Percentage of rivals beaten

Tag Archive for: Percentage of rivals beaten

Evaluating Jockeys by Percentage of Rivals Beaten

3 June 2025/in Dave Renham/by Dave Renham

In this article I will put 35 jockeys under the microscope, writes Dave Renham. These are the riders with the most rides per year, on average, over the past four years. The data has been taken from UK flat racing (turf and all-weather (AW) and the full years 2021 to 2024.

Introduction

I have further limited the findings to mounts sent off at an Industry Starting Price (ISP) of 20/1 or shorter, in order to try to eliminate most of the horses that had little or no chance; and, further, because very big-priced winners tend to skew profit figures.

For this piece I will primarily examine the data using ‘Percentage of Rivals Beaten’, although I also plan to look at strike rates and A/E indices. Percentage of Rivals Beaten (PRB) is a calculation based on a horse's finishing position in relation to field size. It makes key distinctions between a horse finishing, say, fourth in a seven-horse race (PRB 50%, three rivals beaten, beaten by three rivals) and finishing fourth in a sixteen-horse race (PRB 80%, twelve rivals beaten, beaten by three rivals). We express the PRB as a number between 0 and 1. So, in the examples above, 50% is 0.5 and 80% is 0.8.

As racing researchers we can often be blighted by small sample sizes when analysing, for example, win strike rates. Hence, there is a strong argument to suggest that PRB figures are a more accurate metric, simply because they make datasets bigger: they award a sliding performance score to every runner in every race, whereas win strike rate only awards the winner a score with all other finishers getting zero.

Today's offering has a slightly different flow from usual I will be writing it "as I go along". In other words, I’m sharing the research and my thinking process stage by stage, rather than doing all the research and then writing about my findings afterwards. Thus, my main commentary will appear to be in the present tense. If that makes sense, let's crack on (and if it doesn't, it soon will!)

Top Jockeys' PRB: Overall

I will start by sharing the average PRB figures for each of the 35 jockeys over this four-year period. They are ordered alphabetically across two graphs:

To provide a benchmark, the average figure when combining these jockeys was halfway between 0.58 and 0.59, so 0.585 to be precise. Oisin Murphy has the highest PRB figure, 0.64, followed by five jockeys tied on 0.62 – William Buick, James Doyle, Rob Havlin, Jack Mitchell and Danny Tudhope. Tom Eaves, Cam Hardie and Andrew Mullen have the joint lowest PRB figure of 0.54.

It should be noted that all riders in this sample are above the 0.5 PRB benchmark and so even the lowest in the cohort are out-performing the norm.

Top Jockeys' PRB: ISP 6/4 or shorter

Although I have restricted qualifiers to those priced 20/1 or shorter, there are clearly some jockeys who have more rides at shorter prices than others. Hence, I am assuming that jockeys should have higher PRBs because of this. To help analyse and potentially confirm this hypothesis I am going to look at the percentage of rides each jockey had with horses priced 6/4 or shorter. The table shows the splits:

There is a huge variance here, from William Buick with more than 13% of his rides sent off 6/4 or shorter, to Cam Hardie at less than 1%. Of the six jockeys with the highest average PRBs I noted earlier, five of them were in the top six for the highest percentage of rides (highlighted in blue in this table). Therefore, we can see there is a strong looking correlation between price and PRB, as we should expect.

Top Jockeys' PRB: ISP 12/1 to 20/1

It makes sense next to look at the percentage of rides each jockey had when the qualifiers were bigger prices in order to consider both ends of the price spectrum. Therefore, below is a table showing these percentages when considering percentage of rides from runners priced 12/1 to 20/1.

The three jockeys with the highest percentages (shown in blue) are the jockeys who had the lowest overall PRB figures shared earlier, namely Tom Eaves, Cam Hardie and Andrew Mullen: this is further evidence of clear positive correlation. Also, the lowest four percentages in this group are for Messrs Buick, Murphy, Doyle (James) and Mitchell.

At this early point in my research I am starting to appreciate that despite the fact that PRB is a really useful metric, for this type of research the price of runners is also very important and can significantly sway the balance one way or the other. Hence, the market will be factored in for the remainder of what follows.

Top Jockeys' PRB by Price Range

Having established the importance of the starting price, I have decided to calculate PRBs for different price bands for all 35 jockeys. The brackets I am going to use are again based on Industry Starting Price and they are as follows:

In the table below I have collated the PRBs for each jockey for each price band. The average figures for all jockeys in the list are shown in blue at the bottom of each column, and I have highlighted any PRB that is at least 3% above the average or at least 3% below the average. The 3% ‘above group’ (positive) is highlighted in green, the 3% ‘below group’ (negative) is in red.

