Regression to the mean

Question: Can I apply regression towards the mean with Horse ratings?

TRB Response: The concept of regression toward the mean definitely applies to horse performance and ratings. It's a key concept.

Each horse has a given level of talent and over time their form is made up of runs that are peaks beyond their general level of talent and runs well below.

A horse that has typically rated 88-89 in its good runs and suddenly rates 93, shouldn't necessarily be expected to rate 93 next start, especially if it's well exposed in terms of a number of career starts and runs this preparation. The horse is more likely to regress next start, perhaps not all the way to 88-89 but still lower than 93.

Think of it like a cricketer that might have an average of 40, then comes out one inning and scores 100. If you are rating the expected performance of that cricketer next time, you intuitively wouldn't expect him to get 100 again.

The higher up the rating scale a performance was and the more it was above the horse's general level of talent, the more likely it is to regress. Horses in lower class races with lower ratings still regress, but if you have been an 80 rater in the past, plus or minus a little and suddenly rate 85... then it's easier / more common to hold that level of performance than a horse that generally rates around 90 on their best form and suddenly rates 95.

The same applies if the horse suddenly rates poorly. There are all types of reasons why horses rate lower and you can typically expect them to progress more towards their median level of form next start. Consecutive poor runs or sustained periods below past mean levels of performance can also mean a horse is out of form or perhaps past their best. You have to use your judgement.

When it comes to younger, lightly raced horses and those early in their campaign then you need to be more liberal with your judgements. Horses can be 'on the up' and improving to still establish their typical level of talent. Again, the level of ratings plays a role. It's very easy and common for horses to go from 80 to 85. It's more difficult, but still certainly possible for horses to improve from 90 to 95. It's much more difficult and rare for a horse to improve from 95 to 100 and hold that as a typical level of good form. SPs can also play a role. If a horse has been starting short in the market then there is an expectation of talent (relative to that race) and improved ratings can be trusted more than if a horse starts at long odds and overachieves against expectations.

There are no fixed rules or algorithms you can apply. It really is an art form that you can develop over time and is very much on a horse-by-horse basis. The test is to ask yourself "in the context of this horse's age, starts, stage of prep, overall history and recent form... keeping in mind regression to the mean (positive or negative), what rating figure reflects a level of performance that is reasonable to expect from this horse?"

There's no reason why you can create a couple of different race/rating scenarios if there are key horses that present some uncertainty in terms of recent form against their mean. How does it change your view of the race if you make different assumptions about how a key horse might regress either up from a poor run or down from a spike? How does that affect your betting decisions?

The purpose of adjusting ratings is simply to help identify good betting prospects so always keep that in mind. Don't get caught up in the mechanics of tinkering with ratings and forget the actual purpose. There's no real science or precision, they're just assumptions and forecasts that help lead to betting decisions that make sense to you, where the prices on offer seem worthy for the uncertainties faced. If your view on a race doesn't offer any clear betting opportunities then it's typically best to move on to another race. There are plenty of opportunities that present themselves without spending a large amount of time trying to mould a race to fit an existing belief or outcome... whether it be the betting market or a horse you would like to back.