This is the third article in a series in which I have been looking at run style bias in relation to trainers, writes Dave Renham. In this piece, I'll drill down looking specifically at trainer data from two-year-old (2yo) races. As with the previous articles (read them here and here) I have looked at 8 years' worth of data (1/1/14 to 31/12/21) and included both turf and all weather racing in the UK.

The focus is all race types (handicaps and non-handicaps) and all distances. I have not used a 'field size' restriction this time as around 95% of 2yo races had six or more (my usual cut off) runners anyway. I have explained the phrase 'run style' in the first two articles of the series but for new readers here is a very quick recap.

Run style is concerned with the position a horse takes up early on, usually within the first two furlongs of the race. Here on geegeez.co.uk run style is split into four categories as follows:

Led (4) – essentially those runners that get to the lead early
Prominent (3) – horses that track these early leader(s)
Mid Division (2) – horses that settle mid pack in the early stages
Held Up (1) – horses who begin their race near, or at the back of the field

The number in brackets is the run style score that is assigned to each section.

Run style is often linked with the word 'pace' because the early pace shown by horses in a race determines their early position. Hence, the words 'run style' and 'pace' are often used essentially meaning the same thing, though some commentators feel 'pace' is more associated with speed than racing position: this is why we differentiate. Each Geegeez racecard has the last four run style/pace figures for each runner within a table on the 'Pace' tab. That looks like this:

 

2yo horses may often have fewer data as some would not have run four times (indeed Clear Day in the example above has run only three times). This, hopefully, is where the trainer run style data shared below will prove its worth.

To help with this piece I have primarily used the Geegeez Query Tool – a tool that is available, and potentially game-changing, for all Gold subscribers. I then used my Excel knowledge to help crunch and interpret the data gathered.

Which trainers' two-year-olds led early most often?

To begin with, let us look at which trainers saw their 2yos take the early lead the most (in percentage terms). I have included trainers who have had at least 200 such runners over this 8-year period:

 

 

To offer some sort of comparison, the average percentage of all 2yo's that lead early stands at 14.6%. The trainers with the highest percentages are certainly worthy of further analysis.

(Charlie &) Mark Johnston

It is no surprise for regular readers to see Mark Johnston at the top of the pile, as we've previously discovered his modus operandi is typically to send horses forward. Nevertheless, it is an incredible statistic that more than 40% of his 2yos have led early. Mark is training with his son, Charlie, from the current season so it will be interesting to see if anything changes. I doubt it, but it is a good idea to keep an eye on such things. [Editor's note: at time of writing, the father/son Johnston team have led with 25 of 49 two-year-old runners, 51%, so little has changed at this stage]

Let's look at the Johnston stable breakdown in terms of percentage distribution across all four run styles:

 

 

Almost four out of every five of their 2yos either race prominently early or lead. To show how this bucks the general trend, compare Johnston’s stats to the overall 2yo run style stats for all trainers:

 

 

The real differences lie either end in the ‘led’ and ‘held up’ sections. This clearly illustrates how differently Johnston thinks about run style. If we look at individual years, we can see the percentage of his runners that lead has been consistent throughout:

 

 

The range, 36.1% to 46%, shows his methods have changed little over time.

In terms of race distance, we can see that in general it does not matter too much in terms of how likely a Johnston 2yo will lead. The breakdown is as follows:

 

 

It is only when we get to races beyond a mile that we see the percentage drop; even then, it is still very high when compared to other trainers.

The following table, sourced from the Pace Score section on the Query Tool, shows perhaps why the Johnston stable tend not to hold their 2yo (or indeed any age) runners up:

 

 

Hold up horses have been successful for the Johnston team just 5.5% of the time, with losses equating to just under 67 pence in every £1. That's not good for punters and, more materially from a training perspective, not good for owners. Meanwhile, early leaders won 26% of the time (incredible for owners) and would have made a profit if we had successfully decided upon which of his 2yos would actually lead (awesome for clairvoyant punters).

Archie Watson

Archie Watson is second in the standings when it comes to percentage of 2yos that took the early lead during the sample period. The most striking stats I found were when I looked at his record with 2yos that started favourite or second favourite (see below):

 

 

The differences are quite mind blowing. When we combine his 2yo's sent off in the top pair in the betting and that were held up or raced mid-division early, they produced just six winners between them from 62 runners; this equates to less than 1 in 10 winning. Watson's 2yos which led early and were top two in the betting won on average more than four times as often, at a 43% clip.

Which trainers' two-year-olds led early least often?

As we have seen earlier in this series, not all trainers are keen for their runners to take an early lead. Below is a list of the trainers with the lowest percentages in terms of horses that led early:

 

 

The eye is immediately drawn to James Fanshawe: just 1 of his 203 2yos have led early. It should be noted that Fanshawe has a relatively small crop of 2yos each year but, even so, this is remarkable. It is also worth noting that if a 2yo Fanshawe runner has raced prominently they have won 18% of the their races; compare this to the 4% win strike rate for his held up 2yos.

Some other well-known trainers appear in this table: the likes of Marcus Tregoning, Roger Varian and Roger (joined now by son Harry) Charlton to name but three. The Charlton data is worth expanding upon. Firstly let me breakdown his 2yo runners in terms of percentage of run style across all four run styles, as we saw earlier for Johnston:

 

 

A huge chunk of his 2yos tend to be held up, and nearly 65% of them have not been pushed up with or close to the pace early. Now look at the strike rates for each run style category:

 

 

It is the pattern we should all expect by now, but it begs the question why does Charlton hold up 43.8% of his 2yos when only 8.5% of them go on to win? Likewise why does he send just 8.4% of his 2yos out into an early lead when a huge 32.7% of them win? In general, it is likely to be that the Charlton runners may be incapable of getting to the front early, or that they are raced with at least one eye on the future; but the pattern is clear. Perhaps further schooling at the starting stalls might be beneficial.

Trainer run style averages

In order to give us a more complete picture, I have produced trainer run style averages, in exactly the same way that I did in the first article. To recap, I simply add up the Geegeez pace points for a particular trainer's two-year-olds and divide the total by the number of runners. The higher the average the more prominent the trainer’s horse tends to race. I have looked at overall pace averages rather than breaking down by handicap v non-handicap figures. The reason for this is that 79% of all 2yo races are non-handicaps. Also it saves some space!

For the record, the trainer run style average for all 2yos is 2.29. Have a look for your favourites below.

 

 

I have mentioned before that how you deploy these averages is personal choice. In 2yo races, especially when the horses have not run many times before, I believe the data can prove very useful. Let me give an example of a 2yo race run in April of this year.

 

 

As can be seen from the Geegeez PACE tab, only three of the horses had previously run and only once each. If we look at the trainer run style averages it looks likely that the Johnston runner will lead:

 

 

As the result below shows below, the Johnston runner Beautiful Eyes did lead, and also went onto win:

 

 

It is interesting to note that Karl Burke had the second highest number in the run style average table for this race, and his horse raced prominently and came second. Of course, the run style of all 2yo horses are not always going to correlate with the trainer averages. However, these averages can help us build up the most likely scenario of how the early stages of a race are going to be run even when horses have never raced before.

Here is a second example of a race from earlier this year, again it occurred in April:

 

 

Once again there was very limited run style/pace data from previous races to help form a picture of how the race may pan out in the early stages. The trainer run style averages for this contest were as follows:

 

 

Archie Watson comfortably had the highest run style average at 2.98, with David Evans earning the second highest. As it turns out the runners from these two trainers disputed the early lead and finished 1^st and 2^nd.

 

 

As I mentioned earlier this ‘prediction’ method won’t always work, but it is a useful starting point, particularly in 2yo races (or other race types where there is little no previous form).

Run style and market rank

To finish with I want to combine market rank with run style for the 2yo data from 2014 to 2021. The following graph looks at the percentage of runners that took the early lead in relation to their market rank:

 

 

What is clear from this strong correlation is that either market factors influence the running style of certain horses, or the running style of certain horses influences the market. Favourites led early in nearly 27% of all 2yo races in the eight year study period, almost double the average figure for early leaders of 14.6%. Horses occupying the next two places in the betting led in just over 20% of races but, as can be seen, once we get to horses outside of the top six in the betting, getting to the early lead was not easy for this group (less than 8% of them managed it).

This should come as no surprise. Less fancied horses in general are going to be slower than fancied horses, certainly over the full race distance; so it makes sense that this scenario is quite likely to occur early in the race as well as at the finish line. Of course, there will be occasions when an outsider is ahead of the favourite in the first furlong because trainer habits will have an effect or because the market has simply miscalculated the ability of a horse. Sometimes those horses will remain in front at the end of a race: shocks happen! But those are the exceptions.

Combining trainer run style data with market rank looks a potent combination. All Geegeez Gold users have the opportunity to dig even deeper than I have by looking at individual trainer run style statistics combined with market rank inside the Query Tool. To give you a taster, here are the top ten trainers in terms of percentage of runners which led early when sent off favourite (to qualify - 30 favourites minimum):

 

 

So Robert Cowell and (Charlie &) Mark Johnston favourites led more than half the time: that could be useful to know!

That's all for this episode. Please leave any comments, questions or thoughts below.

- DR

p.s. the next instalment of this series contains some of my most detailed research ever - stay tuned!