Tag Archive for: geegeez Gold

Combining Pace and SR Ratings, Part 2

14 July 2026/in Dave Renham/by Dave Renham

A four-week trial of combining Pace with SR Ratings, Part 2

This the second in a double header where I am sharing a very simple ‘system’ which I tested over a four-week period from May 1^st 2026 to May 28^th 2026, writes Dave Renham. Before reading on, it would make sense to read the first piece if not done already and that can be found here.

Not everyone is a fan of systems, but my racing journey was enriched by reading books and articles by the racing systems ‘king’, Nick Mordin. Hence, although I don’t tend to use rigid systems per se when selecting my bets, there are systematic elements that I do lean on within the whole process.

It should be noted that in order to find these system selections, a Geegeez Gold membership is required as that includes access to some of the key data needed.

To recap here are the ‘system’ rules:

UK Handicap races only
Race Distance 5f to 1m1f
The Percentage of Rivals figure (PRB) for the ‘led’ group on the pace tab for the specific race conditions must be 0.60 or higher
Horse must have one of the top three pace scores for the race
Horse must have one of the top three SR ratings for the race

Note that if there were two joint third rated pace horses then both would potentially become a qualifier if either or both of their SR rating positions were also in the top three. If there were three or more joint thirds, then I would count the two with the highest most recent race pace scores. In terms of SR rating joint thirds, I need to include them all as I have no way of splitting them. Also, if a horse had only three previous pace race scores, I would use their pace average figure to compare with the four race averages that most horses have. In this case, the horses with the three highest averages across either three or four past races would count. I did not consider horses with one or two past scores.

I mentioned in the first article that I am only using handicap races, as they give us the most reliable results for this type of idea.

In that first part I shared the results for weeks one and two. Again, just to recap, these are below with profit/losses based on BSP (Betfair SP) less 2% commission.

Week 1

Week 2

First two weeks combined

So, all in all, the first fortnight delivered some positive early findings. Of course, the sample size at this juncture was still relatively small. However, a further positive after these 49 races was that there had been 11 second places as well.

Time to move into week 3.

Week 3

The results for the third week were as follows, with winners highlighted in red. There were four qualifiers where the PRB ‘led’ sample size was small and those horses have an asterisk next to their name:

The last two days of the third week (20^th and 21^st) saw an excellent run of results with six wins from just 13 qualifiers. Previous to that there had been four winners from 29 showing a small loss of £3.88. The overall figures for the week ended up like this:

We saw the best weekly strike rate to date, but there were less near misses with only three second places (five thirds). Hence, there could be a valid argument that this week was a tad lucky, with so many winners. However, as we know it is better to be lucky than good. Although even better to be lucky and good!

I was three weeks in now and the overall results read as follows:

All three weeks were consistent with win rates around the 20% mark and solid profits each time. Will we see a similar pattern in week four? Well, let’s find out.

Week 4

The results for the final week are shown in the table below with winners once more highlighted in red:

The final week of the four-week experiment went with a bang rather than a whimper with ten winners from 29 runners. There were five seconds, too, so over half of the qualifiers finished in the first two. The profit and loss figures are shown below:

A huge profit and returns across the seven days of over 242 pence in the £. Yes, this bottom line was helped by a BSP winner priced 38.53 in Partisan Hero, but even if we take that winner out, after commission, profits still stood at +33.56.

With the four weeks up, I can now share the final totals:

This has been a pretty amazing run of results across the 28 days to create such figures. Of course, we cannot get too carried away, as 120 bets across four weeks is essentially a snapshot. However, it is a snapshot that makes me want to expand my research to other weeks and months and see what they bring.

So far, I have shared just the results without doing any additional digging into the underlying numbers; I'm sure there are some extras for us starting with the average PRB figures for the system qualifiers week by week. The graph below shows the splits:

When analysing PRBs we need to appreciate that the average figure for all runners in all races is 0.50. This is the benchmark to judge against. Anything above 0.55 is usually seen to be positive and once we get to 0.60 and above then we have a strong positive. Therefore, for three of the four weeks to have average figures for all qualifying horses of over 0.65 is exceptional. Even week 3’s figure of 0.58 is very decent. These findings give me much more confidence in the small four-week sample I have looked at.

PRBs are a useful metric because in many respects they expand sample size by considering all runners in each race. For the record the overall average PRB for qualifiers across all four weeks was a very strong 0.65.

I want to focus now on the 30 winners and see which run style they employed in the individual races where they came home in front. Being in the top three of the racecard pace figures for their specific races means I would have expected a led or prominent style in the race to be more likely than a mid-division or hold up one. The graph below shows the splits in terms of percentage of runners within each run style group:

As predicted (and hoped), we see that most of the winning qualifiers had a led or prominent style; in fact, five from every six winners did. This helps cement further how effective the four racehorse pace figures in the Geegeez Racecard are in terms of helping to predict run style.

Finally, I wanted to look at the results based on their actual ranking within the SR rank and pace ranks. Hence top rated in both would count as a 1,1 in the table below.

Of course, as we have nine combinations the sample sizes for each are very small. However, what was interesting was when I split the results up comparing those where at least one of the two rankings was top (e.g. 1/1, 1/2, 1/3, 2/1 and 3/1) compared with those that weren’t. Here are the splits:

Those with at least one top ranked position did much better. Again, the important caveat is that the samples are quite modest. However, there is a big difference between the PRBs for each group with the 'top ranked' group averaging 0.67 and the other at 0.59.

That's enough for this two-parter but I hope, if nothing else, the performance of this Geegeez Gold system, albeit over just four weeks, has provided food for thought. Personally, I will be doing more digging into this system, that is guaranteed!

Until next time...

- DR

More on Combining Pace and SR Ratings

7 July 2026/in Dave Renham/by Dave Renham

A four-week trial of combining Pace with SR Ratings, Part 1

Back in May of 2025 I wrote a couple of articles looking at 5-furlong handicaps with the focus being the scores for each horse found in Geegeez Racecard pace tab, writes Dave Renham. In the second of those pieces, which can be read here, I looked at a set of results based on the top three horses ranked in terms of pace scores in their specific race, when they also appeared in the top three of Peter May’s SR ratings for said race.

The findings were extremely promising over the whole year for horses in the top three of both as the table below shows:

Considering these results came from combining all UK courses, some of which do not strongly favour horses that race close to, or on the pace, I think the results were even more impressive.

In this article I will revisit that idea but before getting to the meat and bones, let me briefly discuss the pace tab and SR ratings, especially for new readers.

The pace tab shows the running styles of the horses for a maximum of their last four races. Each past running style is given a score of between four and one. The breakdown is as follows:

4 – Front runner / early leader

3 – Prominent racer

2 – Raced in midfield / mid division

1 – held up near or at the back early

Past number crunching has noted that early leaders / front runners have a strong advantage over other running styles at some of the shorter distances. This front running bias does not occur at every ‘shorter’ course and distance, but at a significant number.

The one tool that should be able to help us is the Geegeez Gold pace tab. Logic dictates that the higher a horses’ pace score total, the more likely it is to lead and, if it leads, then over time such runners are at an advantage offering us punters good value. Let me share an example of a 5f handicap race run on 25^th June of this year focusing first on the pre-race pace ratings:

The first point to note, before we look at the pace totals for each runner, is the performance of early leaders over this course and distance (C&D).

The most important number for me is the PRB figure of 0.64 for early leaders/front runners (the ‘Led’ group). That suggests front runners have had a significant edge over this C&D in the past. Hence, this is the type of race where, if we can predict the front runner and back it, we should make good money over time. I will come back and discuss the C&D PRBs later in the piece.

In the image above I have ordered the runners by their pace scores and we have two in double figures, namely Iris Dancer and Mr Cool. Pre-race it seemed likely that one of these two would lead. Here is how the race panned out:

Iris Dancer did indeed take an early lead and made all the running. Mr Cool was pressing the lead early but eventually finished fifth. The race followed the most likely pattern and the most likely result. Now, of course, there will be plenty of times when a race does not pan out like this, but the odds should generally be in our favour.

Just before I share my new research, I want to briefly talk SR ratings. These are derived from a neural network developed by Peter May. They are much more than a measure of speed; they include a number of form considerations making them a sort of composite of, in Racing Post terms, RPR (Racing Post Rating) and TS (Topspeed) - both of which we also publish in the daily racecards. Crucially, though, the SR figures are not published on every racecard in Christendom, and so fly under the radar considerably more.

Now, going back to that previous race, let me change from ordering the horses by pace rank to SR Ratings:

Here we see Tarlac top rated on 61, Penny Mountain second on 58 and Iris Dancer third on 56. Hence, based on my previous research article looking at horses being in the top three of both the pace rankings and the SR rankings, the winner Iris Dancer, was the only horse to appear in both top threes.

So, onto today's new research. I thought it would be interesting to come up with a very simple ‘system’ similar to what I did in that original article and test it over a recent four-week period. I chose the first four weeks of May of this year, 2026, to do the testing.

Here are the ‘system’ rules:

UK Handicap races only
Race Distance 5f to 1m1f
The PRB figure for the ‘led’ group on the pace tab for the specific race conditions must be 0.60 or higher
Horse must have one of the top three pace scores for the race
Horse must have one of the top three SR ratings for the race

It should be noted that if there were two joint third-rated pace horses then both would potentially become a qualifier if either or both of their SR rating positions were also in the top three. If there were three or more joint-thirds, then I would count the two with the highest most recent race pace scores. In terms of SR rating joint-thirds, I need to include them all as I have no way of splitting them. Also, if a horse had only three previous pace race scores, I would use their pace average figure to compare with the four race averages that most horses have. In this case, the horses with the three highest averages across either three or four past races would count. I did not consider horses with one or two past scores.

As before I have stuck with handicap races as they give us the most reliable results for this type of idea, but I have expanded the distance consideration from 5f only to races of 5f up to and including 1m 1f. The other main difference is that I am only considering C&Ds where the past data indicates a significant front running bias (PRB of 0.60 or more). For the record, I have excluded any rare three-runner races as all three horses would always qualify in those races under these rules.

OK, let me share my findings. For the rest of this article, I will look at results across the first two weeks, and next week I will share weeks three and four, as well as some thoughts on the four-week trial as a whole.

Week 1

There were 29 qualifying horses in week one and the table below gives each individual result. In the table I have added eight columns – date, course and horse being the first three, then the fourth column gives the PRB figure for ‘Led’ runners under the race conditions for that C&D race (i.e. based on the going and number of runners grouping). Column five gives the finishing position of the horse, then their Betfair SP, SR rank of the horse, and its pace rank.

The results for this first week were as follows with winners highlighted in red:

For the record, I have asterisked (*) Nakatomi simply because the sample size for the pace ‘led’ PRB figure came from a very small sample. Most of the PRB ‘Led’ samples came from a very decent number of horses which adds confidence to the overall findings.

Here is the summary for week 1:

A very positive start, although with the caveat that there were only 29 qualifiers. Having said that, there were six horses that finished second and a further four that finished third. Therefore, over half of the qualifiers finished in the first three.

Week 2

After a great start could the momentum be maintained in week two? The first race in week two where we had qualifiers was the 3.40 at Chester on 8^th May. Note that the ‘Led’ PRB for these races conditions (6f, Good going; runners 11 to 14) stood at 0.67 so it comfortably matched system rule three.

I will share both top-rated runners using the SR ratings and then show top rated pace runners to help clarify the selection process. Ordered by SR ratings first:

We had three joint third SR ratings, which actually is extremely rare, but it did mean I had five potential horses that could become a qualifier. Onto the pace rankings now:

A clean top three here, with Rosenpur and Miraculous appearing in both lists, hence they were both qualifiers. The race result was thus:

I haven’t given all the finishing positions as we were only concerned with Rosenpur (who won) and Miraculous who finished fifth. An excellent winning BSP for Rosenpur of 9.54 compared with the ISP of 6.5 (11/2). Also, we should note that Rosenpur led early and made all the running.

Week two had started on the best possible note and below is a table with all the results from that week including those two.

A quieter week with 20 qualifiers and the weekly summary was as follows:

Four winners from 20 this time and, as with Rosenpur, the other three were good prices (BSP 7.5, 7.27, 12.74), hence the excellent profit figure and ROI%. There were also five more horses that finished second.

After two weeks therefore the overall figures read:

We see a very impressive set of numbers here, but I do appreciate it's from only 49 qualifiers. Having said that, the winners do not seem to have come out of the blue because along with the ten winners, there were 11 second places, too.

This incredibly simple system, for which Geegeez Gold members are able to work out qualifiers on a daily basis, has started very positively. Will weeks three and four continue this trend? All will be revealed next time... until then...

- DR

Query Tool v2 is Finally Here (Very nearly!)

7 July 2026/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

It's been a loooong time coming, but the upgraded Query Tool is very, very nearly ready. In fact, it is ready and we'll be deploying it to the site this week, probably tomorrow (Wednesday). In this post - video and, underneath, text/images - I'll show you what's new and, I hope, improved.

One very important point to note, which I'll return to in what follows, is that QTv2 is being released in 'beta'. This means we've tested as far as we can, but the multitude of different browsers and devices, as well as the almost infinite combinations of variables, means there may well be some residual unsquashed bugs. So, if you spot anything, please do let us know. Thanks.

OK, video explainer below. Text and image explainer underneath that. Updated user guide here (see page 138 onwards).

So, at long last, welcome to Query Tool v2. What's new and what's the same? At this point, the interface is largely the same and all of the features you're accustomed to are as they were, including the charts and QT Angles functionality. We do plan to update the interface - the way you interact with the data - later in the year because it's a) a little dated, and b) a little limited, particularly in terms of visualisation.

As for what's new, as you can see in the image below, there's a splash of colour and a couple (three actually, one hidden in that snap) of new blocks.

The colour has been used to differentiate between three sets of variables which enable you to quiz our database on each of the current run and now also the previous run and penultimate run of a horse. Note the (BETA) labels for the last run/2nd last run blocks.

We've used a nice pastel yellow for CURRENT RUN. This houses the parameters you know and love from the original QT. To these we've added a light blue block for parameters relating a horse's last time out run, and violet shading for its second last run.

We've also removed the BOLD CAPS from from the variables to further differentiate and to be slightly less offensive on the ol' peepers.

Clicking on CURRENT RUN reveals the third new block, CAREER STATS. This is also in beta, so do notify of any quirks.

In this block, you can filter the database by runs and/or wins, and by a number of subsets thereof.

For example, if you wanted to know which trainers have performed well with horses making their first start in a flat handicap over the past two years, choose 'Runs (Handicap)' 0 to 0 and group by trainer. N.B. don't forget to select 'Handicap' under the RACE parameters, otherwise you'll get all of the classy non-handicappers that have never (and may never!) run in a handicap.

In the breakdown data here, I've sorted by PRB and displayed only those trainers with 25+ qualifying runners in the time frame. There are plenty of familiar names in there.

Let's move on. I'll click the RESET button and also remember to clear my '25 runs' stipulation (if you don't see the expected data in the main part of the window, always check you haven't got a filter of some sort in the top area).

This time, I want to see if a runner's pace score in its last (couple of) race(s) is predictive in five furlong UK turf races.

It looks like it is. As per the below, those with a score of 3 (prominent) or 4 (led) last time out won at better than 13%, while those racing in the middle or back of the field early last time won at less than 11%. This was replicated in the each way percentages as well.

What about second last run pace score?

A similar pattern emerges, though the dichotomy is not quite as stark. In fact, it's fair to say that the strongest stat might be avoiding those horses that scored 1 (held up) in both runs. Let's query that to see if it's correct. Here, I've expanded the time window to five years:

Breaking the output down by year shows that it's been a bumpy ride, no doubt predicated on some big-priced winners:

Taking it all the way back to 2009 shows the same difficult route to Betfair SP profit:

But figures have been largely consistent across all of win%, place% and PRB.

The inference is that, generally speaking, in five furlong turf races horses that were waited with in their most recent two starts are best avoided. Meanwhile, those that raced at least midfield in their most recent two starts would likely form the basis of solid wagering going forwards, regardless of whether the race is a handicap or not:

These are just a couple of examples of the sort of research that can be undertaken with the extended Query Tool.

[Note that the pace data in our development database is not updated as much as the live one that you use, and so if trying to replicate the above you should expect slightly different results, and apologies for any confusion that might cause!]

Please also note - and I make no apology for labouring this - that, in case you didn't know ;-), we're in beta mode here so stuff might by wonky. Please shout if you see anything more weird than wonderful and we'll investigate, and fix as necessary.

Good luck!

Matt

Instant Expert for 5f Handicaps: An Analysis

18 May 2026/in Dave Renham/by Dave Renham

An investigation into the Geegeez Instant Expert, Part 2

This is the second of a two-parter in which I take a look at the Instant Expert feature in more detail, writes Dave Renham. Last week, I shared some initial findings connected with 575 British 5f handicap races run in 2024. It would make sense to read that piece first here, if you have already not done so. This week I continue my digging and. as I did last time, sharing my findings as I dig! So, at this juncture, I am unsure whether the findings this week will be as interesting or indeed as positive as that first half.

Recap

To recap quickly, the 5f handicaps I am looking at exclude 2yo handicaps (known as nurseries), but include all other age classifications. Any profit/loss quoted will be calculated to Betfair SP (BSP) less 2% commission on any winners.

As I mentioned last week, the Instant Expert tab can be found on the top of the Geegeez Racecards, between the Profiler and Pace tabs.

Instant Expert provides the Geegeez Gold community with some useful, and quickly digestible, horse information. It has a unique ability to summarise the form profile of every runner in the race into this single view. Instant Expert covers the form of each horse in terms of wins (or places), runs, and win (or place) percentage for each of five key areas namely going, class, course, distance, and field size. This can be seen in the screenshot below where I have clicked the tab for the 2.52 at Beverley run on the 23^rd April of this year:

The parameters I am using in terms of generating the percentages for each horse in each category are:

2-year placed form (see top left above the grid), all races and all codes (see top right above the grid).

These parameters are the same as I used in last week’s article because clearly the data across both articles needs to be consistent. The display is colour coded to help us see things more clearly at a glance: green for a higher percentage rate, amber for a middling percentage, and red for a low percentage. Horses with no form under a certain condition have grey figures.

In the above example I have ranked the runners by their scores which are based on The Shortlist scoring system (The ‘Sh’ column). Green percentages score three points, amber percentages score one, grey percentages zero and red percentages minus one. Hence across the five areas a horse can obtain a top score 15 (five greens), whereas the lowest score would be -5 (five reds). Users may change the parameters and dropdowns so, for example, if you prefer to look at 5-year win only form just click the relevant circles. Once this is done, the Shortlist scores will change.

In the illustrated race above I’m Next had the maximum score of 15, while Novello Lad and Ventura Express were joint second on nine, with Trilby ranked fourth scoring seven points. For the record, the result of the race was as follows:

The top ranked horse I’m Next went onto win this race, priced at 11/8. Trilby, the fourth ranked runner, was second at 9/2 while one of the joint second ranked runners Novello Lad came third at 10/1. Based on what we found out last week in terms of the performance of the rankings, this type of result will occur much more than say three horses ranked near the bottom coming first, second and third.

A line on ranking methodology

Before getting into the meat and bones of this second piece, let me briefly discuss ranking or rating methods for a few lines. The key to a good set of ratings/rankings is not whether the top-rated/ranked runners make a long-term profit. Of course, that would be an added bonus but, essentially, to measure the effectiveness of a rating set we need to look at the win strike rate and the percentages of rivals beaten (PRBs).

The top-rated/ranked runner should have the highest win percentage, the second highest should win next most often, and so on, gradually reducing for the other runners. Ideally there would be a significant difference in strike rate between say the top-rated with the fourth rated, and likewise with the fourth rated and the tenth rated, and so on. This type of finding would ideally be mirrored in the PRB stats.

It was noted in my first article that the PRBs produced the perfect graph when assessing Instant Expert Shortlist score rankings with the PRBs decreasing from highest ranked to lowest. The win strike rates also showed the right type of pattern although the seventh and eight ranked were marginally out of kilter scoring maybe 1 to 1.5% higher than would be expected. Overall, the rankings had the right 'feel' based on all the stats I uncovered.

Instant Expert Rank #1 by Betfair Starting Price

Having set the scene, coupled with some key recaps, let me start to crunch the numbers once more. Last time, I looked into the combined performance of the top two ranked runners in terms of their Instant Expert (IE) Shortlist scores across different areas. To start with here, I want to focus solely on the top ranked runner starting with...

Instant Expert (IE) top ranked runner by Price

I want to see whether the BSP prices of the top ranked horses make any difference to returns. I have split qualifiers into different price band groupings and here are the splits:

We would expect higher win rates for shorter priced runners, but it is interesting that the performance of the bigger priced top ranked runners, those BSP 15.0 or bigger, has been very poor. Overall, these qualifiers have managed just three wins from 111 runs (SR 2.7%) for a loss of £57.07 (ROI -51.4%). From this sample, it is clear that top ranked runners based on their Instant Expert Shortlist scores give the best value if priced under BSP 15.0.

Instant Expert (IE) top ranked runner by Price (4yo+ hcaps only)

Last week it was noted that the top two ranked runners combined performed far better when contesting 4yo+ handicaps compared with other race classifications. Therefore, I want to look at the top-ranked runners in 4yo+ handicaps using the same price band analysis. Here are the splits:

There were no wins at all for those priced BSP 15.0 or bigger – 0 from 37 to be precise. The well fancied runners, those under BSP 6.0, produced very solid looking results albeit from a modest sample size of 98 qualifiers. The overall results for horses priced under BSP 15.0 are 43 wins from 200 runs (SR 21.5%) for a profit of £41.83 (ROI +20.9%).

Instant Expert top ranked runner by Age (4yo+ hcaps only)

Sticking with 4yo+ handicaps for a minute, I want to see if the age of the horse has made any difference when it comes to the top ranked runner.

Based on these findings it does seem that once we get to horses aged seven or older, performance of the top ranked runner tails off notably. This older age group produced just two wins from 41 starts with losses of close to 81p in the £. Yes, it is small sample but comparing the PRBs, five- and six-yrear-olds combined had a PRB of 0.64, while those aged seven and up were significantly lower on 0.54.

Of course, finding value selections is the key to successful betting and looking for positive angles helps us in this regard. However, it is also important to try and find negative angles in order that we can discard (or at least downgrade) certain runners from our shortlisting process. If we can narrow down the field it will implicitly improve our chances of finding value selections. It needs to be said that we can never be 100% confident that discarded selections won’t win, because a handful always will; but if these runners represent very poor value, then in general they are worth discarding. We know we can't back every winner!

In the first article we saw that those ranked ninth and tenth produced significant losses of nearly 46p and 30p in the £ respectively. Combining their two records they delivered just 33 wins from 754 runners (SR 4.4%) for losses of £272.75 (ROI -36.2%).

Instant Expert rank of 9 or lower by Class

Now I appreciate that these runners ranked ninth or lower will only be relevant in races of at least nine runners, but I still want to share a few stats I have found for this group.

A look at these lower ranked runners by Class of Race. Here are the splits:

As expected, we see losses across the board. However, there is no clear pattern in terms of whether lower ranked runners have struggled more in higher or lower classes. The Class 5 returns are somewhat out of kilter, but a BSP 50.0 winner is responsible for making the ROI% lower than perhaps what it should be. What the stats from different classes do tell us is that these runners struggle when racing in all class levels.

IE rank of 9 or lower by Race Classification

I am now going to split the races by 3yo only, 3yo+ and 4yo+ races to see what effect this has had on these lower ranked horses:

We see a lower strike rate in both the 3yo only and the 3yo+ group. These two also endured substantial losses to BSP. The 4yo+ group produced the best strike rate and almost broke even. However, all is not what it seems, as two of the 4yo+ winners were priced at BSP 50.0. Taking those two out and losses for the remaining 258 qualifiers would have been substantial, equating to around 38 pence in the £.

Digging a bit deeper, here are the numbers when we restrict all nine and lower ranked runners to those that were priced under BSP 15.0.

So even as we move towards the more fancied end of the betting market losses remained steep, although the 3yo only data set is small.

One final stat I wish to share before moving on is the performance of these lower ranked runners (9+) when aged three. If we look at all 3yo qualifiers that had a BSP of 10.0 or bigger, just one of these runners won from 192 qualifiers. Losses were £166.50 (ROI -86.7%).

Individual Instant Expert Shortlist scores by PRB

Having looked at a plethora of Instant Expert Shortlist ranking stats across both articles, let's now consider the performance of individual scores. As noted earlier, these scores range from the highest, 15, to the lowest, -5. I am going to examine PRBs first and, in order for them to fit within the graph, I have combined next door positions. Hence the highest two possible scores have been combined (15 and 13), followed by the next two (12 and 11), all the way down to the two lowest scores of -4 and -5. Remember, it's not possible to achieve a Shortlist score of 14.

We see a very similar graph to the one I published in the first article which I referenced earlier – the one that examined the PRBs for different ranked runners. The higher ranked runners had higher PRBs, and we see the same pattern here. The very lowest scores have commensurately low PRBs so, on this evidence at least, horses with a score of minus two or less can generally be safely ignored. Even those scoring 0 or -1 have a note of caution about them with a PRB of just 0.45. For the record, the top score of 15 produced the highest PRB, at 0.62, with a score 13 achieving a PRB of 0.58, while 11 and 12 both scored 0.56, and 9 and 10 both hit 0.55: that's very pleasing linearity indeed.

Instant Expert individual scores

Let me now share the strike rates, profit/loss, and ROI percentages for each individual Instant Expert (IE) Shortlist Score.

Some of the sample sizes are relatively small such as for the scores of 10 or 12, which is one of three reasons why we cannot necessarily be seduced by bottom lines for these individual scores. The second reason is that some individual scores will have been skewed somewhat by a big priced winner or two. One such example is a BSP 80.0 winner for the -5 group. If we weed that winner out and examine the other 237 runners which scored -5, the ROI% drops to -35% and correlating far better with all the other horses whose Shortlist score was a negative value. The third reason is that a score of 1 or 2, or indeed even smaller, could actually be the highest Shortlist score in the race. One such example of this can be found from a Yarmouth race in July 2024 which is shown below:

Merrimack was top ranked with a Shortlist score of just 2. For the record, he went on to win the race.

When looking at the strike rates for individual Shortlist scores there was not perfect correlation in terms of the strike rates always dropping as the scores decreased. However, when looking more generally the right strike rate pattern emerged and this can be seen to best effect when we group the highest IE scores together (11, 12, 13 and 15) and compare them to the lowest (-1, -2, -3, -4, -5).

Grouping like this does help to create more robust sample sizes and also smooths the data. This is further evidence of the potential effectiveness of Instant Expert, although I appreciate I have only looked at 575 UK handicap races, all run over 5f and from a single year, 2024. However, for this sample the correlation between ranking results and individual scores is primarily positive and makes me more hopeful that other result sets will produce similar results.

Instant Expert individual scores (4yo+ hcaps only)

It makes sense next to look at the 4yo+ handicap only data as these races to date have shown the most positive findings. Due to modest sample sizes, I have grouped the individual Instant Expert (IE) Shortlist scores into bands:

More positive correlation with both win rates and returns. This is replicated once more when we compare the PRBs:

For the record, horses scoring between 11 and 15 in 3yo only and 3yo+ handicaps had a lower PRB of 0.56. As with my first Instant Expert research offering, it seems 4yo+ handicaps see the Instant Expert at its most effective, when there is generally more data available from which to populate the scores and colour codings that the view thrives on.

Summary

In this article horses with negative IE Shortlist scores have performed very poorly across the board and are horses I believe we should be ignoring nine times out of ten. This type of performance is similar to what we noted with horses ranked ninth or lower in the first article. The same pattern is occurring – the higher ranked runners have totally outperformed lower ranked runners; higher individual Shortlist scores have outperformed lower scores.

Clearly, these two articles have only scratched the surface as far as the Geegeez Instant Expert is concerned. However, the early findings have shown that this tool has real potential to help pinpoint runners which should offer good value and others which likely represent poor value. Making money over the longer term when betting is about finding value. If we can do this regularly enough, we will come out in front. And, crucially, using tools like Instant Expert means we'll have fun in the process!

The Instant Expert tab is something I always look at when analysing races for potential betting opportunities and, I hope via these two articles, I have converted more Gold members to do likewise.

Until next time...

- DR

A First Look Under the Instant Expert Bonnet

10 May 2026/in Dave Renham/by Dave Renham

An initial investigation into Geegeez Instant Expert

I wonder how many people who read my weekly articles here are Geegeez Gold members? If you are, I think you're going to like this one... and if not, maybe it will give a gentle nudge to try things out, writes Dave Renham.

Introduction

One of the many benefits of being a Gold member are the daily racecards. These give far more ‘bang for our buck’ compared with other racing sites. Below is a screenshot of the racecard for a contest at Wolverhampton on Tuesday 5^th May of this year:

As we can see, there is the usual type of information we would see on most racecards such as the draw, recent form figures, name of the horse, age of the horse, weight carried, the trainer, jockey and official rating. However, we also get to have the Racing Post Rating (RPR) for each horse, the Racing Post Topspeed figure (TS), and Peter May’s excellent ratings (SR).

These extras are just the start, as if we cast our eyes to the top of the Racecard we can see some tabs: Full Form, Profiler, Instant Expert, Pace, Draw, Trends and Odds. Clicking on these tabs presents a wealth of further information and data in order to assist with our race analysis. There are also some icons just below these tabs (between the word ‘Horse’ and ‘Age’) with breeding data, jockey and trainer form etc, etc. So, to coin a phrase, these racecards are literally gold!

In this article I want to focus on the Instant Expert tab in a little more detail. This tab is exclusive to Geegeez, and the tab lets us see a variety of key data for horses as shown below using the same race as above:

Instant Expert gives us punters key horse information. It has a unique ability to summarise the form profile of every runner in the race into a single easily digestible view. It covers the form of each horse in terms of wins (or places), runs, and win (or place) percentage for each of five key areas namely going, class, course, distance, and field size. It also compares today’s official rating for each horse with their last winning official rating. The display is colour coded to help see things more clearly at a glance: green for a higher percentage rate, amber for a middling percentage, and red for a low percentage. Horses with no form under a certain condition have grey figures. It should be noted that the past data is taken from UK and Irish racing only so will not include overseas form.

Users have the ability to change parameters easily. In the above example I am using placed form over two years. This is my favoured combination, but at the touch of a button we can change this to one or five years, or even look beyond that for older horses using the ‘All’ tab. We can look at win data only if preferred over any of the four-time frame options. Also, we can expand individual areas if required; so for example when I look at All Weather races, after looking at the stats for the specific going, I will click the going tab, so it includes both standard going and standard to slow to give me extra data. Clearly, we can do the same with all other variables should we wish and I will often do this for longer distances on the distance dropdown. For example, if a National Hunt race is being run over 2 miles 1 furlong, I will expand a little by including past results from any race from 2 miles to 2 miles 2 furlongs.

In addition to drilling into each horse within the five key areas, we are able to look at sires, trainers and jockeys within those areas too should we wish. Sire data is particularly useful for 2yo races for example, when the majority of runners have little or no past form to speak of. Trainer data is something most punters like to look at, and I will always look at the trainer data when analysing any race. This Instant Expert tool is exceptionally useful, and we can crunch so much data within a matter of a few seconds.

In the last few months there has been a new addition to the Instant Expert template, namely a new column with scores based on The Shortlist scoring system. Green percentages score three points, one point for amber, none for grey, and minus one for red. Hence across the five areas horses can obtain a top score of 15 (five greens) and the lowest score is -5 (five reds). The scores update when you change the variables and dropdowns above the main data grid. In the screenshot I shared above, the Shortlist column has the heading ‘Sh’ and is highlighted in the green box.

In this race, we can see Kento had the highest Shortlist score of 11, whereas Tomarlo had the lowest figure of -3.

Having set the scene, I am now going to share some research I have undertaken connected with these Shortlist scores. I have used a data set that I used for run style / pace articles a year ago, in May 2025. I will be looking at 5f handicap data from the whole of 2024 with the exception of 2yo handicaps, or nurseries, due to limited past horse form.

This sample covers 575 races and over 5000 horses with their individual Instant Expert Shortlist totals. This, therefore, is a reasonable sized sample, and I must say that adding over 5000 Shortlist scores to my Excel sheet was not the quickest process! [We're very grateful! - Ed.]

After all the leg work to input these individual scores, I hoped that I would find something worth sharing! Let’s see…

I am going to write this research up while crunching the numbers so, at this point, I do not know which way this will go. Normally, I do all the research, then crunch the numbers and then write it up. Therefore, it will read more in the present tense than usual.

So, to recap, the highest possible Shortlist score is 15 and the lowest -5 so one area I plan to look into will be the performance of different individual Shortlist scores. I guess the hope is that the higher scores, 12, 13 and 15, (14 is not a score that can be achieved), would certainly win more often and hopefully prove better value, than the lowest scores such as negative values of -1 to -5. However, there is a slight caveat to this because there will be plenty of races that are contested by horses with modest or poor long-term records and hence a Shortlist score of 3 or 4 could in fact be the highest Shortlist score in the race; so I will also need to rank the Shortlist scores in each race. One would hope and indeed expect that the higher ranked positions would win more often than the lower ranked ones and, assuming that is the case, does that lead to a difference in returns?

In terms of these ranked positions, there will be some horses in most races who have the same Shortlist score and therefore they will be treated as ‘joints’. In other words, if we have a five-runner race and the Shortlist scores are 13, 11, 8, 8 and 0 then the horses scoring ‘8’ with be ranked both ranked third, there will be no fourth ranked horse, with the horse with ‘0’ being ranked fifth. Fairly obviously I hope the 13 score would be ranked one or top, with the ‘11’ second.

In this article I will start by looking at the results with no price cap as I just want to examine the raw data first. Obviously, the odd big priced winner could skew the bottom lines, but I will mention it when that happens. I will probably use a price cap at various junctures, I’m just not sure exactly when as yet.

Instant Expert Shortlist (Sh) – Performance by Rank

My starting point is to look at the ranked positions of the Shortlist scores across all the handicap races in my sample. I began by comparing their win percentages / win strike rate. The graph below shows the breakdown:

The hope was that we would see the usual sliding scale from left to right as we normally do when we look at any type of ‘ranking’. Generally, that has been the case here, although those ranked seven and eight are slightly out of kilter. I think because there were a good number of ‘joint’ positions we should expect slightly more overlap than normal. The good news though is when we look at either end of the graph we see the top ranked runners winning the most often and those ninth or worse winning the least. Not only that but the top ranked have by far the best win rate and likewise those ninth or lower have by far the worst.

To build the picture further I want to look at the PRBs (Percentage of Rivals) for each rank. This metric essentially creates much bigger sample sizes as the figures are affected by all the runners in every race. Hence, I am hoping this graph will have the near perfect left to right sliding scale:

0.58 is a strong PRB for the top ranked and we essentially see our ‘normal’ type of graph when analysing the performance of ratings or in this case rankings.

So, how does all this equate then into profit and loss returns? As I tend to do in all my pieces, I will use Betfair SP (BSP) less 2% commission on any winning bets for these calculations. Here are the splits:

As far as I am concerned, it makes excellent reading if focusing solely on either end of the table. The top ranked runners have by far the best win strike rate, while the top two ranked runners combined have seen very small losses overall. Meanwhile those ranked ninth and tenth have produced significant losses of nearly 46p and 30p in the £ respectively. These definitely look to be horses to avoid.

However, we can see that those horses ranked seventh and eighth both made decent overall profits. At first glance this is not ideal as I would have preferred to see any profitable figures at, or near the top of, the table. However, both of their profit figures have been skewed by one big priced winner in each case. A horse called Rainyniteingeorgia won at Lingfield on 22^nd December 2024 at the odds of 127.21 when ranked seventh. Take that winner out, and the remaining 427 horses ranked in seventh place would have made an overall loss. In terms of the eighth ranked winners Big Nut won at Musselburgh in August ‘24 at odds of 80.0 which wipes out a huge chunk of the eighth ranked runners’ profits.

Instant Expert Shortlist (Sh) – top two ranked runners

I want now to dig into the top two ranked runners in a bit more detail as when combined together they have not been that far from breaking even (losses of just under 2.5 pence in the £). With these combined results I am going to see if they have performed better or worse within different areas. To begin with, I am going to look at Class of Race. Here are the splits for each Class level combining the results of the top two ranked runners:

The figures do not fluctuate massively. Lower win rates in Class 2 and 3 races is what I would expect as these races tend not only to be very competitive but have bigger fields. Bigger fields mean lower win percentages. Also, the data sets for both of these class brackets are relatively small. Returns are similar across the different classes, although in Class 4 races the losses were a tad above the rest.

How about if we split the top two ranked runners by the Race Age Classification? There are three main types being 3yo only races, 3yo+ (races for horses 3 years old and older), and 4yo+ (4 years old and older races). There was one 6yo+ race in the sample which is obviously too small to worry about!

Before digging into the stats for each I am hoping to find slightly better results for the more exposed runners, namely the 4yo+ races. I am also expecting that the 3yo+ races would be the least predictable with less exposed 3yos running against more exposed rivals. However, I have been known to be wrong before! Right, let's see how the top two ranked runners have fared across the three different age classification types:

The 3yo only results are interesting; definitely better than I would have expected. The results for the 3yo+ and 4yo+ races seem to match my initial hypothesis which is pleasing. There is a significant difference in the strike rates between 4yo+ and 3yo+ races for these top two ranked runners. Also, the 4yo+ results for these runners see very decent returns of close to 16p in the £, compared with losses of over 17p in the £ in 3yo+ races.

However, before we get too carried away, the 4yo+ figures did have a BSP winner of 55.0 which accounts for a decent chunk of the overall profit figure. Therefore, it makes sense to implement a price cap when comparing these 3yo+ and 4yo+ results to avoid bigger priced runners potentially skewing bottom lines. Thus, for the next dataset I will include only horses that were priced under 15.0 BSP. Let's see if this changes things or not. I will build up the tension by first comparing the win and each way (win & placed) percentages for each:

We see positive correlation with the 4yo+ top two ranked runners having outperformed their 3yo+ counterparts in both the win percentage and the EW percentage. This also correlates so far with what saw before I introduced a price cap. What about the profit/loss returns? The table below shows the splits and does the 4yo+ group come out on top as before?

They have indeed, and comfortably so. Based on this sample, and I appreciate it is just one sample of races, it seems that the top two rated runners from the Instant Expert Shortlist scores should be considered a strong positive in 4yo+ handicaps.

Sticking with 4yo+ handicaps, how have the top two rated performed when they finished first, second or third LTO? Let’s find out:

These figures are very impressive with each one showing a profit. I should add a caveat that each individual LTO position sample is fairly modest. However, to create a bigger sample, if we combine the results to include any horse that finished in the top three LTO we get 50 wins from 195 (SR 25.6%) for a healthy BSP profit of £54.17 (ROI +27.8%). Based on these findings, horses that finished in the first three LTO require close scrutiny if racing next time in a 4yo+ handicap when in the top two of the Instant Expert Shortlist scores.

Shortlist Rank Comparison: 3yo+ Handicaps vs 4yo+ Handicaps

Before finishing I want to share the PRBs for all individual Instant Expert Shortlist ranks for both 4yo+ handicaps and 3yo+ handicaps. We have seen already that the top two ranked runners have performed much better in 4yo+ races. How about using a PRB comparison across the board? The chart below gives us a neat graphical comparison:

This graph perfectly illustrates why, for this sample of races, the ranking of Instant Expert Shortlist scores would have worked better in 4yo+ handicaps compared with their 3yo+ counterparts. The graph highlights the edge to higher ranked runners in 4yo+ handicaps coupled with the fact that the lower ranked runners perform less well when compared to the lower ranked 3yo+ groupings.

It is important to reiterate that what the graph shows is what we should expect from drilling into each horse in terms of their record on the going, in the specific race class, at the course, over the distance, and within a specific field size grouping. Handicappers aged four or older, as a rule, would have raced more often than 3yo handicappers over a full 2-year period. This would be especially true in the first half of the year, as 3yos only have two full years of races once they reach the end of their 3yo year, and that's assuming they started racing early on as juveniles.

More data for each of the five areas should be expected to be more robust and reliable.

When I started this research, I was heading into the dark somewhat. I assumed I would have enough decent material for an article and, as it turns out, I have plenty more to delve into. And I haven’t even started to look at the individual Shortlist scores from 15 to -5 yet. All in good time!.

Based on my findings so far, the Instant Expert tab is one that is not only unique to Geegeez, but it really might have the potential to almost single-handedly improve our bottom line. That certainly seems the case for the data analysed to date.

Next week I will continue my research and share the rest of my findings.

Until then...

- DR

Two New Features in Query Tool

24 March 2026/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

We've added a couple of new features in Query Tool - woohoo!

The first is 'Course Characteristics', which covers the direction of the track, its profile (flat, undulating or very undulating), its general configuration (galloping, sharp or very sharp), and any specific configuration elements (sharp bends, uphill finish).

And the second, which might be more general fun, is DSLR, or 'Days Since Last Run'. This has a few nuances, which I'll quickly explain. Selecting a 'from' of 0 on DSLR (i.e. no days since last run) will bring back data even though it is impossible for a horse to run multiple times in one day.

What's actually happening is that we're using zero to mean 'horse is making its first start in UK or Ireland'. That in turn can mean one of two things: an unraced horse making its debut, or a horse with overseas form making its UK/Ire debut. Our dataset does not extend to overseas form so if, for instance, a French trainer brings a horse to UK for the first time, that runner will show up in the zero DSLR cohort.

Although we don't have point to point form in our database, we do normally have the number of days since the most recent point run, so horses with experience between the flags will generally not show up in the zero DSLR cohort.

I told you it was a bit nuanced!

Anyway, it's a really interesting way of looking at things like trainer performance, especially at this time of year when juveniles are making their debuts and horses are returning from their extended winter breaks.

This short video explains more about the new features and includes a couple of quick examples.

Good luck!

Matt

Gold Updates: February 2026

12 February 2026/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

At the end of a busy week which has already featured a major sectional data enhancement to the Full Form tab, we now unleash a raft of small changes designed to make your life easier. Let's crack on with them...

Enhanced Tracker buttons on the racecard

Historically, there was a cumbersome process whereby you could add a horse (or trainer/jockey/sire) to your tracker via the little star icons on the racecard... but you then needed to go to the Tracker page to add or update any notes. No more. Now, a click on the star brings up a dialogue box enabling direct entry of notes. If you already have an entity tracked, you can remove it via a 'remove' button that displays next to the 'update' one. That's better...

Breeding suffixes on Full Form and Profiler tabs

Adding breeding country suffixes to horse's names is a small but quite useful little tweak. Not much else to say on that one.

Shortlist scores on Instant Expert

Ooh, what's this? We've added a new column to Instant Expert with scores based on The Shortlist scoring system. Three for green, one for amber, none for grey, minus one for red. Top score 15, bottom score -5.

The scores update when you change the variables and dropdowns above the main data grid. Nifty.

'Last 5 Years' option on Draw Tab

Sometimes the draw bias over a particular track and trip changes. Maybe the rail configuration was amended, maybe a different irrigation system was put in place, occasionally they dig up the entire track and re-lay it. These things have a bearing on any bias that previously existed, and our 'last 5 years' button allows you to quickly compare the data from 2009+ with that for the most recent half decade. Simple dimple.

Each way terms added to the Odds Tab

Jeez, it took us a while to get to this. But, finally, we are there. Number of places and the fraction of the win odds paid are now displayed on the odds tab for each bookmaker. Useful.

Report tweaks galore and a bug fix

We've added csv download buttons to most reports and course selector dropdowns on all reports. And we've added odds to most of them, too.

You're welcome!

Hopefully there's something useful there for most users. A gazillion more updates planned for this year, but that's a goodly bunch this week to get you started.

Matt

p.s. As always, you can view the full User Guide here. There's a lot inside Geegeez Gold including things you probably didn't even realise were there! Any questions, let me know below.

A January Random Roundup

15 January 2026/in Horse Racing Blog/by Matt Bisogno

In this video post, I outline what's happening now and next on geegeez.co.uk, and beyond these shores. Specifically:

- Geegeez Feature Upgrades
- Australian Open AI Play
- Tix... PLUS?!
- Racehorse Syndicate Updates

The geegeez feature updates are first up so feel free to skip the rest if it's not of interest to you.

Enjoy!

Matt

p.s. link to the Nirvana du Berlais ex Futura syndicate detail is here >>

QT Updates Coming Today

2 December 2025/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

A short post, with some video content if you'd like it, to let you know that we'll be updating Query Tool today. We're adding some new input variables as well as a new output column.

Details of those can be found in this recent post. Or in the videos below (first one is a general overview including the new variables; second one outlines a couple of important changes to the way QT Angles works).

IMPORTANT: QT may be unavailable for a period this morning (Tuesday 2nd December) while we make the upgrades.

These changes are also reflected in the latest version of the User Guide, which you can view here.

VIDEO #1: Query Tool Overview

VIDEO #2: QT Angles Enhancements

Matt

Geegeez Autumn Update

17 November 2025/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

It's getting darker earlier, and there's a distinct chill in the air this week. The turf flat season is a fading memory and close at hand now is the National Hunt season proper as well as a serious programme of all-weather racing. Both make for excellent punting!

While it might seem like we've been quiet on the development front this year, we've actually added some big (and small) features in 2025.

The TRENDS tab was introduced in January, and has become part of my 'go to' considerations for a race.

In April, we rolled out Betfair data (Betfair SP, Place SP, in running high and low prices) across the site.

And in September we added 'AvOR', the average official rating of each race, to help you compare today's race quality with runners' recent outings.

You'll also have noticed this year that parts of the site have had a 'facelift' - updating them to be easier on the eye as well as instructive to your wagering.

And we're not done yet!

We're currently working on some further Query Tool updates, as well as more costmetic upgrades. Allow me to show you a few things from the development site.

Query Tool

We'd originally decided on a small number of new variables in QT to release as soon as possible... but once we started digging we found a few technical elements that ideally needed improving. That has delayed release unfortunately, but I'm still able to show you a couple of bits here.

New Variables

We've added some new variables for you to interrogate. These are:

Season

You'll soon be able to search by season, across flat, all-weather, and National Hunt seasons in both UK and Ireland. This is great for analysing, for example, trainers' and jockeys' progression (or regression!) and - especially for the non-flat seasons, where things are not aligned to calendar year - looking at overall stats more generally.

Below for example I've selected the trainer Anthony Honeyball, whose yard geegeez.co.uk sponsors and with whom we syndicate a number of horses, and 'grouped by' season (having only selected UK and NH race types - Anthony has also had flat winners and winners in Ireland during this time).

Anthony Honeyball stats by season

Owner

You'll also be able to drill down by owner before long. And, because some ownership entities have, literally, hundreds of ownership names, we've added a 'select all' button within the search facility. Here's an example using the geegeez.co.uk ownership entities. You'll be able to dig in the weeds of J P McManus, the Ballydoyle cartel, and/or anyone else you fancy.

Racing Post Rating and RPR Rank / Topspeed and TS Rank

And we've added Racing Post Rating and Topspeed, as well as RPR/TS Rank within a race (e.g. RPR rank 1 means the top rated horse on Racing Post Ratings). This image shows the top two ranked Topspeed horses' collective performance over the past two years in UK handicap hurdle/chase races. This is bound to include some big-priced winners skewing the data, but it's not a bad starting point for further analysis by any means!

PRB

The eagle-eyed may have noticed that we've added a new column on the right hand side of the results output in Query Tool for PRB (Percentage of Rivals Beaten). This is a very useful metric, especially for smaller sample sizes, where a number from around 0.55 (55% of rivals beaten) is a positive, and anything below (0.45) is somewhat of a negative. It's definitely a number to keep an eye out for when creating your QT Angles.

A New Look for the Reports

As well as QT upgrades, we're also working on our report suite and, specifically, on making it a little more modern in look (the data remains the same excellent content you're already familiar with). They'll look like this:

Timeline for Implementation

I've already got this wrong once, which is annoying/embarrassing, and so I'm loth to make the same mistake again. However, I'd be very hopeful they'll appear on your screens - the report changes at least - by the end of the month.

Query Tool upgrades should be an early Christmas present, expect them online in the first half of December with a following wind.

And... in January, we'll be kicking off the New Year in style. More on that another day.

Thanks as ever for being a geegeez subscriber, it means a huge amount to me - to all of us - and it enables us to continue to invest in your racing site. Thanks again!

Matt

Ad Hoc Almanac Race Preview

9 May 2025/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

It's Friday, there's lot of racing - much of it high quality - and so why not do a race preview, or three?

In the video below, I've tried to showcase a few of the more accessible components of Geegeez Gold and Lite, as well as throwing in a couple of the less well-trodden areas of what we have.

As ever with such videos, the main point is to showcase what's inside Geegeez racecards rather than to pick winners. Of course, I hope to hit one of those, too!

Before watching the video, a quick polite reminder that I'll be taking the Flat Track Almanac link down on Sunday so time is running out. You need to be premium (Gold or Lite) subscriber to see the Almanac download link on your My Geegeez page, and you can take a 30 day trial for £1 here.

Good luck

Matt

How to Play the Early Days of the Flat Season

28 March 2025/in Horse Racing Blog/by Matt Bisogno

The excitement of a new flat turf season is upon us once more, as we look forward to the Brocklesby, the Spring Mile and of course the Lincoln tomorrow at Doncaster. But that buzz can quickly give way to head scratching as we start to ponder which horses that have been absent for most of 200 days might be fit enough on this first spin of the year.

Here are two or three thoughts that might help with sorting the not today's from the ready's in our quest for some extra, erm, readies...

Trainer Form

How can we know if a horse is fit? Some talk about trainer form, either historical at this time of year or recent in the past few days; and it's not unreasonable to think in those terms. But a hitch at home - say a gallop getting washed away, or a problem with a high pollen crop in a nearby field - can upend history and delay a trainer's team for a few weeks.

Meanwhile, recent form cases are often built on the basis of just one or two runners which, while better than zero, is a very flimsy sample size.

Combining the two may be better than relying individually on either; and trying to squeeze a bit more meaning out of small samples by using percentage of rivals beaten (PRB) rather than win, or even place, strike rates seems sensible, too.

Here's a list of trainers who traditionally hit the turf flat season running: this group has 50+ UK flat turf runners in March/April across the past four seasons and an Impact Value of 1.25 or higher, and they're ordered by win strike rate.

Early Season Trainers: those who hit the ground running

It's no surprise to see Charlie Appleby at the top of the pile. We can either disregard Mark Johnston or consider combining son Charlie's form into the mix - personally, I'll ignore that row. William Haggas and Charlie Hills appear to be fast starters and potentially offer a small bit of value; whereas the quick from the blocks efforts of Team Gosden and Sir Michael Stoute are undone for us punters by typically short prices when they've won.

Further down the list may be where the more interesting characters hang out. The likes of Eve Johnson Houghton and Ben Haslam and Jack Channon are worth drilling into a bit further.

This second table is taken from the Trainer Statistics 14 Day report, with 5+ runs and a PRB of more than 0.55.

Trainers in recent good form and with runners at Doncaster

There is no crossover between the tables; no trainer appears in both tables. Karl Burke was just off the 14 day table, with a very good PRB of 0.55; and Julie Camacho was just off the early season four-year form table with an IV of 1.24. Burke has six runners at Donny on opening day, all fancied, and Camacho has just Lattam in the Lincoln, also well enough found in the early market.

You can draw your own inferences from the tables, but a couple of observations from me:

- Keep an eye on the runners from the Haggas yard tomorrow. Are they well supported? Did they run with credit, even if not winning?
- Ditto Charlie Hills. (fwiw - nothing at all - my one token interest in the Lincoln is on his 33/1 rag, Galeron, so I'll be watching keenly enough)
- We know horses from the Clive Cox, Marco Botti and Jennie Candlish yards are running very well at the moment
- A few other names on that list - Kevin Frost, Richard Hughes and your pick from the rest - are worth noting, too.

Here's another way of looking at trainer form...

Slow starters and expensive from an ROI perspective

This table comprises eight trainers who tend to start on the turf slowly. They all have a pretty painful ROI, too, with the possible exception of Phil Kirby. Tim Easterby has actually saddled 26 flat winners on the turf in March or April since 2021, but they've come at the expense of 423 non-winners. And an ROI of -51%!

These are all good trainers, but for differing reasons they tend to fare poorly in the early weeks of the flat turf season. (Note, any one of them could have a magnificent month, caveat emptor, small samples, etc).

Here's the 14 day trainer form table similarly flipped on its head and sorted by percentage of rivals beaten (PRB), lowest at the top:

Recent middling form

Are these trainers to avoid? Probably not, at least not solely on the basis of the table above. But theirs might be horses to have a question mark against unless you really like the rest of the profile. Again, I'm not saying they can't win - duh - but I am suggesting I'll personally be a little less forgiving when trying to frame a case for any of these.

In summary, trainer form is much more nuanced than "Charlie Farley had a winner yesterday so he's in form". Combining longer-term early season performance with recent evidence based on PRB might be a good way to get a better handle on the subject.

Price Movement

A much shorter section here. How can we know if a horse is expected to run well? Look for the blue on the odds grids! This is actually not a terrible strategy in general, but at this time of the year - when punters not connected to stables must take fitness on trust - the markets are a really helpful barometer.

The problem with betting 'blue' horses is that by definition we've already missed the price. Furthermore, it is often the case that such horses drift back out again closer to the off - those subsequent drifts do not stop horses winning!

I religiously check the odds grids at this time of year, especially for less obvious horses which I then try to 'reverse engineer' a form case around.

The Geegeez ODDS tab only starts to show blue (shortening) and pink (drifting) from 9am on race day. We deliberately ignore the cheap moves overnight, before BOG (for those who can get it) comes in and at a time when a very small amount of money can move a horse's price materially. You can still see that price movement on our grids by clicking the little coloured chart icon:

That will open a window displaying either a table or graph (it remembers which one you last looked at), as follows:

Table view of odds movement since first show the night before

Note that we also don't clutter up the table with millions of rows showing tiny odds moves back and forth - who needs or wants that? Instead, we publish a couple of overnight price rows, then a morning (7am) row, and then hourly from 9am, and then every 15 minutes from one hour before race time when prices may fluctuate more meaningfully and frequently.

In my opinion, that's a much better digest of the price movement of a horse or race, and a lot easier to absorb.

Here's the graph view:

Odds chart, configurable to view all or some runners; and best or average price change

There's a bit more going on here with various configuration options. You can vary the start of the time period, choose average odds or best available, and add/remove horses from the view. Hovering over any line on the chart will display the odds for all chart lines (runners) at that point in time.

It's really useful and, if you're not currently using this tab, I'd recommend you start doing so.

What else?

What else should we consider at this time of year perhaps more so than generally? Well, one to think about is the influence of draw and run style. I've written about this ad nauseum, as has Dave Renham. You can check out much of our work by typing 'draw' or 'pace' or 'run style' into the search box on this page.

Specifically for tomorrow's big mile handicaps, the Lincoln and Spring Mile, the draw chart looks like this (16+ runners, between good and soft, handicaps):

Doncaster 1m handicaps, 16+ runners good, good to soft or soft, 2009+

The main blue line represents PRB3 (the average percentage of rivals beaten of a stall and its immediately neighbouring stalls - so, for stall 3, it's the average PRB of stalls 2, 3 and 4). This is a way to flatten out any daft-looking outliers and attempt to make things vaguely meaningful.

50%, or 0.500, is a figure meaning runners from a stall were beaten by as many rivals as they beat; so more than 50% is positive, less than 50% is negative. Positive meaning can start to be implied at around 55% (depending on the size of the sample) and negative at around 45%.

What is noteworthy from this chart, then, is that virtually the entire line exists between 0.45 and 0.55. One might argue that close to either rail is a positive - as it often is at many courses in huge field straight track races - and that right in the middle is no man's land. Having said that, here are the winning stall numbers of the Lincoln and Spring Mile since 2013, in numerical order:

1
1
2
3
3
3
4
4
5
6
8
10
10
12
15
15
15
16
17
20
21
22

You can win from anywhere, but the middle third has had the toughest time of it overall.

At this time of year more generally, keep an eye on any potential changes to draw biases: there can be a small edge before the market fully catches on. For example, when Chester introduced a false rail on the bend into the straight it reduced (though failed to eliminate) the inside draw bias. That was an opportunity to get solid value on wider berthed horses whose win chance was a little underestimated. It still is to a small degree but, like everything in the dog eat dog world of punting, the market corrects soon enough.

Keep an eye out for the next material change.

And finally...

I had hoped to share a major new addition ahead of the start of the flat season, but it's not quite ready. We should have it online next week and, without explicitly stating what it is, here's a sneak preview - you'll be able to figure it out!

Good luck with your flat season play. Obviously, geegeez has you increasingly well covered - check out our brand new whizzy bangy sales letter if you're not yet on board and see if there's anything that can maybe help you (hint: there is!).

Matt

New for 2025: Introducing the Trends Tab

2 January 2025/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

Happy New Year to you! Here's hoping 2025 is another year packed full of thrilling action on the track, and some terrific bets landed. More importantly, I wish you and yours the very best of health for the coming year.

Meanwhile, on geegeez.co.uk, we've got a brand new TRENDS racecard tab, woohoo!

It looks like this:

And on mobile, something like this:

Clicking on the TRENDS tab will reveal up to the last ten renewals of the given race (currently - we're working on adding all renewals going back to 2009 when our database began). Of course, if a race has been run fewer than ten times, you'll see commensurately less data rows in the table.

Let's take a look at the tab in more detail. Here are the trends for a Cheltenham handicap chase from yesterday.

