Is Recent Trainer Form Important?

Spring is upon us, and the turf flat kicks off this weekend, writes Dave Renham. It is my favourite time of the year as all of my winter research can be unleashed in an attempt to hammer the bookies! Of course, despite all the hard work put in it, it does not guarantee profits for the year ahead. No doubt it will be the usual rollercoaster of good weeks, bad weeks, and indifferent ones. Hopefully, though, there will be more good than bad!

In this article I am going to investigate recent trainer form to see what I can unearth. Newspapers, betting sites and some pundits seem to place a lot of stock in trainer form; I must admit that I tend not to, but I am prepared to change my mind depending on what I discover: have I been missing a trick all these years? Let’s see.

I have taken flat and all-weather data from 2021 to 2023 for UK racing and profits/losses have been calculated to Betfair Starting Price less 5% commission (readers can do better than those results by selecting the 2% commission option in your Betfair account).

My focus is going to be on 14-day trainer form, but with a caveat. The caveat is that a trainer must have had at least 20 runners during that time frame. I’m adding this to make sure we eliminate small samples which are unreliable.

To try and explain how small samples can be unreliable, a trainer could theoretically have had five runners in a 14-day period with one win and four losers giving them a recent win strike rate of 20%. Generally, a strike rate of 20% for trainers at any time is considered very good. However, firstly that 20% strike rate is based on very limited data. Secondly, let us imagine all five runners had been odds-on favourite – that would not be crying out good trainer form. To be clear, I’m not suggesting that 20 runners in a 14-day period is the perfect number of runs, but it seems as sensible an arbitrary figure as any.

General 'Recent Form' Trainer Statistics

Time to review the first set of data. Here, I am looking at trainer strike rates – both win and each way linking with the win percentage individual trainers had achieved in the previous 14 days.

 

 

This initial piece of evidence suggests, at least from a strike rate angle, that trainers who had been in better form over the past 14 days outperform those whose recent form had been less good. Both the orange line (Win SR%) and the blue one (EW SR%) show solid correlation with the graph, on an upward trajectory.

However, as a recent article of mine suggested, strike rates are not the be all and end all; we need to look at value and profit / loss. The table below gives us a breakdown of this:

 

 

This presents a less clear picture but, looking at the returns, the trainers who have scored 5% or less with their last 20+ runners over the most recent two-week period have provided the worst returns. Looking at the trainers with the highest recent strike rate (31% or more), it appears that the betting market has compensated for this to a great extent given losses of more than 13p in the £.

Of course, the data shared so far looks at all trainers combined. This gives us a general starting point, but to get a better overall perspective we need to split the trainers into groups. This is because comparing the 14-day win strike rate of Charlie Appleby, say, with Liam Bailey makes little sense. Appleby has a 29.6% strike rate during the time frame compared to Bailey’s 4.3%. In addition, Appleby’s A/E index of 0.94 is nearly double that of Bailey whose figure stands at 0.48. Hence in a scenario when Appleby and Bailey have secured a 15%-win strike rate in the past 14 days, we should be aware that Bailey is performing well above his norm and Appleby well below. Looking at all trainer data therefore gives us a flavour but cannot paint the full picture.

To mitigate somewhat for this, I am going to consider split trainers into groups based on their annual win rate while analysing recent form (14-day results).

I have split them into five groups: those with an annual strike rate of 8% or less, those between 9% and 12%, 13% to 16%, 17% to 20%, and finally trainers winning at a rate of 21% or more.

Trainers with a yearly win strike rate of 8% or less  

Looking at all trainers who had run at least 200 runners from 2021 to 2023, around a fifth of them had an overall win strike rate of 8% or less. With this group I would not expect to see many qualifiers in the higher 14-day strike rate groupings.

 

 

The biggest group of qualifiers has occurred in the 6-10% 14-day strike rate (SR%) bracket and they have made a BSP profit. However, this is down to two unusually big-priced winners of 451.93 and 350.0 which skew the figures considerably. What is clear is that once trainers in this group hit 16% or more with 20+ runners in a 14-day period they do start to have better results. Combining the results of the 16%+ group they have returned 10p in the £ with a solid A/E index of 0.93 from a sample of around 400 runners.

Trainers with a yearly win strike rate of between 9% and 12% 

This group of trainers is the biggest and includes runners from the stables of Richard Hannon, Richard Fahey, David O’Meara, Kevin Ryan, and Jim Goldie to name but five. Let us look at the overall win and each way strike rate first:

 

 

The graph shows that this group of trainers have performed quite poorly in terms of win strike rate if their 14-day Win SR% had been between 21 and 30%, which is surprising. I expected the orange and blue lines to rise gradually and relatively smoothly from left to right.

When we compare the A/E indices we see a similar pattern:

 

 

Combining the runners from the 21-25% group and the 26-30% group would have lost over 16p in the £ to BSP. Even the 31%+ group lost 15p in the £ despite a decent A/E index. It seems that the market is aware when these types of trainers are showing good recent form and prices have been more than adjusted to account for it. Here is a tabular view of this group:

 

 

The 0% group has performed better than expected. However, as we will see, this turns out to be an outlier when considering the rest of the research I share.

 

Trainers with a yearly win strike rate of between 13% and 16% 

This group of trainers includes Charlie Johnston, Archie Watson, and Clive Cox. Here are their stats:

 

 

There are better strike rates across each grouping as we would expect from higher general strike rate trainers, but these handlers look worth avoiding if they have had no winners from 20+ runners in the past 14 days. Losses of over 26p in the £ is steep albeir on a smaller sample size.

