Handicap Ratings in UK racing

Handicap Ratings

Handicap Ratings: giving everyone a chance

A day doesn't go by when a racing writer or broadcaster will refer to a horse as being 'six pounds well in', or 'racing from out of the handicap', or 'favoured by the weights'.

But what does that actually mean? And are there any anomalies that we can use to our advantage? In today's post, I'll look at handicap ratings in the context of UK racing, and try to find any related chinks in the armour of the betting masses.

So let's get stuck in, starting with handicap ratings.

Handicap Ratings

What are handicap ratings and why do we have them?

Handicap ratings, in their simplest form, are a set of numbers used to define the ability of a group of horses. Many people keep their own set of private numbers, figured out according to method and structure (allied to a hint of subjective opinion).

The reason for these ratings is to group horses together in races, called handicaps, where each animal has a theoretically equal chance of winning. There are numerous other reasons as well, such as helping to ascertain the value of a horse in training at sale, and as a means of establishing the merit of horses in different years or generations... or even in the same year.

From a betting perspective, the main purpose of handicap ratings is to understand which horse or horses might be favoured by how much lead they carry. More on that in a moment.

Who produces handicap ratings?

As I mentioned above, there are many people who keep their own records and ratings on horse performance. These are known as 'private handicaps' and, assuming the owner has some talent when it comes to rating races and horse performance, they have real credibility when it comes to betting.

The reason for this is because information from a private handicap is, by definition, not available to the general public. So, whilst in the main, a private handicapper's interpretation of a race will align with the 'official' view, sometimes there will be horses who stand out on private ratings. For example, a horse which is 7/1 with the bookies but is clear top-rated on a private handicap, may be a cracking wager if the private handicapper knows what he's doing.

But the main producer of handicap ratings, and the official rater of horses in the UK, is the British Horseracing Authority (BHA), led by Phil Smith. There is some excellent information on what the BHA handicappers do here.

How are handicap ratings produced?

In most cases, a horse must run three times under rules before being allocated a handicap rating. These runs give the handicapper an opportunity to see the relative merit of a horse, when set against other horses in the context of a race. Of course, the race conditions - distance, going, course constitution, etc - may also play their part, and makes the job of handicapping horses initially the biggest challenge... especially when some trainers may want to 'disguise' the ability of the beast in question.

This is from the BHA site again:

Once a horse has qualified to run in handicaps, we will usually publish a handicap rating for it.  A full list appears every Tuesday morning on the BHA’s Racing Administration website.

Most handicaps are limited to horses with handicap ratings in a specific range.  If a Flat race is for horses rated 56-70, for example, then nothing rated higher than 70 is eligible for entry.  Horses rated lower than 56 are allowed to run but they would normally be given weights as if they were rated 56 irrespective of how much lower than this they are actually rated. 

It would not normally make sense for the trainer of a horse rated 46 to enter.  He would have to carry ten pounds more than the weight that we think would give it an equal chance.  Any horse running under those circumstances is said to be “out of the handicap”.  It would be better to run in a race where he would have his proper weight and a proper chance.

In most races, the deadline for entering a horse is noon six days before the race. 

Once that deadline has passed, the BHA publishes the list of horses entered together with the weights they have been set to carry.  The trainers then looks at the opposition, consider all the variables and decides whether they want the horse to run in the race.  On the Flat that decision has to be taken by ten o’clock on the morning two days before the race.  Over jumps the deadline is usually ten o’clock on the morning before the race.

How do handicap ratings transfer to weight allocation?

In order to convert ratings into weight, we need also to factor in a third variable, 'class'. Class in the context of horse races can be most simply described as a range of ratings which seeks to match horses of similar ability against each other.

Each point on the rating scale equates to a pound of weight in a race.

For instance, the above 56-70 example seeks to match horses rated within a stone of each other (70, the top of the range, minus 56, the bottom of the range, equals fourteen; and there are fourteen pounds in a stone).

So if Dobbin was rated 68 and Mister Ed 60, then Dobbin would carry eight pounds more weight in the race than Mister Ed (68-60=8). The specific weight they'd carry - as opposed to the relative weights outlined just now - would depend on who was the highest rated horse in the race.

The highest rated horse, who in this example could not be rated more than 70 (except under a certain condition which I'll get to later), would generally carry 9-10 (nine stone ten pounds). The other horses, rated lower, would carry one pound less per ratings point below the highest rated.

Again, Dobbin would be set to carry 9-08 (rated 68, two pounds below top rated/weighted), and Mister Ed would lug 9-0 (another eight pounds lower rated, and therefore weighted).

Still with me? Great!

Class Band Exceptions

Sometimes when a horse wins well, the owner and trainer will be keen to race him again quickly. This is because, as you've seen above, the handicapper does not reproduce his handicap every day. Rather, he does this weekly.

This presents an opportunity for a fit and well horse to run again before being re-assessed by the handicapper. In a fairly arbitrary attempt to counteract this, winners who are quickly turned out prior to re-assessment are given a five or six pound penalty in flat races, and seven pounds in jumps races.

The handicapper may, in due course, consider the merit of the horse's win to be worth more, or less, than five, six or seven pounds. But five or six (or seven) pounds is the excess a nag will shoulder for racing again after a win but before the official rating has been assessed.

For instance, today - Monday 16th January - Niceonefrankie carries the penalty in the 3.10 Plumpton. Note the 7x next to his name, denoting the fact that he carries the arbitrary penalty prior to tomorrow's reassessment.

But do horses running again before reassessment win often enough to justify their support? The answer is interesting...

In turf flat races since 2008, horses running off the same official rating in a handicap having won a handicap last time, won 57 of 222 races - about 25% - and showed a negligible profit of 10.33 points.

In all weather races in the same time period and under the same conditions, 29 won from 138 runners (21%) and showed a loss of 28..33 points.

And in jump races the story was similar to all weather, with 52 of 228 winning (23%) but losses incurred of 72.22 points.

What I find most interesting is what we see when we compare the performance of horses running under the seemingly arbitrary penalty, and the relative performance of those last time out winners who raced subsequently only after they were re-assessed.

The former group (quickly turned out, no re-assessed) won collectively 138 of 588 (23.47%) for a loss of 90 points (-15% ROI).

The latter group (won last time, re-assessed by the handicapper before running again) won collectively 3,461 of 20,380 (16.98%) for a loss of 2759 points (-13.5% ROI).

Notice how one's negative return in percentage terms is actually better for re-assessed horses. The implication, and take away, is this: horses running quickly under a penalty for a last time out win are overbet. They win more often, but they are less profitable. So the question, as ever, is this: do you want winners? Or profit? 😉

How does weight rank equate to performance?

We've seen so far that horses are weighted according to their rating, and relative to the other horses in their race. But, aside from understanding the mechanics of handicap races, how does this help us punters turn a profit?

Let's look at the performance of flat horses performance in handicaps by weight rank, i.e. highest weight is ranked 1, second highest 2, and so on.

Position in Weights Bets Wins WinStrike P/L Places PlaceStrike
1 23619 3014 13% -4128.71 7916 34%
2 19710 2411 12% -3648.09 6403 32%
3 19527 2254 12% -3478.47 6011 31%
4 19271 2008 10% -4355.82 5686 30%
5 18888 1909 10% -3460.54 5437 29%
6 18116 1606 9% -3700.75 4767 26%
7 16861 1295 8% -4214.45 4180 25%
8 15298 1138 7% -3019.92 3772 25%
9 13271 937 7% -2883.93 3019 23%
10 11112 670 6% -3075.44 2215 20%

Look at the beautiful linearity between a horse's position in the handicap weights, and its win strike rate. Top weighted horses win most often, second top rated win second most often, and so on.

But, further, look at the place percentages. The results are exactly the same, in terms of weight rank and performance.

Of course, this by itself will not make us all rich, as the P/L column testifies. But when betting horses, or putting systems together, it is instructive to understand the relationship between weight rank and winning chance.

Another way of looking at this is to note that the top four in the weights collectively win almost half of all turf flat weights (47% to be exact). So you could expect to collect something half the time by just blindly dutching the top four in the weights.

The above table doesn't contain ROI figures, which are of course important. So let's factor that data in now.

Pos in Weights Bets Wins WinStrike SP_PL ROI
1 23619 3014 13% -4128.71 -17.48%
2 19710 2411 12% -3648.09 -18.51%
3 19527 2254 12% -3478.47 -17.81%
4 19271 2008 10% -4355.82 -22.60%
5 18888 1909 10% -3460.54 -18.32%
6 18116 1606 9% -3700.75 -20.43%
7 16861 1295 8% -4214.45 -25.00%
8 15298 1138 7% -3019.92 -19.74%
9 13271 937 7% -2883.93 -21.73%
10 11112 670 6% -3075.44 -27.68%

We no longer have the same linearity, in profit/loss terms. But look closely, and you'll note that the top weight is still the least losing horse on average. And the top three are - barring what might be a slight anomaly with the fifth rated horse - the least losing trio of horses.

So focus your turf handicap wagering on the top three in the weights, and you can expect to win 37% of the time and lose as little cash as possible from such an arbitrary approach.

Ignoring discipline, and looking at all races - turf flat, National Hunt, and all weather - in UK gives this table:

Pos Wgts Bets Wins WinStrike SP_PL Places Pl Str Win ROI
1 20310 2688 13% -3061.1 6783 33% -15%
2 17176 2136 12% -3164 5473 32% -18%
3 16876 2023 12% -3286.5 5255 31% -19%
4 16736 1853 11% -3303.9 5004 30% -20%
5 16147 1749 11% -2525.5 4757 29% -16% ??
6 15213 1471 10% -3098.5 4215 28% -20%
7 13942 1162 8% -3040.4 3575 26% -22%
8 12167 917 8% -3130.1 3052 25% -26%
9 10407 809 8% -2081.2 2469 24% -20%
10 8328 566 7% -1868 1790 21% -22%

This table covers the period from 2008, and the win/place strike rates have that same perfect curve to them, from top weight to tenth ranked. (Lower weight ranks broadly conform, though as the sample sizes get smaller, so anomalies creep in. These are not material from a statistical perspective).

Again, we see that the top three weights are the least unprofitable of the set, with the same curious exception of the fifth weight. It could be that this is a point in the market where odds are greater than they ought to be. Or it could be anomalous.

I'm inclined to believe it's a curio and, mercifully, it has no monetary merit in any case - more losing less cash, rather than actually winning anything! - which relieves me of the temptation of investigating further.

Is weight change in handicaps even important?

Weight in racing is a divisive subject at the best of times, and there are plenty of learned students of the game who will attest to ignoring weight change in handicaps.

Nick Mordin wrote in his excellent book, Mordin on Time,

"The normal sort of weight swing that occurs from one race to another can only affect a horse by a few lengths, and if this is enough to cost the typical horse you bet a chance of victory, then your bets are much too speculative".

Whilst I don't agree with the ferocity of the statement, I think the general principle is true. My own interpretation is that, given the average weight of a thoroughbred racehorse might be estimated at 1200 pounds, adding seven pounds to the burden may not slow the beast down that much.

There is however a more material point about the nature of weight in handicaps. And that is this: horses who win or run well are allotted additional weight. But horses who run poorly have their rating reduced and therefore carry less weight subsequently. But which group performs better: those carrying more weight? Or those carrying less?

In the same book, Mordin quotes a US study by a chap called Rennets Alexandria, who found a large sample of horses which were running under nigh on identical conditional for two races in a row, and always over the same distance.

The study showed that the group of horses carrying more weight in the second race required, on average, an extra three pounds of weight to slow them down by one length in the race.

Compare this with the group of horses which were carrying less weight in the second race who required, on average, a drop of 6.2 pounds to improve their performance by one length in the race.

Weight of course is only one variable, and we need also to consider the fact that horses gaining weight are generally improving in their form cycle, and those losing weight are regressing in theirs. That alone, argues Mordin, is not sufficient to vindicate such a considerable disparity between the two sets of horses.

He contests that class must also be a factor. After all, when a horse wins a 0-70 race off topweight, he will be obliged to race in a higher class race next time (at least, after the handicapper has re-assessed his performance). In that higher class race, he will probably carry less weight, but against better animals.

Let's use British examples to illustrate the same point. The first table below shows those horses which won last time out, over the same distance as their next race, which was a handicap. That next race was in a higher class, albeit carrying less weight.

Bets Wins WinStrike P/L Places PlaceStrike
6662 957 14.37% -1059.1 2465 37.00%

A one in seven win rate, and a loss in ROI terms of 15.9%

Now then, this next table shows a similar group of horses which won last time out, over the same distance as their next race, which was a handicap. But in this case, that next race was in the same class or lower, and the horse carried more weight.

Bets Wins WinStrike P/L Places PlaceStrike
6548 1249 19.07% -786.92 2883 44.03%

Here, we're confronted with a roughly one in five win rate, and a negative ROI of 'only' 12.02%

In other words, horses carrying more weight in the same (or lower) grade after a win comprehensively out-perform those upped in class after winning.

Jockey Allowances and Handicaps

Apprentice (flat races), amateur (both codes) and conditional (National Hunt) jockeys are generally all entitled to claim an allowance, depending on their ability - measured in terms of races won to date - and the conditions of the contest in which they're riding.

Some trainers consider these jockeys' allowances more than offset their relative inexperience and inability (in some cases, though not all!)

For instance, a seven pound claiming jockey on the flat may have already ridden nineteen winners in his/her first season. Alternatively, they may still be seeking their first win after a hundred or more rides under rules.

It's difficult to quantify the value of amateur jockeys, except by looking at the profit and loss tables. The most recent of these, which advertise the prospects of Richie Killoran, Kielan Woods and Lee Edwards over jumps; and Harry Bentley, John Fahy and Ryan Clark on the flat; can be seen here.

Latest apprentice/conditional jockey stats.

Final takeaways on handicap ratings

With almost half of all UK races run as handicaps, a decent grasp of the mechanics is a necessity if you aim to make betting pay. But that alone will not put your nose in front. For that, we need to dig more deeply and consider actual performance of groups of horses versus the conventional wisdom.

We've seen in the above that a horse's winning chance is directly proportional to its position in the weights, with those at the top favoured.

We've also noted that following horses quickly turned out prior to official re-assessment may win more often than those whose next run is post-assessment, but they also lose more money.

And we've learned that it is generally better to look to horses carrying more weight in a similar class race than those carrying less weight in a better class race, after a win last time out.

None of these pointers will in or of themselves make you a winning punter. But each, when factored into your own betting approach, and drawn upon as you study the cards - along with all the other variables which make racing such a glorious puzzle - will bring you closer to the elusive goal of making a profit from your betting.

Matt

p.s. Please do rate this post, and/or share it, using the buttons below. That's what they're there for! 😉 And, of course, feel free to leave a comment.

Your first 30 days for just £1