Betting Strategy / Simon Rowlands / 14 July 2010 / 2 Comments Free £25 Bet

Our expert blogger continues his series on race standardisation with an in-depth look at assessing handicaps

"In effect, handicaps test the accuracy of the official assessment of horses - which is what determines the different weights those horses carry - in a real-life environment. The accuracy of your or my assessments of those horses is a consequence of that."

Following last week's blog on race standardisation, the next step is to consider how similar principles can be extended to races in which we know much more about the runners involved.

While we may have a greater knowledge of the runners, that should not lead to acceptance of so-called yardstick handicapping. It remains best to get a handle on the form of a race overall before making assumptions about which horses have or have not "run to form".

Handicaps are the easiest races of all to assess, which is useful as they represent a large proportion of all races in Britain. The crucial difference between handicaps and the races we looked at last week is that we need to consider a horse's performance relative to its mark and not in absolute terms.

In effect, handicaps test the accuracy of the official assessment of horses - which is what determines the different weights those horses carry - in a real-life environment. The accuracy of your or my assessments of those horses is a consequence of that.

In simple terms, handicaps are won by horses that are proving they are well handicapped. Again in simple terms, horses that finish close up are often proving that they are well handicapped - if not quite so well handicapped - too. Horses that are beaten may have excuses, but horses that finish down the field repeatedly are usually showing that they are not well handicapped.

There are exceptions to this - such as with horses that are greatly advantaged or disadvantaged by the run of the race or with handicaps which are far weaker or stronger than the norm - but exceptions are exactly what they are, and they can usually be spotted in any case.

In general terms, a mature horse winning a 12-runner handicap can be expected to have exceeded its BHA mark by about 5.5 lb. A runner-up can be expected to have exceeded its mark by about 2.5 lb, a third to have dipped about 0.5 lb below its mark, a fourth to have dipped about 3.5 lb below its mark, and a fifth to be about 6.5 lb below its mark.

Of course, horses do not conveniently finish at regular 3 lb intervals, and nor do they always run in 12-runner races of equal strength. Adjustments need to be made for the specific margins between the horses and for the specific races in which they run.

The smaller the field, the weaker the race, and the more restricted the race (such as when confined to a given age or sex) the weaker it is, too. Nonetheless, a process similar to that which was explained last week, involving weighting according to position at finish, will result in a ballpark figure of acceptable accuracy.

Here is an example of how it works in practice (for the purposes of this illustration, it is assumed that no adjustments need to be made for field sizes, restricted races, age range and horses running from out of the handicap etc).

Standard: +5.5, +2.5, -0.5, -3.5, -6.5 (first 5 ratings compared to handicap marks).
If all horses ran off the same mark of 100 and returned at 5 lb intervals (as in last week's example), the calculations would be: winner 105.5 rating on standard; second 102.5 rating on standard (making winner 107.5); third 99.5 (winner 109.5); fourth 96.5 (winner 111.5); and fifth 93.5 (winner 113.5).

Weighting this as in last week's example comes up with 107.9. If the winner in this example were to be rated 108, then the second would be 103, the third 98, and so on.

Of course, most handicaps involve horses running off different marks. The arithmetic is a bit more involved but the principles are the same.

Taking last week's Bunbury Cup at Newmarket, won by St Moritz, as an example, the first five ran off marks of: 96; 105; 97; 95; and 93. And I make the difference at the weights: winner 8 lb < second; winner 6 lb > third; winner 9 lb > fourth; and winner 11 lb > fifth.

The standardisation calculation is as follows: winner's rating 96+5.5=101.5; second's rating 105+2.5=107.5 (in which case winner's rating is 107.5-8); third's rating 97-0.5=96.5 (winner's rating is 96.5+6); fourth's rating 95-3.5=91.5 (winner's rating is 91.5+9); and fifth's rating 93-6.5=86.5 (winner's rating is 86.5+11).

The weighting results in 100.6, according to Zipf's Law, but the figure needs to be revised upwards to take into account the larger field size and more competitive nature of the race than normal. A standard figure for the winner in this instance would be 102 on the official scale, making the runner-up 110 (we have already established that it is 8 lb > the winner considering the weights), the third 96, the fourth 93, and the fifth 91.

As with last week, it is crucial to test the validity of this initial assessment against subsequent events and to revise it where necessary. The evidence over many years is that race standardisation applied to handicaps is a good starting point in the large majority of cases.

Next week, in the final part of this short series, I will deal with non-handicap races in which, unlike in last week's example, we do know plenty about the horses involved and in which past form may be a consideration.

Tags: Bunbury Cup, Race Standardisation, St Moritz, Yardstick Handicapping, Zipf's Law