And the winner is…

I recently invited readers to submit an article by way of applying to write on the site on a more regular basis. The brief was that the article should be well-researched, cover an interesting angle, and have its rationale teased out. I received a dozen entries, and these were my top three. See if you agree…

3rd place: Adam Hills – ‘No Name Trainer Track Angles’

Anyone who has ever dabbled in having a flutter on the geegeez (pun intended) will be familiar with the phrase “horses for courses” and there is merit in awarding mental ticks to horses who are able to demonstrate favorable (and profitable) form at any given course. What is perhaps less obvious is the idea of “trainers for courses”. 

The aim of this series would be to put a handful of trainers under the spotlight to investigate their historical trends in terms of the performance both on a course-by-course basis but also drilling much deeper into associated micro-angles. The basis for this article was derived using the pre-defined reports on the GeeGeez website.

These angles specifically focus on some of the perceived unfashionable trainers who I feel often fly under the radar and therefore offer significant value in the market.

Here is an example: Derek Shaw, running a handicapper at Chelmsford – boasts a record of 36 winners from 263 runners since 13th August 2013 (5 years). Those 36 winners returned a pleasing but not earth shattering +9.29 win PL to a pound stake. If you drill into that just a little deeper and remove any horse aged 2 or 3, then that win PL figure jumps up to +36.79 with 33 winners from 205 runners. Scratch at the surface just that little harder by adding a further condition of those horses who carried a higher weight (9-04+) and we see some quite astounding results:

  • 22 winners from 75 runners;
  • a strike rate of 29.3%;
  • a win PL of +84.79 (to £1 stake at SP);
  • an ROI of +113.1%;

The Derek Shaw’s of this world cannot compete with the powerhouse trainers of this world, but they have their place and they have their day.


2nd place: Richard Hunt – Jockey Performance by Distance

In horse racing, a lot of time is spent by so-called experts looking at a horse’s breeding. This will hopefully infer whether the horse will be a sprinter, middle distance runner, or need three and a half miles around Ffos Las in the mud.

In this article, I wanted to look at the person on top i.e. the jockey, and whether the jockey’s performance is related to the distance of a race. In many other sports, the competitor is good at a specific distance: for example, Usain Bolt is the world’s fastest over distances up to around 200 metres but wouldn’t stay four miles in the proverbial horsebox. Similarly, in cycling, some guys are all about sprints while others need the Tour de France to show their best form.

Hence I have looked at how jockeys perform by distance of a race. There are obviously a lot of variables involved and in order to reduce the variation as much as possible some restrictions have been used in the dataset.

I am only considering handicap flat races in the UK that were run on turf since the start of 2015. We are all aware that some novice and maiden races are often used for teaching a young or inexperienced horse and so might not have the cut and thrust of a handicap. Currently handicaps form the majority of races on a typical mid-week card so it is not a great restriction in terms of numbers. The other consideration is getting sufficient number of rides in a particular category and as such I have grouped races distances into three categories:

Sprint: up to 7.5 furlongs

Middle: from 7.5 furlongs to 10.5 furlongs

Long: anything above 10.5 furlongs.

The resulting analysis is shown in the table below. Jockeys in alphabetic order by first name.

Your first 30 days for just £1


  Long Middle Sprint Long Middle Sprint Long Middle Sprint
Jockey A/E A/E A/E Strike Rate(%) Strike Rate(%) Strike Rate(%) ROI to SP ROI to SP ROI to SP
Adam Kirby 0.8 1.2 1.1 11 16 14 -47 -2 -7
Andrew Mullen 1.2 1.0 1.0 11 7 8 -14 -6 -18
B A Curtis 1.0 0.8 1.1 13 9 13 -4 -49 -14
Daniel Tudhope 1.1 1.2 1.0 19 18 15 -6 10 8
David Allan 1.3 1.2 1.0 19 16 12 42 7 -15
David Probert 1.0 0.9 1.2 12 10 15 -44 -21 16
F M Berry 0.9 1.2 0.9 12 13 10 -24 -1 -23
Franny Norton 0.9 1.2 0.9 14 17 11 -49 9 -17
Graham Lee 1.0 1.0 0.8 10 10 8 18 -30 -39
James Sullivan 1.7 1.0 0.9 14 8 9 42 -13 -27
Jamie Spencer 1.0 1.0 1.2 18 14 16 -9 12 3
Jason Hart 1.0 0.9 1.1 11 8 12 -4 -31 -5
Jim Crowley 1.1 1.2 1.1 16 19 17 -12 -4 3
Joe Fanning 1.2 1.0 0.9 21 13 11 15 -20 -19
Josephine Gordon 1.4 0.9 0.9 15 10 9 21 -25 -38
Luke Morris 0.8 1.0 0.9 15 12 9 -35 -37 -21
Nathan Evans 0.5 1.2 1.4 4 13 12 -70 -11 -5
Oisin Murphy 0.8 1.0 0.9 13 14 10 -20 -10 -34
P J McDonald 1.1 1.4 1.2 13 14 13 -12 1 3
Paul Hanagan 0.9 1.0 1.0 10 13 13 -40 -14 -12
Paul Mulrennan 0.9 1.0 1.0 11 10 11 -36 -23 -29
Phillip Makin 0.8 1.0 0.9 12 11 10 -21 -18 -33
Richard Kingscote 1.3 1.4 0.9 18 19 12 14 18 -12
Silvestre De Sousa 1.1 1.2 1.1 22 21 18 -11 0 -4
Tom Eaves 1.0 0.9 1.0 8 7 7 -13 -14 -27
Tom Marquand 0.9 0.9 1.1 9 8 11 -39 -37 -26


I have restricted the table to jockeys with at least 100 rides in each distance band and overall at least 600 rides. Note this is rides in handicaps only and so certain top jockeys like Ryan Moore do not appear in list.