The picture is not much brighter at the other end of the scale – trainers that seem to be in flying form with a 14-day SR% of 31% or more have made losses of over 14p in the £. Their strike rate is relatively poor, too, at 15.09% and their A/E index a modest 0.86. These are the two key discoveries from what is essentially another mixed bag of findings.

 

Trainers with a yearly win strike rate of between 17% and 20% 

Let’s see if this smaller group of trainers can give us some stronger patterns. Ralph Beckett and Sir Michael Stoute are two trainers who are included here in the following results:

 

 

When these trainers are out of form (14-day SR% of 5% or less) they look worth avoiding. Combining the top two rows of data we see significant losses of 27p in the £. Conversely, when they hit the dizzy heights of 31% winners or better in the past 14 days this cohort has edged into profit. I would guess that it will come as less of a surprise for such trainers to hit these levels from time to time and perhaps the market has failed to properly adjust.

Overall, these stats are strong with a decent strike rate of close to one in four, returns of over 7p in the £ and an excellent A/E index of 1.04. In general, these stats correlate better with what I would have expected to see.

 

Trainers with a yearly win strike rate of 21% or more 

We are now looking at a very select band of trainers including Charlie Appleby and Willie Haggas. Here are their win strike rates based on recent form:

 

 

I want to focus on win percentages (hence no EW ones on this graph) to help illustrate how little difference there is in some of the numbers. For example, when this group of trainers had previously had a 6-10% win SR% over the past 14 days their win rate was 19.8%. When it was 21-25% in the previous 14 days their win rate was virtually the same at 19.9%. Let me share the fuller picture:

 

 

The first thing to point out is that a 14-day SR% of 5% or less is extremely rare in this collective, as one would expect. Secondly, I want to highlight the poor performance of the 31%+ group which incurred losses of over 18p in the £ and produced a disappointing A/E index of just 0.83.

There is no easy explanation as to why the 26-30% group have fared so well in comparison. My reading of the data is that when these yards seem a little under par (previous 14-day SR% of 11 to 20%) their runners may offer a little bit of value. I guess punters could be put off by the relatively modest recent strike rate, but essentially these runners are still scoring close to each trainer’s norm.

At this juncture the picture is quite muddy when it comes to 14-day trainer form. The strongest and most important finding is that we can say that a 14-day win SR% percentage of 5% or less is a negative.

*

Individual 'Recent Form' Trainer Statistics

It is time now to look at individual trainers. With each trainer plying their trade in a different way, one would hope there might be more insights to glean here. Fifty of the leading trainers are shown in the table below in terms of their win strike rate (missing cells are due to limited data):

 

 

Before going into more detail let me share their A/E indices with you as well. I have colour coded the A/E indices with positive figures of 1.00 or more in green and negative ones of 0.70 or less in red.

 

 

Individual Trainer Angles to Note

These are what I think are the most significant findings:

1. Andrew Balding does unexpectedly well when his recent 14-day SR% is less than 5% returning 37p in the £ on all runners. He also made a BSP profit with his 6-10% group.

2. Ralph Beckett looks a stable to follow when he hits the 14-day win SR% of 26% or more. An A/E index of 1.05 and a small profit to boot.

3. The Crisford stable has a bizarre set of strike rates and A/E indices:

 

Both metrics correlate with each other which gives us confidence in the findings, but the data is suggesting that the poorer the recent form of the stable the better.

4. Don’t be put off by low recent 14-day win strike rates for Eve Johnson Houghton. When her strike rate was 5% or less in the previous two weeks her runners have offered good value. Backing all runners blind in this context would have secured a small profit.

5. When Alan King’s 14-day SR% has been between 21% and 25% he has managed just one winner from 62 runners. Losses of 90p in the £ would have been recorded backing horses from a so called ‘hot’ stable.

6. Daniel Mark Loughnane has the type of profile I was expecting more of. When his 14-day SR% is 5% or less his record is very poor. When it hits 26% or more his record has been excellent. He does a look a clear case of “avoid when his stable is cold, take advantage when the stable is hot”.

7. When the Kevin Ryan stable is out of form, his runners are probably worth swerving, especially if he has failed to record a winner from 20+ runners in the previous 14 days.

8. Saeed Bin Suroor’s A/E indices for the four groupings in which he has enough data are all above 0.90 suggesting his stable perform in a similar vein regardless of very recent form.

9. Grant Tuer shows a similar pattern to Crisford implying that the poorer the recent form of the stable the better from a betting perspective.

 

*

Summary

So, what are the main takeaways from this research? When looking at the general picture, I think the data indicate that poor 14-day form has more of an effect on performance than good 14-day form. ‘Cold’ stables with 14-day win rates of 5% or less from 20+ runners are best avoided (with a few individual exceptions).

In terms of when a stable has been ‘hot’ over the past 14 days we see an uptick in win rate, but this does not guarantee value or profitability due to such form being fully exposed in the market.

Regarding when a ‘hot’ trainer should potentially be followed I would use the individual trainer table of A/E indices looking for figures close to, or greater than, 1.00.

I plan to revisit this idea in the future, looking at a slightly longer prior time frame to see what difference, if any, that makes. I could also delve into NH trainer stats as well. But that’s for another day. I hope you have enjoyed this piece, and good luck with your early flat season betting.

• DR

Other Recent Posts by This Author:

• Top Ten All-Weather Front-Running Biases
• National Hunt Racing Systems for 2024/25
• A Look at All-Weather Returners
• Hat Trick (Plus) Seekers in NH Racing
• Sires on the Sand: All Weather Stallions To Note