The colour coding helps to highlight jockeys that seem to perform above the norm and those that may have performed below what might be expected within each price band. There were three jockeys who obtained two ‘greens’: Robert Havlin, Clifford Lee and Kieran O’Neill. And there were four jockeys who obtained two or more ‘reds’: William Buick (3), Holly Doyle (2), Joe Fanning (3) and Rob Hornby (2).

Top Jockeys' PRB: All-Round Performance

I am thinking that another way we could analyse these data is to simply add up each jockey’s six PRB figures in the above table and compare them. Below, then, are the riders with the top ten combined PRB figures when adding the six values together:

It could be argued that these are the top 10 performing jockeys from my original list of 35 as their totals are based on the overall performance across different price ranges. From looking at these findings I would be happy to see one of these ten riding a horse I am keen to back. Rab Havlin, who has consistently shown positive figures in the research to date, tops the list on a combined total of 3.99. (0.88 + 0.76 + 0.68 + 0.65 + 0.55 + 0.47).

Next, here are the lowest ten combined PRB totals from our sample of the top 35 riders:

As can be seen, we are talking small margins here so despite these ten being at the bottom we know that they are all still top-notch riders. However, in terms of PRB figures within certain price bands, they have performed with slightly less success than the rest of the jockeys in this sample.

To complete the set here are the remaining jockeys (positioned 11^th to 25^th) with their PRB totals. Due to the bigger group, I am using a table rather than a graph:

Top Jockeys: Other Metrics

I stated earlier that PRBs are arguably the most accurate metric but it always prudent to consider other metrics where possible in order to attain a stronger 'feel' for the data.

We know that finishing fifth in an 18-runner race will produce a better PRB figure than finishing eighth in the same the race, but usually finishing fifth does not make punters money (unless those generous bookie types are offering extra places).

At this point, then, I am thinking about the key battles in terms of finishing first rather than second and, therefore, I am going to share the wins, runs, strike rate, profit/loss and A/E indices for all 35 jockeys. As with the PRB data this does not include rides on horses priced over 20/1 ISP. Profits and losses have been calculated to Betfair SP less 2% commission. The A/E indices are based on Betfair prices and any figure above 1.00 has been coloured in green:

Somewhat surprisingly, 18 of the 35 jockeys have secured a profit which is impressive considering there are not any really big BSP winners to skew the returns. In fact, the highest winning BSP was 46.0 and there were only three winners in total above BSP 40.0, and only 23 above BSP 30.0 (out of total of nearly 12,000 winners).

Rossa Ryan, Saffie Osborne and William Buick have the best ROI%s (above 7%), and they each have one of the top five A/E indices. Impressively, Ryan has made a blind profit in each of the four years, Osborne and Buick matching that feat in three of the four years surveyed. There are two jockeys that made a loss in each of the four years, namely David Allan and James Doyle.

Conclusions

All this is helping me, and hopefully you, to start building a more complete picture of jockey performance; or, at least, the performance of these 35 top riders. The PRB data have given us an extra layer on top of the usual metrics we focus on. However, it is becoming clear to me that for this type of jockey-based research we do need other metrics (win percentage, profits, A/E indices, etc) to bring betting utility to the party.

I am just starting to expand the jockey PRB research into other areas and there is plenty more to share; so I have come to the realisation that this article will spawn a second piece. Thus, it is probably too early to draw any key conclusions from the research so far as there are more pieces of the puzzle to add.

However, next week I have a Royal Ascot article ready to go, so it affords me a little extra time to do further digging for part two of this jockey deep dive!

- DR

Early Flat Season Trainer Form

26 March 2025/in Dave Renham/by Dave Renham

After the thrills and many spills of the Cheltenham Festival attention now turns to the start of the turf flat season, writes Dave Renham. Saturday 29^th March is the starting date this year and the crowds will descend on Doncaster for a card that includes the first big handicap of the season, the Lincoln. In this article I am going to look at some early season trainer form and trends. Data are taken from 2019 to 2025, although in 2020 there was no flat racing in the early part of the season due to Covid.

