This month we will move away from the examination of past form statistics and look at the relationship between starting price (SP) and 'true' odds.
Most of you will be aware that, on average, betting prices give a fairly good indication of a horse's true chances. Not a perfect indication, but close enough for many punters' purposes.
For example, pre-post prices are used by numerous systems as a guide to horses' true chances. They are also used by exotic betting punters with the assistance of quinella and trifecta formulae for determining exotic bet sizes.
The main difference between betting prices and true prices is that betting prices, whether they be bookie or tote prices, are generally under the odds so as to ensure a profit margin exists for the bookie/tote.
In the case of the tote, the profit margin is around 15%. For bookmakers the margin is similar although it varies from race to race depending on how well the bookie can frame his book We will take the 15% figure as the average profit margin, and hence the average loss suffered by all bettors.
In this article we will examine bookmaker prices and see how they compare with true prices. In particular we will try to find out whether the bookmaker's profit margin (punter's disadvantage) varies with price. Doing so will assist the punter who wants a 'magic' formula to convert bookie prices into 'true' prices. Perhaps more importantly it will identify at which prices, if any, the punter has a better chance of making a profit.
Probably the first to study this topic was Don Scott way back in 1955. His findings were revealed in his books Winning, Winning More and Winning In The 90s. By surveying the results of nearly 20,000 starters in Sydney, Scott found that the disadvantage to the backer got worse as the price lengthened. In other words, the shorter priced horses were the ones upon which punters were most likely to be able to beat the bookies' profit margin. Scott concluded that horses in the range 1/1 to 7/2 provided the least disadvantage to the backer.
To see whether this conclusion is still valid, I set my computer to examine the SPs and true winning chances for nearly 200,000 starters during the two-year period August 1, 1988 to July 31, 1990.
The first test I carried out was on horses with SPs of even money. Horses in this category won only 43% of the time meaning that you would need a price of 11 /8 to break even. Since this 'breakeven' price reflects the percentage strike rate of the even-money horses, it represents the 'true' price of even money horses. By backing these horses at SP you would have lost 13% on turnover.
Horses starting at 6/4 won 35% of the time. This means that the true price (or break-even price) of these horses was 15/8. By backing these horses at 6/4 you would have lost 13% on turnover.
Horses starting at 2/1 won 28% of the time meaning that you would need a price of 5/2 to break even. By backing these horses at 2/1 you would have lost 12% on turnover.
So what about horses starting at 5/2? They won 25% of the time. To break even you would need a price of 3/1 about these horses. Backing them at 5/2 would result in a loss of 11% on turnover.
As far as SPs at 3/1 and greater, the disadvantage to the punter just got worse and worse. Backing all horses at 3/1, 4/1 and 5/1 resulted in a loss of 15% on turnover. To break even, prices required were 15/4,9/2 and 6/1 respectively.
At 6/1 the loss was 18% (8/1 required to break even) and at 8/1 the loss was 99% (10/1 required to break even).
Backing horses at 10/1 and 12/1 resulted in a loss of just over 30%. To break even, prices required were 15/1 and 16/1 respectively.
Backing horses at 16/1 and 20/1 resulted in a loss of just under 40%. To break even, prices required were 25/1 and 33/1 respectively.
Backing horses at 25/1 and 33/1 resulted in a loss of around 50%. To break even, prices required had to be double that available, namely 50/1 and 66/1 respectively.
Finally, backing horses at 50/1 and 100/1 resulted in a loss of around 60%. To break even, prices required were 125/1 and 300/1 respectively.
SUMMARY
| Price | Loss |
|---|
| Evens | 13% |
| 6/4 | 13% |
| 2/1 | 12% |
| 5/2 | 11% |
| 3/1 to 5/1 | 15% |
| 6/1 | 18% |
| 10/1 to 12/1 | 30% |
| 16/1 to 20/1 | 40% |
| 25/1 to 33/1 | 50% |
| 50/1 to 100/1 | 60% |
It is fair to say that by considering price alone, the punter's best chances are, as Don Scott found, on the shorter-priced horses. In fact the findings in this article suggest the best horses to back are those that start at odds of less than 3/1. While this will not guarantee profits, bets at odds of less than 3/1 should fare slightly better than the average loss referred to previously of 15%.
Backing horses at odds of 3/1 and longer simply results in a loss to the punter in excess of the average 15%. In fact it rises rapidly to a massive 60% for horses at 50/1 or more. Unless you have a good reason to think long priced horses might have much better chances than suggested by their prices, you should leave these horses alone.
In concentrating your betting on horses at odds of less than 3/1, your potential loss which is already better than average can be improved further by trying to get over the odds. As we have seen, you only need to get a bit over a quarter of a point extra about short-priced horses to break even. Any more and you will make a profit.
One way to get overs is to anticipate market moves. If you believe your selection is going to firm then you should get on early. If you think it is likely to ease then you should wait and get the best odds. Unfortunately it is not always easy to predict which way the market is going to move so getting overs in this way is not always easy.
Another way to get overs is by looking for the bookie who offers a price that is higher than all the other bookies. For example, if a few seconds before jump time all the bookies have a particular horse at 7/4 and you suddenly get 2/1 from one solitary bookie. It is best to leave this sort of approach until just before jump time as by then you will be in a position to know whether you are getting the best of the prices available during betting.
Of course we shouldn't forget the tote. In addition to being able to get the best bookie prices in the ring, on course punters also have the advantage of being able to compare tote and bookie prices and getting overs on the tote.
YOUR LETTERS
While in the middle of researching this article I received a letter from Mr B. E. of Taree, N.S.W., on the very topic of prices and the punters disadvantage. Mr B. E. was familiar with Don Scott's conclusion that on the basis of price alone, horses starting at evens to 7/2 presented the least disadvantage to the punter. By obtaining the best price for horses in this price range, Mr B. E., claims a 15% to 20% profit could have been made over the last 18 months.
Readers may note, as Mr B. E. did, that Don Scott concluded the disadvantage to the backer for evens to 7/2 horses was just over 5%. The results of my research outlined earlier in this article however suggest the figure is just over 10%. Even if it is over 10% in the longer term, the required 11/8 about even money horses, 15/8 about 6/4 horses, 5/2 about 2/1 horses and 3/1 about 5/2 horses is still achievable.
It is worth noting that Scott's study was of Sydney starters only whereas my study included metropolitan and provincial starters from four States. Next month I will take a closer look at the even money to 7/2 group to see whether the disadvantage to punters is less in any particular State or at metropolitan tracks as compared to provincial tracks.
If we can find a situation where the disadvantage to the punter on short priced horses is even less than average then your task of making a profit from getting overs will be made even easier.
Mr M. M. of New Zealand sent me a very interesting letter asking about statistics on beaten margins versus finishing positions. While realising the importance of beaten margins for rating past performances, being a New Zealander, Mr M. M. does not have access to margins for horses finishing further back than fourth. To help him out, I will start an analysis of beaten margins versus finishing positions next month.
As well as helping our fellow punters in New Zealand, we should find out whether all placegetters are equally likely to win their next race or whether those that finished closer to the winner last time out are better propositions. We will also look at horses that may have been beaten by half the field or more yet were still only a few lengths behind the winner.
Click here to read Part 5.
Click here to read Part 6.
Click here to read Part 1.
Click here to read Part 2.
Click here to read Part 3.
By Neale Yardley
PRACTICAL PUNTING - JUNE 1991