4 drawn horses in 24 £1 tricasts you would have received a whopping return of £9664.97.
The reason for the change has almost certainly been down to course maintenance. In the old days, even with small fields and the stalls low, virtually all horses tacked across to the far side where the quicker ground was. Nowadays most runners stick to this near side (low) and essentially it looks like the ground near to the rail is quicker than the centre and the far side.
Chester 5f - There have been 15 qualifying races over 5 furlongs from ’08 to ‘10 with the draw split as follows:
Chester is renowned as the course with arguably the strongest draw bias in the country and the 5f figures continue to back up the theory. The previous 3 seasons also had a strong low draw bias:
A potential problem with Chester is that punters and bookies alike know how strong the bias is and hence most people think that profiting from it is difficult. Having said that, if you had backed the 4 lowest drawn horses in all 15 races since the start of 2008 you would have remarkably made a profit – indeed each of the 4 individual stalls has been profitable as the following table shows:
So perhaps there is still mileage in backing well drawn horses at Chester. For the record horses drawn 10 or higher in 5f Chester handicaps since 1997 are a dismal 1 win from 230 runners! Indeed only 11 other runners have placed.
Finally let us look at a bias at a longer distance:
Pontefract 1m - There have been 25 qualifying races over 1 mile from ’08 to ‘10 with the draw split as follows:
The mile trip at Pontefract sees a significant edge to lower draws with very high draws winning just twice in the last 25 races. Indeed draws 10 or higher have produced just 1 win from 105 runners in this time frame. Low draws have enjoyed an edge in the past over this trip but as the ’05 to ’07 stats show, it was not as strong:
The edge was still significant, but the recent bias seems stronger. I am not sure why this is the case, but it is clear that when betting over this C&D we should be ideally looking for a lower drawn horse than a higher drawn one.
The good news therefore for draw punters is that biases still exist, but perhaps more importantly we need to keep track of biases that are weakening or changing. This way we will stay ahead of the crowd and gain more value.
Copyright echo date("Y");?> PunterProfits.com