Selected Trainers: First Ten Runs

We see in the racing press plenty of stats connected with a trainer’s recent form, be it the last seven, 14 or 30 days, or their last ‘x’ number of runs. For some punters this information is really important and forms an integral part of their selection process. With that in mind, one question I am keen to address in this article is connected with recent trainer form. I want to try and establish whether the first few runs of the season from a particular stable is indicative of how their runners perform up to the end of April. Also, I will be looking at whether a similar level of performance each year is achieved by trainers up to the end of the first full month of the season. I just would like to clarify that the data shared in this piece has been collated starting from the day of the first turf flat meeting through to the 30^th April in each season.

In order to make this piece manageable I have decided to focus on a selected group of trainers who tend to have a good number of entries in the early weeks of the season. This includes some of the big guns, namely Charlie Appleby, William Haggas and John/Thady Gosden.

My starting point was to work out the PRB (Percentage of Rivals Beaten) figures for each trainer over their first ten runs of each season. I felt that using the PRBs would be the most accurate way of determining how well a stable was performing over those ten runs. Clearly, I could have used win strike rate but over such a small sample size we could potentially get a blurred picture of how well the horses are actually running. Here are my findings.

N.B. I have combined the figures for the Johnston stable although of course Charlie Johnston is now in sole charge:

Early Season Trainer Form: Selected trainers' PRB figures

Before doing a comparison with their records up to the end of April for each year, the table does highlight that we cannot guarantee exactly how well each stable will get out of the blocks each season. Taking Richard Hannon as one example, in 2023 his first ten runners of the turf season hit a PRB of only 0.46, but last year in 2024 it was up at a huge 0.83. Likewise, the Johnston stable has seen wide variances with three PRBs below 0.35 and two hitting 0.65 and above. Now of course ten runs is a small sample but by using PRBs it does give us a better idea of the very early form of a specific stable compared with other metrics. I believe the numbers shared in this table also help to highlight that each year is different and even if stables traditionally start the season quickly, there will be years that for whatever reason things will progress more slowly. And of course, vice versa.

Selected Trainers: To End of April

Let's now take a look at the annual PRBs for each trainer covering the start of the turf flat up to the end of April. Essentially, for most years this equates to roughly the first five weeks of the season.

Early season trainer form: up to end April annually

As might be expected, fluctuations year by year in the PRBs are now less pronounced due to the much bigger datasets, although two of Charlie Appleby’s figures differ quite markedly - from 0.81 in 2022 down to 0.63 in 2024. Likewise, the Gosden stable saw a big difference between their 2019 figure of 0.71 and their 2023 one of 0.51.

Now that we have these two sets of figures we can try to address the earlier question of whether the first few runs of the season from a particular yard are indicative of how their runners will perform up to the end of April. In order to do this, I have picked out some of the trainers to analyse in more detail.

Specific Trainers: Early Season Form

Charlie Appleby

If we look at Charlie Appleby’s performance with his first ten runners in 2019, 2022 and 2024 we can see he has quite well aligned PRB figures (0.62, 0.64 and 0.66). In 2019 and 2024 he maintained a similar level of performance up to the end of April hitting 0.66 and 0.63. However, in 2022, his PRB figure for the longer timeframe soared to 0.81. That year he had 23 winners from 55 runners up to April 30^th equating to a strike rate of just under 42%. From those similar starting PRBs in 2019 and 2024 he managed a longer-term strike rate of 29.2% and 29% respectively. It is difficult to say why those early five or so weeks of 2022 panned out so well for the stable compared with 2019 and 2024 when they started off in the same vein. It perhaps underlines how challenging it can be to predict future trainer form based on a smallish sample of runs.

Mick Appleby

Next for the microscope is Mick Appleby. The graph below shows the comparison:

Mick Appleby: early season form, 2019-2024

The graph shows that Appleby has been consistent in terms of overall performance in the weeks up to April 30^th (the orange line) - four of the five years saw PRB figures within a very small band ranging from 0.44 to 0.46. In 2022 he did have a better overall start to the season hitting 0.53 over those first few weeks, and that year he had started fast with a 0.63 figure for his first ten runners. The 2023 season saw an even better start with a 0.66 10-run figure, but that form tailed off quickly ending up at 0.46 for the longer time frame. Looking at this data tells me that the first ten runs of the year for Mick Appleby would not necessarily have given us a good guide to how the next few weeks would have panned out for his runners.

Going back to his PRB figures for all runs up to 30^th April, despite having similar ones, the correlation with the win strike rates is not completely ‘positive’ as the graph below shows:

Mick Appleby: early season win strike rate comparison

Yes, the best year was 2022, (16.1%), which correlates with the highest PRB figure of 0.53, but there is a big variance between 2019’s strike rate of 14.3% compared with 2023’s 3.2% figure. This is despite having very similar PRB figures in those two years (0.44 and 0.46 respectively).

As with all metrics, any single one does not necessarily give us the best picture. Clearly in 2019 and 2023 the Appleby runners were generally running at the same level overall – the PRB figures show that. However, in terms of winning races 2019 saw many more winners than 2023.

This type of number crunching is an excellent reminder of why racing can be difficult to profit from. Let’s imagine for example we back 20 horses to win in one month, if all of them run really well but all finish second, we still would have lost all 20 bets.

Andrew Balding

Next is Andrew Balding. Again, I have graphed the comparison between the PRB of the first ten runners with that of all runners to the end of April each year.

Andrew Balding early season form: comparison of first 10 runs with all to end April

2021, 2022 and 2023 mirrored each other with both the 10-run PRBs and the all runs to end of April PRBs very close together. In 2019 and 2024 we saw a similar pattern, with the stable flying out of the blocks in those first 10 runs and then slipping back to more normalised figures based on a larger sample.

When looking at those early weeks of the season up to the end of April, Balding does tend to perform at a similar level year on year. If we look at his win strike rate from the start of the season up to the end of April, we can see that in four of the five years they were between 17 and 19%:

There is positive correlation between the PRB figures and the win strike rate in those four years. We saw earlier with Mick Appleby that we don’t always get that positive correlation, and for Balding the 2022 figures paint a similar story. That year saw a lower strike rate despite a similar PRB figure to other years. This highlights once again why it is a good idea, where possible, to look at more than one metric when analysing a set of results in order to get a broader and better overview.

Tim Easterby

Tim Easterby has a lot of runners but his overall strike rate year on year is quite low, both early in the season and taking the season as a whole. Hence his first 10-run PRB figures are the lowest of the trainers mentioned taking the five years as a whole. 2022 saw a poorer start than usual but, by the end of April, he had pulled back to very similar five-week figures as achieved in other years.

Looking at his PRBs for those early weeks up to the end of April, we can see that there is only 0.04 between the highest and lowest ones. Essentially, at the beginning of the season, Easterby has followed a similar pattern every year with similar outcomes. Not surprisingly his win strike rate up to the end of April each year has been low as the table shows:

Tim Easterby early season win strike rate

Personally, I rarely back Tim Easterby horses even if they appear to have ticks in several boxes. For me, finding good value in his team is tricky. On the plus side, his patterns of performance rarely surprise us.

William Haggas

William Haggas has had very consistent longer-term PRB figures (up to the end of April) ranging from 0.60 to 0.67 over the five different years. His PRBs for the first ten runs are more varied as we would expect given the smaller sample size. However, it seems that, year on year, runners from the Haggas stable perform in a similar fashion. Again though, the win percentages up to the 30^th of April have varied much more as the table shows:

William Haggas early season metrics

As we can see, the two highest PRB figures of 0.67 in 2019 and 0.64 in 2023 did not produce the two highest win rates. In fact, they produced the lowest win rates by some margin. 2019 was definitely unlucky for Haggas in those first few weeks as they had 12 second places from their 42 runners that year. Against that, Haggas had only six winners hence the 14.3% strike rate. We talk about luck in racing, and regardless of how good a punter one is, luck and variance are ever-present, sometimes massively.

I would not worry too much about what sort of numbers Haggas posts after his first ten runners this season. We can be fairly confident that his team over the first month or so will run to a similar level to previous years. Whether they win at around 27% or 14% I cannot say, but for readers that back any of his, let’s hope it is nearer 27!

Richard Hannon

For Richard Hannon I want to compare the two sets of PRB figures side by side as I did for Mick Appleby and Andrew Balding.

Richard Hannon early season PRB figures

The orange line represents the longer-term figures up to the end of April and, aside from the 0.47 figure for 2022, the rest lie between 0.51 and 0.59 showing that Hannon's runners perform at roughly the same type of level at this stage of the season year on year.

What I find interesting is the difference between the first 10-run figures for 2023 and 2024, which was huge. 2024 was his best start at a massive 0.83 PRB, 2023 was his worst at just 0.46. However, by the end of the first month although 2024 ended up ‘better’ in PRB terms, the gap was quite small at 0.06 (0.57 v 0.51). Indeed, looking at the win percentages for these two years there was less than 2% in it. 2023 saw a 10%-win rate, 2024 stood at 11.8%.

This is another reminder that looking at a handful of races may not be as important or as useful as some punters/pundits may think; and I am not just talking about the first ten starts of the year. It is essentially the same thing when looking at any 7-day trainer form snapshot throughout the season when a trainer has had ten runners or so during that period. Is that really a reliable enough sample on which to judge how the next few weeks are going to go for the yard in question?

Charlie Johnston

The final yard I want to look in more detail at is that of Charlie Johnston (and its recent incarnations), formerly run solely by father Mark, then by Mark and his son Charlie, and since 2023 by Charlie on his own. Here are the two sets of PRBs:

Charlie Johnston / Johnston yard early season form

I mentioned earlier the huge variances in their opening 10-run figures (the blue line), but despite that the longer term PRBs are all in the same ballpark lying between 0.50 and 0.57. I don’t think the performance of the first ten runners will be that relevant again this year when it comes to predicting what will happen in the subsequent weeks to the end of April. However, we can be fairly sure how they will perform over the longer five-week time frame.

Selected Trainers: Win Strike Rates to end April Annually

To finish off let me share the win strike rates for all trainers for each of the five years based on their runners from the start of the turf season to the end of April:

Selected trainers: early season win strike rates 2019-2024

These percentages can vary markedly year on year, as I meantioned earlier when looking at the performance of the Haggas yard. Luck plays its part for all trainers every year, be it good luck or bad. A few bobs of the head in a finish can make a big difference to the win rate; hopefully Geegeez members will be on the right end of tight finishes more often than not!

That is almost it for this week but for before closing I will put my head on the block and predict the win strike rates and PRBs for all of the stables mentioned in this article from the start of the Doncaster Lincoln meeting this year to the end of April. Here goes:

Projected early season win percent and PRB figures for selected trainers

Hopefully, most of these projections will be close to their mark.

Until next time,

- DR

Horse Racing Metrics: A/E, IV, PRB

20 November 2023/in How to Bet on Horses/by Matt Bisogno

Throughout this site, in editorial content and on our award-winning Gold reports and racecards, there are references to various measures of performance or utility: horse racing metrics. Although some of the concepts may be new, their application – and therefore your understanding of them – is generally straightforward.

This article offers a brief run down of the metrics used, notably Impact Value (IV), Actual vs Expected (A/E) and Percentage of Rivals Beaten (PRB). In the following, I explain how the metrics are arrived at; but if you’re not a geeky type, simply make a note of the ‘what to look for’ component for each one.

Impact Value (IV)

IV helps to understand how often something happens in a specific situation by comparing it against a more general set of information for the same situation.

For example, we can get the IV of a trainer’s strike rate by comparing it with the average strike rate for all trainers.

Let’s say a trainer saddled 36 winners from 126 runners, a strike rate of 28.57%, during the National Hunt season.

And let's further say that, overall in that season, there were 3118 winners from 26441 runners. That’s an average strike rate of 11.79%.

We could simply divide the two strike rates:

28.57 / 11.79 = 2.42

Or we could do the long version, which at least helps understand the calculation. It goes like this:

('Thing' winners / All winners) / ('Thing' runners / All runners)

In this case,

(36 / 3118) / (126 / 26441)

= 0.011545 / 0.004765

= 2.42

What to look for with IV

An IV of 1 is the 'standard' for the total rate of incidence of something. A number greater than 1 relates that something happens more than standard, and a number less than 1 implies it happens less than standard.

The further above or below 1 the IV figure is, the more or less frequently than ‘standard’ something happens.

The example IV of 2.42 means our trainer won at a rate nearly two-and-a-half times the overall trainer seasonal average: 2.42 times, to be precise.

Note that very small data samples can produce misleading IV figures.

IV3

IV3 is a derivation of IV created by us here at geegeez.co.uk to help ‘smooth the curve’ on chart data. You can see examples of this when looking at draw data on this website.

IV3 simply adds the IV of a piece of data to the IV's of its closest neighbouring pieces of data, and divides the sum by three.

For example, the IV3 figure for stall five at a racecourse would be calculated as:

(IVs4 + IVs5 + IVs6) / 3

where IVs4 is the Impact Value of stall 4, the lower neighbour of stall 5, whose IV3 we are calculating, and IVs6 is the Impact Value of stall 6, the upper neighbour of the stall whose IV3 we are calculating.

Thus, in the below example which shows stalls 1-5, the IV3 figure for stall 2 is the average of the IV figures for stalls 1, 2 and 3:

(1.98 + 2.27 + 2.55) / 3 = 2.27

As with IV, the greater the value the better, with anything above 1 representing an outcome which occurs more frequently than standard.