Familiar statistics such as A/E, strike rate and ROI to SP presented.

Most jockeys have A/E in the range 0.8 to 1.2. The interesting ones are those with higher or lower figures.

James Sullivan has an A/E of 1.7 in long distance races with an ROI of 42% while average figures in the other distance groups.

On the other hand Nathan Evans only has an A/E of 0.5, a strike rate of 4% and a 70% loss on ROI in long distance races while in sprint distances has an A/E of 1.4 but still a small loss on ROI.

What does this mean for my investments? Synopsis ends here


1st place: Jon Shenton: The Curious case of Paddy Mathers…

Often when I’m cycling through the countryside in a vain attempt to keep the sands of time at bay I think about racing angles, racing data and systems. Endlessly trying to find ways to beat the bookies. It’s not the money, it’s the numbers, the maths and the perfect combination of where cold hard facts collide with traditional sporting theatre.

On a recent sojourn around the beautiful Leicestershire countryside (yes really), I was thinking about horses that had displayed differing levels of aptitude on left handed and right handed courses and how this could be used as an angle for more effective wagering.

I wondered whether the same might apply to the pilots we know and love (most of the time), gut feel being that is was a ridiculous notion and a jockey would be equally as competent riding around both left handed and right handed bends. I mean, I’m no horseman by any stretch of the imagination but I can’t think of a single logical reason why there should be any difference?

With curiosity piqued I fired up the excellent horseracebase to analyse the notion, using all flat turf or all weather races ran since 1st Jan 2013 around LH or RH bends. Races on straight tracks have been excluded from the number crunching (for now).

Using the Actual/Expected ratio we can carry out some quick high level comparisons of jockey’s performance between their runs around left and right handed bends. To mitigate irresponsible conclusions from small sample sizes only jockeys who have 100+ runs on both left and right handed bends have been considered.

The table below shows those with a stronger preference for the left hand bend, simply by evaluating the variance in their A/E performances across the respective directions.

Jockeys whose performance going left-handed is markedly superior to their right-handed performance

Jockeys whose performance going left-handed is markedly superior to their right-handed performance

A couple of immediate observations, if we look at the column RH A/E this shows the ratio of actual wins against expected for the named jockeys when riding right handed, we can see that on the face of it there is some serious underperformance, with eyewatering losses in terms ROI to match, a
whopping 80% of your stakes in the case of Paddy Mathers and Dougie Costello.

The column LH A/E given the comparison when these jockeys go left handed, we can see that the performance is generally much more in line with market expectations, although it doesn’t look (at first glance anyway) that use of this data is a launchpad to the private jet and riches. Interestingly
the aforementioned Paddy Mathers does outturn a small, tiny return of just over 1% if you backed him on all 733 of his left handed runs. Comparable to investing in a high street savings account over the same period, and tax free too.

I feel inclined at this stage to investigate Mr Mathers stats a bit more closely, the variances between LH and RH performance is so notable, impossibly large for my mind. Down to chance or an underlying trait in his riding ability?

One possible explanation may be that the trainers he rides for specialise at tracks with a LH turn, meaning the reason for the variance could be more associated with the trainer than the jockey. Looking at the table below it shows the stables that Paddy Mathers has ridden for most frequently
on turning tracks from the start of 2013, interesting to note that if you back him every time he jocks up for Richard Fahey you’d make a few pennies over nearly 300 runs.

Stables for whom Paddy Mathers has ridden most frequently since the start of 2013

Stables for whom Paddy Mathers has ridden most frequently since the start of 2013

So let’s look at their combined performance on turning tracks taking our Paddy out of the equation to see if there appears to be any bias to a particular direction.

Selected trainers combined performance for jockeys other than Paddy Mathers since the start of 2013

Selected trainers’ combined performance for jockeys other than Paddy Mathers since the start of 2013

Not really is the answer. In fact looking at win percentage these trainers are more likely to have winners at RH tracks than the opposite direction.

So in the case of Paddy Mathers, Dougie Costello and the up-and-coming Charlie Bishop (to name but three) they appear to be either very unlucky in the rides they have secured around right handed bends or, perhaps more likely, they genuinely are not as competent at riding right-handed as they are left-handed.

To be honest, it still doesn’t make logical sense to me, though it’s now a factor I’ll look out for when trying to resolve punting puzzles.


I’d like to express my gratitude to all readers who took the time to send in an article – the standard was consistently high – and, with a following wind, we’ll be hearing more from our winner in the near future.


p.s. which of the three articles did you enjoy the most? What sort of information would you like to see researched in future posts? Leave a comment and let us know!

Your first 30 days for just £1
0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *