Analysing Market Data in Handicaps
In this article I will be looking at a different approach to analysing market data, writes Dave Renham. I will not be using the traditional methods of rank – favourite, second favourite, etc. – or using a specific price or price band. My focus is going to be solely on handicap races. The data has been taken from the last four full years of UK racing (2020 to 2023) and I am looking at races of between 6 and 14 runners.
My plan is to split the betting market into three thirds – in the same way that I do when I analyse draw bias. The idea here is straight-forward: horses with the shortest prices will be classified in the top third of the market, more mid-range prices will be classified in the middle third of the market, while bigger prices/outsiders will be classified in the bottom third of the market. I have used Betfair SP rather than Industry SP to do this. The reason is that with Industry SP there are more prices that are the same, which can make the splits into thirds more challenging.
One slight downfall of this method is that there is not always an even split due to the actual number of runners. However, I will use the same principles as I do for my draw bias work meaning that it should balance out fairly as the table below shows:
Hence, with 7, 10 or 13 runners, the middle third gets the extra runner, while races with 8, 11 or 14 runners the top and bottom thirds get the extra runner. Hopefully therefore, we will get a fairly accurate reflection of how the market behaves.
Each third will produce a percentage figure – in simplistic terms let us imagine a series of 100 races. If the top third of the market won 61 races, the middle third won 28 and the bottom third won 11, then the market percentages would be 61% for the top, 28% for the middle and 11% for the bottom. Clearly, we are not going to get ‘equal’ percentage splits of 33.3% for each group given the inherent accuracy of the betting market.
Handicap hurdles
I want to look at handicap hurdle splits first, and these are shown in the pie chart below:
These figures show the expected bias to the shorter priced runners (top third of betting). Close to two in every three races has a winner emerging from that portion of the market. Just one in ten races goes to the outsider group (bottom third of the market) which is perhaps slightly fewer than most people may think, me included.
Next, I grouped the races into smaller fields (6 to 8 runners), medium sized fields (9 to 11) and bigger fields (12 to 14) looking for differences. The findings were as follows:
As you can see the market bias to the front end has strengthened as the field size increases. This is a useful nugget to be aware of, especially if you strongly consider prices and market position. It also makes some sense if you consider the likely prices on offer within the specific runner groupings.
The next port of call is handicap chases.
Handicap chases
I have split the data for these races in the same way and am interested to see how well the figures match. It’s expected that they will correlate quite strongly, but how strongly? Let’s look at the overall market third splits first:
As expected, the splits are like the hurdle ones, but the outsider group have performed better in these races, striking at one win in eight compared with one win in ten. This is a relatively significant shift considering their outsider group status. I must admit I had expected outsiders to fare worse in handicap chases rather than better, as my perception was that handicap hurdles were harder to predict than their chasing counterpart.
Time to group the races into the three field size groupings as I did earlier. Here are the splits:
These figures do not quite fit the same pattern as handicap hurdles although there are some similarities, such as the top third of the market have the lowest win percentage in the 6 to 8 runner group once more, and by some margin. The bottom third of the market performed best in the 12 to 14 runner group, winning almost twice as often as in handicap hurdle races with similar field size.
Overall, the data is pointing to outsiders having more of a chance in handicap chases, and the front end of the market having more of a chance in handicap hurdles.
All-weather handicaps
Finally, it’s onto the sand to see what patterns emerge there. The overall market splits look like this:
These results are very similar to the handicap hurdle ones. The outsider group have performed the worst here with just 9.7% of wins. The front end (top third) has performed the best of the three race codes, with 64.1% of races going to that group.
Let’s now look at the splits within the field size groupings:
This conforms more to the pattern of the handicap hurdle data with the market more influential as field size grows. The top third of the market are very close to winning two-thirds of races when the field size is 12 to 14. Conversely, the bottom third of the market win only around one in every 12.
For the all-weather cohort, the data covers nearly 7000 races so it is a decent sample! Hence, I thought it would be interesting to split the market data up by class of race. Here are the findings:
There is a definite pattern here, with the front end of the market performing better as the races get weaker. At the other end of the market the reverse is true as one would expect given that initial finding. Hence, in the two better classes of race (2 and 3), the outsider group have performed better than in mid to lower class races (4 to 6). They have particularly struggled in the lowest Grade of Class 6 handicaps winning less than 9% of all races.
To conclude, when we look at the different race codes – handicap hurdles, handicap chases and all weather handicaps – the win strike rates by market splits are similar but there are important and subtle differences, too.
*
Selected Cheltenham Festival Handicap Market Trends
With the Cheltenham Festival not too far away, it is worth looking at past trends of some of the big handicap races. I have looked at two Cheltenham races, the first of which is the Pertemps Handicap Hurdle over 3 miles. I have data for the last 26 renewals and the market splits for the winner are as follows:
These are the sort of figures we have seen throughout this article although horses from the bottom third of the market have slightly under-performed in this race providing less than 8% of the winners. Now this race does have big fields so given much of the data we have seen so far, this is perhaps no huge surprise.
The Plate handicap chase (2m 4½f) is the other race I have looked it, and the breakdown for that is as follows:
Here is a quite different picture with a much more level playing field, implying a far more open contest. Of course, 26 races is a smallish sample, but ‘the world and his wife’ shares Cheltenham race trends, and usually not as many as the last 26 races!
If you are a fan of big race trends, I think this type of market breakdown/analysis is a better idea than say looking at the performance of favourites, or the performances of horses priced 25/1 or bigger, etc. It gives more of an overall feel, in my view.
*
I hope you found this slightly different perspective an interesting read. I plan to use this method to dig into turf flat races at some point in the next few months, and will share my findings then.
- DR
A great article Dave and perfectly timed as I was thinking about the handicap world just a few days ago. Would also be interesting to see where the greatest LSPs reside in your categories.
Thank you, very helpful.
Paul.