N.B. For the lowest and highest stalls in a race, IV3 is calculated from an average of the stall and its sole neighbour (stall 2 in the case of stall 1, and stall H-1 in the case of the (H)ighest numbered stall).

What to look for with IV3

Used on this site mainly in charts, IV3 shows a smoother, more representative curve when looking at the impact of stall position.

Example IV Chart:

Same data plotted by IV3:

Actual vs Expected (A/E)

Whereas IV tells us how frequently, relatively, something happens, as bettors we need to know what the implied profitability of that something is. In concert, they are a powerful partnership, with favourable figures denoting an event that happens more frequently than average and with a positive betting expectation.

A/E, or the ratio of Actual versus Expected, attempts to establish the value proposition (profitability in simple terms) of a statistic. The 'actual' and 'expected' are the number of winners.

The ‘actual’ number of winners is just that. In the case of the IV example above, the trainer had 36 winners from 126 runners. Actual then is 36.

But how do we calculate the 'expected' number of winners?

We use a formula based on the starting price (you could just as easily use Betfair Starting Price or even tote return if you were sufficiently minded - we've used SP), thus:

Actual number of winners / Sum of ALL [entity] runners' SP's (in percentage terms)

So far we know that to be 36 / Sum of ALL [entity] runners' SP's (in percentage terms)

To establish a runner's SP in percentage terms, we do the sum 1/([SP as a decimal] + 1).

For instance, 4/1 SP would be 1/(4 + 1), or 1/5, which is 0.20,

evens SP would be 1/(1 + 1), or 1/2, which is 0.5,

1-4 SP would be 1/(0.25 + 1), or 1/1.25, which is 0.8, and so on.

The sum of our trainer's 126 runners' starting prices, calculated in the above fashion, is 33.15.

Our A/E then is 36 / 33.15 which is 1.09.

We can then say that this trainer’s horses have a slightly positive market expectation, and in general terms her horses look worth following.

What to look for with A/E

As with IV, a score above 1 is good and below 1 is not good, though in this case the degree of goodness or not goodness pertains to market expectation, or what might be summed up as ‘likelihood of future profitability’.

A dataset that shows a profit but has an A/E below 1 is probably as a result of one or two big outsiders winning. Such runners have a low expectation associated with them and are far less likely to represent winners in the future.

Clearly, then, we’re looking for an A/E above 1. But we need also to be apprehensive around ostensibly exciting profit figures when the A/E doesn’t back that up. That is, when the A/E figure is below 1.

Note also that very small data samples can produce misleading A/E figures.

Percentage of Rivals Beaten (PRB)

One of the main problems with assessing horseracing statistics is that we’re often faced with very small amounts of information from which to try to form a conclusion.

For this reason, I personally prefer place percentages to win percentages, as there are more place positions in a small group of races than there are winners. Thus, it tends to lead to slightly more representative findings.

PRB tries to take this race hierarchy a step further and produce a sliding scale of performance for every runner in a race based on where they finished.

So, for example, in a twelve-horse race, the winner beats 100% of its rivals, and the last placed horse beats 0% of its rivals. But what about those finishing between first and last?

The calculation is:

(runners - position) / (runners - 1)

The 4^th placed horse's PRB in a 12-runner race would be calculated as:

(12 – 4) / (12 – 1)

= 8 / 11

= 0.73 (or 73%)

The full table of PRB’s for a 12-horse race is below.

A word on non-completions

There are different interpretations of how to cater for a horse which fails to complete (refused to race, unseated rider, fell, pulled up, etc).

Some exclude those runners from the calculation sample, others use a 50% of rivals beaten figure. The traditional way of dealing with non-completions - the way its creator, Simon Rowlands, has managed them since introducing %RB around 15 years ago - is to recode pulled ups as joint-last (so will be >0% if more than one), and fell etc as neutral (50% of rivals beaten).

Whilst I can see the rationale behind both of those, the approach we have taken is more literal: we assume a non-completing horse to have beaten 0% of its rivals. This is unfair on the leader who falls at the last but nor does it upgrade a tiring faller or a horse pulling up at the back of the field.

There is not really a perfect way to represent non-completions in PRB terms; this is at least a consistent interpretation which is of little consequence in larger datasets or where non-completions are rare (for example, in flat races).

What to look for with PRB

PRB is helpful when attempting to establish the merit of unplaced runs; for example, a horse finishing 5^th of 24 in a big field handicap has fared a good bit better than a horse finishing 5^th of 6.

A PRB figure of 55% or more can be considered a positive; by the same token, a PRB figure below 45% should be taken as a negative, all other things being equal.

The problem with PRB is that it assumes, as per the rules of racing, that every horse is ridden out to achieve its best possible placing. In reality that frequently fails to happen: horses whose chances have gone are eased off and allowed to come home in their own time.

Thus, the further from the winner you get, the less reliable is the PRB figure.

PRB²

As the name suggests, this is the PRB figure, expressed as a decimal, times itself. This is also sometimes written as PRB^2, which means the same as PRB².

So, for example, if the percentage of rivals beaten was 80%, or 0.8, the PRB² figure would be 0.8 x 0.8 = 0.64

The reason this is useful is that it rewards those finishing nearer to first exponentially, as the table and chart for an 11-runner race below illustrates.

The chart lines start and end in the same place but, in between, they are divergent.

The difference in the values is greater the further down the top half of the field a horse finishes, and then gravitates back towards the PRB line in the latter half of the field (where PRB² scores are lowest).

This is significant when looking at, for example, trainer statistics. Let’s take an example where two trainers have the following finishes from three horses, all in eleven-runner races (for ease of calculation):

Using our reference table above for eleven-runner races, we could calculate the PRB’s, using decimals rather than fractions, as follows:

Trainer A: 1.0 + 0.5 + 0.0 = 1.5

Trainer B: 0.5 + 0.5 + 0.5 = 1.5

Both have a score of 1.5 which, when divided by the three runs, gives a PRB rating of 0.5.

But Trainer A had a winner and Trainer B failed to secure a finish better than 6^th, so should we afford them the same merit?

Some will argue yes, but I prefer – and PRB^2 offers – to recognise all that has happened but to reward the trainer with the ‘meaningful’ placing to a greater degree than her perma-midfield counterpart.

Here’s how PRB^2 views the same trio of performances:

Trainer A: 1.0 + 0.25 + 0.00 = 1.25 / 3 = 0.42

Trainer B: 0.25 + 0.25 + 0.25 = 0.75 / 3 = 0.25

This time we see the preference towards Trainer A, who had the same average finishing position but the more worthy finish in that one of his runners won.

That, in my view, is a more meaningful statistic for all that it is not straightforward to know what a ‘good’ PRB^2 figure is.

What to look for with PRB^2

Anything above 0.4 on a reasonable sample size implies ‘good’ performance whereas anything below 0.3 on a reasonable sample implies ‘poor’ performance, though there is some scope for different interpretations between 0.3 and 0.4.

PRB3

PRB3, not to be confused with PRB^2, is used in the same way as IV3 when there is a logical and linear relationship between a data point and its closest neighbours. The example we used in IV3 was stall position and that holds equally for PRB3: it would be the average percentage of rivals beaten of a stall and its closest neighbours. Another example might be the rolling monthly percentage of rivals beaten for a trainer, although this will always be historical in its outlook (we cannot know next month's PRB).

As with IV3, its primary utility is one of smoothing the curve to make patterns in the data easier to spot.

Horse Racing Metrics Summary

Throughout the site, figures relating to Impact Value, Actual vs Expected, and Percentage of Rivals Beaten are referenced. There is nothing to be afraid of; rather, each metric simply provides an appropriate way of easily understanding the data (and, crucially, its utility), and comparing it within the context of the entity under investigation.

Gold Upgrades: May 2023

25 May 2023/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

It's been a while since we've released any new features to Gold, so I'm excited to share a couple of small - but awesome - upgrades with you. The first may be imperceptible (but probably won't be), the second is niche, and the third is a new rating... let's get to it.

1 Remember settings

You know how you always do the same thing when you go to the cards to see the info that's right for you? For me, this means setting the draw tab to 'Actual' and PRB/PRB3/PRB in the three dropdowns; and setting the Pace tab heat map to 'Place%'; and incessantly re-sorting the Instant Expert grid when I change one of the top of the page variables.

Well, no more... yay!

Why?

Because we've introduced some 'local memory' improvements so that things remain as you set them up if you're using the same machine as the one you set them up on. Here's what I mean: let's say you, like me, look at PRB views on the 'Draw' tab.

Up until now, if you were logged out or otherwise left the site, you'd have to re-select the parameters as you like them. But, from now, you'll see those settings remembered and appearing as a matter of course when you go to the tab.

Likewise, that minor irritation of having to re-sort the Instant Expert table after every variable amend:

Not any more. Now you can change anything at the top of the Instant Expert and the grid will remain in the order you had it. Phew, what a relief!

We've added 'remember' functions to all tabs, which will certainly save me a goodly few clicks most days, and hopefully ease your transit around the cards a little, too.

N.B. This is the case if you're using the same machine/browser and don't clear your cache/cookies. Even if you do clear cache from time to time, like me, it's still only an occasional faff to redefine your parameters. And that will remind you of when you had to do it every time and make you grateful 😉

2 Added PRB to Pace 'Heat Map' view and table

I've wanted this for a loooong time. Trying to infer the impact of the combination of draw and run style can be really difficult when the sample sizes are small, which is why I use 'Place %' rather than 'Win %' - because place percent gives meaning to each of the placed horses, whereas win percent only does that for the first one home.

Let's say we only have four ten-horse races in the sample size. That would mean 12 placed horses and just four winning horses. Obviously, then, the heat map could be skewed especially when looking at win percent.

PRB - Percentage of Rivals Beaten - assigns value to every runner in every race, aside from the last horse home whose value is 0 (0% of rivals beaten). In our fictional four race sample above, we now have 40 (or 36 if you exclude Tail End Charlie's) scored horses from which to form some sort of perspective.

Anyway, PRB is better in my view, especially when looking at almost any flat turf race (the all-weather sample sizes tend to be much larger and, therefore, more meaningful in traditional win/place strike rate terms).

Note the PRB option in the dropdown box top right and, directly below that, a new PRB column in the pace table. This will immediately be my 'de facto' setting; and, of course, in this brave new world that setting will be remembered!

3 Introducing Performance Ratings

Everyone loves a rating, right? Here at geegeez we have a fair number these days, what with official ratings, Racing Post Ratings, Topspeed, Peter May's private ratings, and our sectional upgrade figures - and that's assuming you don't use our ratings tool to create and save your own 'R1' numbers! But these Performance Ratings are pretty nifty and I think they'll add value for some users at least. What are they? I'm glad you asked...

These are the BHA's own race figures. They differ from Official Rating (OR) in that they are a measure of the level to which a horse has been judged to have run in a specific race. So, for a horse going up the handicap, OR and the Performance Rating (PR) will be the same.

Here's Sayifyouwill, a last time out winner off 79 - she was raised 3lb to 82 for that, as a result of her PR for the last day win being adjudged to have been 82:

But when a horse runs below its handicap mark, it won't necessarily be dropped to its performance level. For instance, Cry Havoc won a couple of times before running poorly on her most recent two starts:

Looking from bottom to top, we see she was 1st of 14 and, running off 79, was raised to 82. She then won in a field of 11 off her revised OR of 82 and was awarded a new PR of 86. On her penultimate start, when sixth of eight, she ran to a PR of just 69 and yet her handicap mark (OR) remained unchanged on 86. Another poor run, this time eliciting a PR of just 34 when running off 86 and trailing home last of six followed. After that, she was dropped a pound to an OR of 85 (not shown).

So what do these PR figures tell us? Well, they quantify the degree to which a horse may have underperformed and, when looking at the last few runs of a horse, they can help us build a profile of progression / regression in a similar way to RPR.

PR figures will start building as a history from now - we don't have the historic data, unfortunately - and there are some caveats, as follows:

- We will never publish a PR figure where there is no published OR (in other words, before a horse receives a handicap mark)

- We will never publish a PR figure for runs prior to the awarding of a handicap mark

- We will never publish a PR figure for Irish racing

The reason in each case is the same: we don't have access to them! Such data are a closely guarded secret at BHA towers and we are permitted to publish only what is available on the BHA website itself.

NOTE: You need to 'turn on' the PR ratings from the Racecard Options area of your My Geegeez page:

Over time, I feel that these PR figures will be a useful guide to horses' form profile and may also help to shed some light on optimal conditions for more exposed runners.

For more information on Performance Ratings, check out this article on the BHA's website.

That's all for now - I hope there's something of use to you in the above.

Good luck,

Matt

p.s. the SBC Awards votes close tomorrow at 9pm. geegeez.co.uk has been voted in two categories - Best Betting Website and Best Betting Data Resource - and your vote counts! There are also prizes to be won in a draw, as follows:

A Brand New Asus 14″ Laptop Computer
A Brand New Lenovo Smart Tablet
An annual membership to the Smart Betting Club

To vote, click here. It'll take about 120 seconds. And thanks a million!

Membership Login