IS RACE CLASS WORTH INCLUDING IN RACE ANALYSIS?
By Peter May
Anthropologists propose that Class is an obsession of the British and nowhere is this truer than in horseracing and not only in the stands. References to the class movements of runners appear in a significant proportion of race analyses and can often be used as supportive evidence for a selection. It does seem to be a critical factor for many bettors and I have lost count of the number of times I have been asked how I account for the class of a race when determining speed figures. My answer is always the same: I don't. This always surprises the questioner whose view tends to take the normal construct that horses running in better races should be given higher ratings. But this does not apply to speed ratings, for obvious reasons. For instance, if two winners produce the same race time under identical conditions they should both receive the same speed figure even if one was competing in a Listed race whilst the other was winning a maiden contest. I appreciate that many Readers will not agree with this, it is difficult to see how assigning different ratings for the same performances can be justified. Also by how much should the rating of the horse which contested the higher class race be increased?
Although it appears that race class has no part to play in the calculation of speed figures there is some logic to the assertion that horses moving down in class, and hence competing against less able runners, should have a better chance of success, with the reverse true for horses moving up in class. But do the data support this hypothesis, and can it be used to improve our betting profits?
Taking all age flat handicaps run on turf (those not restricted to just juveniles of three-year-olds) as a starting point, over the last five seasons there have been approximately 91,000 performances which have produced a win rate of 10% and a level stake loss at Bookmakers' starting price of 20p/£. Of these runners, horses dropping down in class from their last run, as defined by the BHA, also won 10% of races and lost 19p/£. Those remaining in the same class repeated the 10% success rate but lost more, 21p/£, and runners moving up the class ladder won 9% of contests but lost just 17p/£. These results do not provide the type of support we may have expected for our previous contention that horses dropping in class should find it easier and win more often, though the win rate does drop very slightly for those moving up.
In an attempt to remove extraneous factors as far as possible I reanalysed the data but focused just on the favourites. The overall loss dropped to 10p/£ and the win rate increased to 26% which is in line with expectations. But what about the class movers? Well, those dropping in class improved this win rate to 27% and lost just 8p/£. Those running in the same class mirrored the overall win rate but lost slightly more at 13p/£. But most surprisingly, favourites stepping up in class won at a rate of 28% and lost the least at just 5p/£. So far from the horses dropping in class being the best bets, it would appear from these results that followers of market leaders are best advised to concentrate on those tackling a higher race grade than previously.
The analysis, so far, has produced many large negative figures which is due mainly to the degree of under-pricing by Bookmakers in these races and the fact the we have used starting prices rather than the more favourable early morning prices. So how can we get some plus signs appearing in the results? First, it is necessary to focus on the category of horse which appears to offer the best starting point, namely horses moving up in class. Second, switch to exchange off time prices. And third, look at a range of variables which are logically connected to the class change or which many bettors feel are negative in terms of their influence on the race outcome.
At exchange prices the loss for horses moving up in class changes to a 1p/£ profit which, from almost 21,000 performances, is a decent platform to work from. Three- four- and five-year-olds all made a profit of 5p/£ or more before commission from approximately 14,000 bets which for some would be a satisfactory return, though a couple of slightly negative years from the last five is a little off-putting.
An analysis by course absence, though, is more informative. In general terms horses which have been off the track for an extended period of time are often overlooked, but for horses racing in a higher class contest than previously this combination is particularly profitable. Runners returning from a break of at least 100 days, and stepping up in class in all age handicaps on turf, made 28p/£ from just over 2,300 runs over the last five years. This is a startling figure and although the win rate of 9% is on the low side and there was one losing year in the period, it is a combination I will be monitoring during 2015 with a view to implementing in 2016. Quick returners also made a decent profit from a similar number of runs, but this group of horses is essentially a subset of those generated by the five day rule covered in an earlier article so does not increase our range of approaches.
Slightly surprising, given its prominence in race analysis, was the profit achieved from simply following those runners which won their last race. From over 5,000 performances, last time out winners now racing in a higher class of race made a profit of 10p/£ which I suppose illustrates how negatively a rise in race class is viewed by the betting public (horses which are running in the same or lower class that won most recently lost 6p/£).
Finally the most unexpected result related to the performance of fillies. Usually I would omit fillies from any rules but in this case they are particularly profitable. Over the last five years there were over 5,400 fillies that qualified under the moving up in class rule and of these 11% won, returning a profit at off time exchange prices of 9p/£. The best year was 2011 which produced a 20p/£ profit, 2012 was the poorest with a break even average and the other three years made between 4p/£ and 11p/£. However there were a few long-priced winners, and favourites did lose 5p/£ so backing these in isolation would be a high risk approach.
For flat handicaps restricted to just three-year-olds the pattern is very different. Analysing the same period showed that the average loss in these races was 18p/£ at Bookmakers' starting price, with a win rate of 11%. There was very little difference between the win rates of those horses moving up in class compared to those moving down, but in terms of profit and loss runners appearing in weaker contests lost just 15p/£ compared to -24p/£ for those stepping up. This probably illustrates the way improving three-year-olds are viewed by the betting public as well as Bookmakers who maybe would not want to take any chances with them. Such conservative pricing makes it difficult to find any useful angle in this race classification.