The left hand half of the screen is focused on standard intel, like the winner, trainer, jockey and going. Note the red font for N Henderson: this tells us that Nicky had a runner in the race this year. His runner was Chantry House, which won at 8/1. Convenient 🙂

Clicking on a race date will take you to the result of that race. And all of the columns within the TRENDS tab are sortable, making it easy to see if a specific profile is emerging around any of the variables. For the less obvious column headings, hovering over them will display a fuller explanation - in this case I'm hovering over 'DSLR':

In this example race, we can see that the odds on favourite, Broadway Boy, was younger than most winners (though there were two recent scorers his age). Broadway Boy had been sent off 4/1 favourite last time while most recent winners of this race had been less fancied on that prior spin. Chantry House was third choice in a hurdle race on his last day. The sweet spot on DSLR was four to five weeks (30-35 days) and Chantry House last ran 33 days ago.

The point here is not really to pinpoint winners - after all, looking only at win trends is a narrow field of vision for such a thing - but, rather, to highlight potential red / green flags for horses that you are considering in the round of their overall form profile.

Here's this Saturday's Veterans' Chase Final TRENDS tab:

Lots of former winning trainers, and Sam Brown bids to be the first repeat winner in the past decade. Being aged 13 won't stop him - three such veterans prevailed since the race began in 2016. And it's not been a great race for the top of the market with just three of the nine winners coming from the top four in the betting.

There's quite a narrow band on DSLR - see below - and we're not looking for a recent winner typically. Four of the nine winners were well fancied last time (first or second favourite, i.e. (Rk) (1) or (2)); and one or two runs in the past 60 days is standard, though that will likely be nearly all entries. Five winners finished 6th or worse, or failed to complete last time.

Looking at all that, one might split a tenner at prices on this quartet, though given their uninspiring surface form they may be bigger odds on the day:

So that's the new TRENDS tab. It exists for big races and little races; and it's live for Lite and Gold subscribers right now.

Note, if the tab is greyed out, there are no past editions of that race. [And we're aware of a small bug where the tab is greyed out for second divisions of a split race - we'll get that fixed, but I didn't want to delay releasing this new tab into your information portfolios].

Matt

What to expect in 2024

4 January 2024/in Geegeez Gold, Horse Racing Blog/by Matt Bisogno

The new year is well upon us now and, on this fourth day of January, a few resolutions may remain intact. Chocolate, biscuits, cakes (and especially chocolate biscuit cakes) and beer are largely off the agenda for a bit here - yes, life is currently very dull - but, on a much more interesting note, below are some words around what is on the 2024 agenda for geegeez.co.uk...

Racecard Small Changes

We'll start with a 'not very rock n'roll' update: a collection of small changes to the racecards. Although small, most of them are things many users repeat countless times while navigating the software in search of interesting horses.

22nd January Update

These changes are now live and you can see them in action in the video below - there's a timeline below the video:

00:00 Intro
00:31 Save Racecard Filters (desktop & mobile)
02:45 Actual Race Distance in form blocks
04:16 Full Form UK/Ire filter
05:25 'By Time' Racecard view now has time order dropdown
06:40 Asterisked notes
09:10 Run Style added to Full Result
11:00 Removed 'abandoned' meeting non-runners from Tracker
11:35 Outro

Editorial Explainer

First up is a racecard menu filters 'memory' - currently, a user must select parameters from the racecard menu filters section each time they close and open the cards menu page. If you use the same filters all the time, you have to reinstate them each time. Faff. We'll sort that.

[Incidentally, if you sometimes see there are no races displaying on the menu page, just hit the 'reset' button top right]

Next, an asterisk on the form row when you have a note saved for any/all of meeting, race or runner - to notify you that it's there.

Also, we'll be displaying the specific race distance and any distance amendments when you hover over the 'Race Conditions' on any form row:

And, if you choose to view the racecard menu page 'by time' you can view the race dropdown ordered by time.

If run style is of interest to you, we're adding each horse's early pace position to the full results:

We'll get those small, but perfectly formed, changes live later this month.

Betfair Data

One of the projects for later in the year is to incorporate Betfair data - Betfair Starting Price (BSP) as well as in-running high and low prices. We actually have these data in our system but adding them appropriately to results and into the tools will take a while. But it's on our to do list.

For a lot of readers who have been restricted, some of the BSP results are likely to make eye-opening reading, certainly when compared to SP.

Ratings Model

This is one of those dreaded rabbit holes into which I vowed we'd never delve. Well, we have already sunk a good few hours into the project and we've made some promising progress; but there is much still to do. I'm at the point now where, for the first time, I do believe we can produce a set of ratings that a) finds a lot of winners and b) highlights some value.

The process involves creating separate models for separate groups of races, and if/when we get as far as publication, we'll do it piecemeal. That is, once we're happy with, for instance, our all-weather sprint handicap model, we'll publish numbers for all-weather sprint handicaps. And so on.

There are loads of ratings out there, many of which are very good at finding winners - but due to the fact they're published so widely they are significantly loss-making. Our Peter May ratings get close to break even at Betfair SP with their top rated picks every year, sometimes turning a small profit and sometimes a small loss. And we might not be able to fare better than that.

My main point is that, unless we find something of utility, as opposed to the somewhat ornamental numbers produced by the fashionable houses, we'll not publish.

Query Tool

QT is a powerful means of analysing large chunks of racing data and, once that's done, of saving specific 'QT Angles' to your own account and being notified of qualifiers each day. It's been unchanged for a few years now, and we've aborted a few attempts at an upgrade; but I have so many things I want to add to QT - a majority of them from your feedback and suggestions - and, once we've re-engineered a QT 2.0 engine, it will be relatively straightforward to deploy that extra functionality.

This WILL happen in 2024, it's been too long.

As you can see, apart from the small changes due for release this month, we've got a couple of pretty big 'how long is a piece of string' projects for later in the year. The Betfair element shouldn't be too onerous but I'd like to put some developer time into the modelling next. Very, very loose timeline would be aiming to get some flat rating models on stream for the start of the turf season; then perhaps pivoting to the Betfair and QT projects before reverting to the remaining race code ratings models.

There is a lot of scope for timelines to change, but these are the 2024 resolutions for geegeez. Let's hope they last longer than my personal attempts at self-improvement!

Matt

p.s. away from the bright lights of geegeez, there are a couple of other interesting projects on the go. One, a tote ticket builder, should hopefully be available very soon (I've been using it for a year!), and the other, TennisProfits.com, is a site for tennis traders that we're hoping to make more accessible for bettors, too. I'll share snippets on these from time to time as the year progresses. The tote ticket project especially is one that I think will be of great interest to many geegeez readers/racing punters.

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.

Membership Login

Tag Archive for: geegeez Gold

A four-week trial of combining Pace with SR Ratings, Part 2

A four-week trial of combining Pace with SR Ratings, Part 1

Week 1

Week 2

An investigation into the Geegeez Instant Expert, Part 2

Recap

A line on ranking methodology

Instant Expert Rank #1 by Betfair Starting Price

Instant Expert (IE) top ranked runner by Price

Instant Expert (IE) top ranked runner by Price (4yo+ hcaps only)

Instant Expert top ranked runner by Age (4yo+ hcaps only)

Instant Expert rank of 9 or lower by Class

IE rank of 9 or lower by Race Classification

Individual Instant Expert Shortlist scores by PRB

Instant Expert individual scores

Instant Expert individual scores (4yo+ hcaps only)

Summary

An initial investigation into Geegeez Instant Expert

Introduction

Instant Expert Shortlist (Sh) – Performance by Rank

Instant Expert Shortlist (Sh) – top two ranked runners

Shortlist Rank Comparison: 3yo+ Handicaps vs 4yo+ Handicaps

Enhanced Tracker buttons on the racecard

Breeding suffixes on Full Form and Profiler tabs

Shortlist scores on Instant Expert

'Last 5 Years' option on Draw Tab

Each way terms added to the Odds Tab

Report tweaks galore and a bug fix

VIDEO #1: Query Tool Overview

VIDEO #2: QT Angles Enhancements

Query Tool

New Variables

Season

Owner

Racing Post Rating and RPR Rank / Topspeed and TS Rank

PRB

A New Look for the Reports

Timeline for Implementation

Trainer Form

Price Movement

What else?

And finally...

Racecard Small Changes

22nd January Update

Editorial Explainer

Betfair Data

Ratings Model

Query Tool

Impact Value (IV)

What to look for with IV

IV3

What to look for with IV3

Actual vs Expected (A/E)

What to look for with A/E

Percentage of Rivals Beaten (PRB)

A word on non-completions

What to look for with PRB

PRB2

What to look for with PRB^2

PRB3

Horse Racing Metrics Summary

PRB²