Part 1 of Betting Angle Research: it needn't be this complex!

Dave Renham: A Window into My World of Racing Research, Part 2

On January 25th of this year I wrote a detailed piece which I hope gave readers an understanding of how much work can go into researching horse racing ideas when some, or all, of your research requires a race by race approach, writes Dave Renham. This was the only method of research in the old days before computers and racing databases, and this is what I primarily did going back to the early 1990s.

Of course, nowadays 99% of my articles are sourced by solely using databases such as the Geegeez Query tool, the Geegeez Draw Analyser, etc. However, there have always been significant advantages to this old fashioned slow data collection method, primarily because you do get a proper ‘feel’ and understanding for the data you are collating. You can see patterns that might be missed if simply pressing a button and just getting the raw stats breakdown. The downside is obviously the time it takes to go through race by race.

In my first piece I looked at a specific group of races – these were all-weather handicap races run around a bend with eight runners, over the sprint trips of 5f and 6f. I looked at data for four seasons which equated to 190 races. I chose the sprints simply because I have always been a fan of handicap sprints and most of my bets occur in such races. I looked at the effect of the draw, running style/pace data, market factors and Peter May speed ratings. The key findings from this research were:

  1. As a general rule, on turning sprint tracks a lower draw is preferable due to its position closest to the inside rail;
  1. The top three in the betting combined broke even to BSP;
  1. Horses with higher Geegeez pace/running style total scores based on their last four runs win more often than those with lower totals;
  1. Horses from the top three of the speed ratings scored much more often than those fourth to eighth, although profits/losses between the two groups were similar.


Your first 30 days for just £1

You can review that article here.


So what have I looked at this time? Well, I've started with a similar group of races to before, only this time I have focused on UK 5f handicaps only, on the turf. I did exclude 2yo nursery handicaps as such races have more limited past data. I expanded the field size to include races between 8 and 11 runners, while keeping the same four year time frame (2019 to 2022). This gave me a much larger number of races to research – just under 540 in fact. With a greater data set to examine I hoped that it may lead to some angles that we can use to our advantage in the future.

At this point I should add that there was a small tweak in terms of what I focused on compared with the original article. I did not examine the draw this time, because of the different nature of the various five-furlong course constitutions. There are some straight 5f tracks, some round ones, and also there can be different stall positioning on some straight tracks. All this means there is no uniformity so it made sense to focus on other things. Hence, for this research I not only collated the Peter May ratings again, but I also added the Topspeed speed ratings as well; and I looked at the Geegeez pace/running style data from the past four races, as I did last time. Market rank was also once again looked at.

Now, I could not have done this much research by simply pulling out each race result, then writing each horse, course, finishing position etc into a spreadsheet. That would have meant manually typing in 16 distinct columns of data for 4953 rows of individual runners. I mean, I like my research, but 79,248 cells to type a number or word into is too much even for me! I’m all for finding racing angles that others don’t, but personally typing numbers and words into nearly 80,000 cells would be a tad crazy!!

Instead, I used two databases, one being the Geegeez Query Tool, in order to get most of the columns into the spreadsheet far more speedily. This gave me my starting point as 12 of the 16 columns were done, but there were four missing that I needed to add individually - these being the specific run style (pace) data, market rank and both sets of Speed Rating data. Now market rank was relatively ‘easy’ to add using some excel tricks, but it did leave me the run style (pace) data and speed rating positions to add manually.

As you can imagine, manually adding run style and speed ratings data race by race for over 500 races took some time – a lot of man hours. Thus, for both sets of speed ratings and the pace/run style scores, I decided to add the top three ‘rated’ horses only to the spreadsheet. If I had manually added every single ranking position to those columns the research would have taken twice as long, maybe longer.

Having set the scene it’s almost time to dig down and share what I found, but first I want to show you where you can find the relevant data on Geegeez.

In terms of speed ratings you can go to the ‘CARD’ tab on a specific race and you will see the Topspeed figure for each runner (column headed TS) and the Peter May Rating (column headed SR). The 2022 race from Chester (below) was one of the races in the 537 race sample:



Once I scrolled to this page I could sort the columns to display the horses that were in the top three of each rating column (e.g. the highest three figures). In the example above I have ordered the Peter May figures (highest at the top). We can see that Le Beau Garcon had the highest speed fig (81), So Smart had the second highest (67) with Riversway third (65). Once a race was sorted like this I labelled the three horses 1, 2, 3 on my spreadsheet. I then sorted the Topspeed ratings and repeated the process.

For the run style / pace data I wanted to find the top three horses in terms of their pre-race Geegeez run style/pace total from their previous four runs. To find what I needed I clicked on a race result, and once the result came up I then clicked on the ‘PACE’ tab. From there I ordered them with highest totals first – an example of what I mean is shown below from a 5f handicap at Catterick in 2021:



Here, Autumn Flight was top ‘pace’ rated with 15, Major Jumbo second on 14 and Militia third on 13 and, as with the speed ratings, I labelled the three horses 1, 2, 3 on my spreadsheet. Now, occasionally you will see a horse that had a ‘U’ rather than a number in one of the last four race columns. This occurs occasionally when it is unclear from the in-running comments what pace number should be assigned to that specific run. For these horses I double checked different sources, or even watched the start of the relevant race so that I could add the right figure. I would then recalculate those specific horses’ total.

It should also be stressed that there are times when you get horses with identical four race pace totals, which means it is potentially difficult to get an exact ‘top three’. In the event of tied four race totals, I look at the most recent race first (LR column) and compared the horses who have tied. The horse with the highest figure in that first column would take priority. If the scores are the same for the LR, I would then compare the next column (2LR) and keep going until one out scored the other. Here is an example of such a case:



This Thirsk race saw three horses tie for second with 12 pace points each. However, if we compare the LR column we see Spanish Angel scored 4, Boudica Bay 3 and Dandy Spirit 1. For this race Spanish Angel was placed second behind Birkenhead, and Boudica Bay third in terms of their run style/pace position.

OK, it is time to share my findings...


Topspeed Ratings

Starting with the Racing Post Topspeed figures, and comparing win strike rates.



In truth, it was a little disappointing to see a relatively even split. OK, the 4th+ speed rated horses did have the lowest strike rate, but I had expected / hoped their strike rate figure would have been lower. Not only that, the 4th+ rated horses actually turned a very small profit to BSP.


Peter May Ratings

Let us now look at the performance of the Peter May (SR) Ratings – again comparing the top rated, 2nd rated, 3rd rated and combining those rated 4th or lower:



This is a more encouraging picture – in terms of win success at least. There is a clear drop off when we look at the horses with a speed rating position of 4th or lower. In terms of returns, the 2nd top rated horse made a profit of around 9p in the £ to BSP if backing every single one ‘blind’.


BSP Market Rank

Let's look at market rank now. I was able to rank all horses, not just the top three and here are the strike rate splits:



We can see a familiar sliding scale here: favourites winning close to 30% of the time, whereas outsiders ranked 6th or lower in the betting market combined to score just under 5.5% of the time. Favourites would have almost broken even with a loss to BSP of just 2p in the £; while second favourites got closer still, losing just a penny in the £.


Run style / pace

This is my favourite area of research, especially in sprints, because in general, there is a strong front-running bias in shorter distance handicaps such as these. Therefore I was hoping for some relatively positive stats. Here is the breakdown for the top three Geegeez pace/run style rated horses, as well as the combined results for horses rated 4th or lower:



In general these figures are encouraging, especially when we look at the profit/loss column. The top three rated horses have combined to be far better value than those rated 4th or lower. The strike rates are much closer, although the top rated runners have secured the highest win rate.

When we dig a bit deeper, it is worth noting that the profit and loss figures are not skewed due to the higher rated pace horses having more big-priced winners. In fact, if comparing the odds of the ten highest priced winners from the top rated pace group against the fourth or lower group, we see the following:



Clearly, the lower rated pace / run style runners have had bigger priced winners overall. The average BSP price for these ten runners for 4th+ rated is 51.68; for the top rated it is 33.45.

I have mentioned numerous times in previous articles how important it is to look deeper into profit and loss figures. You need to make sure the bottom line is not giving a false impression.

I want next to look at a couple of angles concentrating solely on the top rated pace/run style runners. Firstly a look at the yearly breakdown:



It is very promising to see that three of the four years turned a profit to BSP. The strike rate in 2019 was a fair bit lower, but when horses win on average 12.9% of the time, it is not unusual to see a 10% strike rate over 150 races.

Now I am splitting the top rated pace/run style results by number of race runners:



Obviously, ignore the diminishing strike rates as field size grows: that is to be expected as it is easier to beat seven rivals in an 8 runner race than it is to beat ten rivals in an 11 runner race. Again, three of the four sections have produced positive returns and the other broke even.

The final area I wanted to look at was combining ratings with the Geegeez pace/run style ratings. So, firstly, how did horses do if they were in the top three of all three? That is, they were one of the three highest Topspeed ratings, one of the top three in the Peter May ratings and one of the top three in the four-race Geegeez pace/run style totals. Well, 316 horses qualified, of which 51 won (SR 16.1%) showing a small profit of £26.96 (ROI +8.5%). This was encouraging.

Let me compare these results to horses that were not in the top three of any of the Topspeed, Peter May and pace ratings. There were of course far more qualifiers – 1775 to be precise. 161 of these won, equating to strike rate of 9.1%, which is quite a difference. However, those lower-rated qualifiers made a slightly larger absolute profit of £50.09, but with a smaller ROI of just 2.8%.

This feels positive overall, especially the strike rate difference (16.1% versus 9.1%). While both groups turned a small profit, when I again dug deeper and looked into the biggest winning prices for each group, we can see the variance in their five highest winning priced runners:



Looking at this clearly demonstrates that horses that are rated in the top three positions of all three - Topspeed, Peter May and Pace ratings - are far better value - and less susceptible to skewing - than horses that lie 4th rated or worse in all three of the ratings.

Before I finish, I guess you may be wondering how horses that were top rated in all three ratings got on? Well, the problem here is number of qualifiers – there were only 24 runners that ticked all boxes over this four year period. Having said that 10 did win (SR 41.7%) for a BSP profit of £34.44 (ROI +143.50). Now whether these runners will continue to perform as well in the future is difficult to say as 24 runners is such a small sample. However, I personally will be keeping an eye for them. [Stop press: since researching this piece, Rajmeister was a 10/3 winner for this angle on the 22nd April]



Well, this has been quite a journey for me and a long one at that. Having to go through race by race is hard work, but ultimately I think the research uncovered some interesting findings. Not only that, it has inspired me to do some more digging around these themes, albeit it will no doubt be a rather slow excavation!

Before I finish I should mention that all BSP profits and losses have taken a 5% commission into account, as that gives the truest reflection of real life returns using that medium. Many punters are now paying only 2% commission, so if that's you, you can mark up the profits published above accordingly.

I hope you have enjoyed this journey and I hope it shows the usefulness/importance of some of the Geegeez data you can get from each racecard. I’ll be back next week looking at run style in smaller field handicaps. Until then...

- DR

Other Recent Posts by This Author:

Your first 30 days for just £1
4 replies
  1. Monsieurbernie
    Monsieurbernie says:

    interesting , think i have my settings for topspeed and peter mays rating to be off my geegeez racecard but i may add peter mays ratings in future… may help me sway for one more than another if i cant pick between two!!

    who is peter may anyway?!

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.