In this, the final part of a three-part series, PPM's computer expert Neale Yardley passes on the results of his computer research aimed at weeding out the important from the not-so-important formguide details.
Last month I suggested that the more wins and places a horse has had over its recent starts, the more likely it is going to be fit enough to win a race.
I also pointed out, however, that it is possible for a horse to have had too many wins and be near the end of its winning streak. An analysis of preparation profiles will help us decide whether this is about to occur.
PREPARATION PROFILESI will use the term preparation profile to describe the variation in a horse's performance as it progresses through its preparation from being first-up to its final run before its next spell.
Studying preparation profiles will help you identify periods of fitness. It will also illustrate the fact that different horses reach their peak performances at different stages of their preparations. Some horses, for example, peak at the beginning of their preparation while others peak in the middle or towards the end of their preparations.
If a horse is going to win a race, it is most likely to win during one of these periods of peak performances. Since not many of us have the time or data to be able to find patterns in every horse's previous preparations, we must rely on a few rules of thumb to predict when these winning performances are likely to occur.
Results show that most horses need a number of runs before they are fit enough to win a race. This is especially true when it comes to the class of races held on Saturdays and public holidays. (The only exceptions are those above-average class horses that have proven on more than one previous occasion their ability to win first-up from a spell.) At the other end of their preparation you will find that horses are less likely to win after too many runs without a spell or let-up.
With the exception of horses that win very early or first-up in their preparation, we can attempt to predict when a horse is going to reach its period of peak performances from the number of runs it is into its preparation. This information is fairly readily available from most good formguides since all we need to look at are the dates of horses' recent runs. (A knowledge of previous preparation profiles will only be necessary when we want to identify the ability of first-up horses.)
The first rule of thumb we can make is that the majority of horses are unlikely to win a race before their third or fourth race into their preparation. These horses are easily picked as any horse that has only had one or two runs during its current preparation will have an or 'S' against its name in its recent form numbers, for example 52X6 for a horse that has had only one run in its current preparation and 3X74 for a horse having had only two runs in its current preparation. The horses we are interested in are those without these spelling symbols against their names.
It is a little bit more difficult to predict when horses are likely to reach the end of a sequence of peak performances since some horses can race for up to six months without needing the benefit of a spell while others only perform well over much shorter periods. It is easy to note when a horse starts preforming poorly (e.g. a sting of wins or places is followed by a number of unsuccessful runs where a horse is unplaced at more than five lengths behind the winner) but it is not as easy to predict a decline in form before it actually happens.
The best way we can try to predict a decline in form in advance is. to refer to the number of starts a horse has had in its current preparation as we just did when tying to predict when a horse is about to begin a winning streak. We can also look at whether a horse has won 'too many' races as we will see a bit later.
Since the best of formguides like the Sportsman, Truth or Sporting Globe give details for horses' last six to eight mm or so, we can at least tell whether a horse has had more or less than this number of runs in its current preparation. In the absence of additional form, I suggest you treat with caution those homes that have had more than six to 8 runs in their current preparation (ie. those for which no spell or let-up breaks can be identified between any of their previous runs listed in the formguide).
You should be careful in applying this rule of thumb because I am not suggesting that horses having had more than six to eight runs in a preparation cannot win. Rather that horses having had only three to six runs since a spell have, on average, better chances of winning.
The following system of points allocation may be useful to help you apply the above-mentioned criteria for identifying when horses are likely to be at their fittest.
- Resuming from a spell attracts a points score of zero.
- One start back from a spell attracts a points score of one.
- Two starts back from a spell attracts a points score of two.
- Three to five starts back from a spell attracts a points score of four.
- Six or more starts back from a spell attracts a points score of two.
Like the points systems I have discussed in the first two articles of this series (for fitness related factors of days since last start and win consistency), the horse with the highest point score is the most favoured.
While on the topic of the first two articles, you should note that best use of the points system described above will be achieved if you add the points obtained to those described in the previous two articles. In this way you will arrive at a total point score that takes into account days since last start, win consistency and number of starts into current preparation.
I made passing reference earlier on to the possibility that a horse can have 'too many' wins. What I mean here is that a horse can only win so many races *m a row before tiring and needing a spell. You should be careful backing horses in this category as their chances of winning are reduced and the price on offer is usually much less than it was a few starts back at the beginning of the horse's winning streak.
FORM
So far m this series I have considered form details that are indicative of fitness. This is because fitness is arguably the most important factor when it comes to picking winners. After all, if a horse is not at its peak fitness then it represents a risk no matter how highly it rates under other considerations.
Probably the next most important factor is form, or rather form improvement. Homes that are performing better from one start to the next are improving m form and expected to continue doing so. Such improvement can be identified by examining preparation profiles (if we graph performance against time) but to do so we need to measure performance.
As many of you will know, performance is best measured by class weight ratings which take into account weight carried, class of race and beaten margin. The formula for calculating such a class weight performance rating is to. take a horse's recent start, allocate it a race class rating, add the weight carried over the limit by the horse and subtract one and a half for every length by which the horse finished behind the winner.
This calculation is best done over at least a horse's last two starts. An upwards or downwards trend can then be identified and a prediction then made as to how the horse might rate at its next start. The prediction can be made by simply assuming that the difference over the last two start ratings will be duplicated over the next two. A couple of examples will illustrate this.
Suppose a horse improves its rating by three from 45 to 48 over its last two starts. Assuming an equal increase in rating to its next start we can predict a rating of 51 for the horse's next start (48 plus 3). Similarly a horse that has rated 57 and 55 can be expected to return a rating of 53 at its next start.
These weight ratings should be adjusted for weight to be carried over the limit in the hones next race if meaningful comparisons are to be made between horses in a race. Those of you with computers can write to me for a free program that automatically calculates past performance class weight ratings in the manner just described.
Everything discussed so far in this series of articles can be used as the basis of a ratings system since all you need do is take a horse's predicted rating and add to it the point scores for days since last start, number of recent wins and number of starts into current preparation. The final number arrived at can then be used to rank each horse in a race so you can identify the top two or three to concentrate your bets on.
In concluding this series I must stress that horses have to be fit if they are to have any chance of winning a race.
My studies have shown that consideration of factors like days since last start and recent win consistency are essential if you are going to have any hope of separating the fit from the not-so-fit horses in a race. Although class weight ratings form an important basis for handicapping too many people over-emphasise horses' chances because they ~ high class weight ratings can make up for lack 'of fitness.
The fact that many punters incorrectly place less importance than they should on fitness means that if you take dose note of fitness when picking your selections, you will be more likely to get good, priced winners. The theory here is that if you do something different from everybody else and it is right then the prices about your selections will be much better than everyone else gets for their selections.
Being different can often be profitable. I will discuss this topic next month in a new article that will tell you how you can make money on the tote by concentrating on bet types that are being increasingly ignored by the majority of tote punters.
Click here to read Part 1.
Click here to read Part 2.
By Neale Yardley
PRACTICAL PUNTING - NOVEMBER 1989