Predicting Mark Johnston Runners, Part 1
When Poet’s Society delivered win number 4,194 for Mark Johnston in the Clipper Logistics Handicap to cement his position as winning most trainer in UK racing history, I read a number of tweets about the impossibility and frustration in predicting how his sizeable battalion of horses perform, writes Jon Shenton.
This rang true: having searched high and low in the past for potential Johnston angles I’ve found it hard to see the wood for the trees. Perhaps the sheer volume of runners essentially evens everything out? Surely though there must be something in that vast amount of data worth discovering? Armed with a renewed vigour I put the kettle on, fired up a database (the excellent horseracebase) and got to work.
The first question to ask, and answer, is, “where do you start?”. To me there is a certain logic in evaluating the yard by age groups, so in flat racing terms it makes sense to run the rule over Johnston’s ample contingent of 2-year-old runners in the first instance.
Two-year-olds
The volume of horses travelling through this yard from a young age is phenomenal. By my calculations there are on average around a hundred new 2YO debutants from the stable per year. Backing them systematically is a complete no go as that would result in a loss of 17% at SP, however if we look at distance as a differentiator, we start to get a different feel:
Mark Johnston debutant 2yo's split by race distance (1st January 2013 to 24th August 2018)
Distance | Runs | Wins | Win% | P/L (SP) | Place% | ROI (SP) % | P/L (BF) | ROI (BF) % | A/E |
5f & 5.5f | 145 | 30 | 20.7 | -6.9 | 40.0 | -4.8 | 10.0 | 6.9 | 0.94 |
6f & 6.5f | 163 | 28 | 17.2 | 4.2 | 36.2 | 2.6 | 40.9 | 25.1 | 1.04 |
7f & 7.5f | 156 | 20 | 12.8 | -35.9 | 27.6 | -23.0 | -18.4 | -11.8 | 0.81 |
1m & 1m.5f | 67 | 5 | 7.5 | -44.8 | 35.8 | -66.8 | -43.2 | -64.5 | 0.68 |
1m1f & 1m1.5f | 4 | 0 | 0.0 | -4.0 | 0.0 | -100.0 | -4.0 | -100.0 | 0.00 |
1m2f & 1m2.5f | 3 | 1 | 33.3 | -1.2 | 66.7 | -39.0 | -1.1 | -36.3 | 1.49 |
Immediately we can see that the 5-6.5f debutants more or less break even to SP with a reasonable return to Betfair SP. In my experience this is usually a good starting point for further investigation and with BOG on the table there would be many worse ways to wager than backing these all blindly. However, I very much doubt you’re reading this to more or less break even, so let’s see if there is a better indicator in the data to improve punting performance.
A key component to consider with trainers often is the gender of the horse, in fact Geegeez’s own Tony Keenan appraised Irish Trainers and their relative performance with fillies, here.
Assessing Johnston’s debutants by gender is very interesting, as the table below indicates:
Mark Johnston’s debutant 5f-6.5f runners by gender (1st January 2013 to 24th August 2018)
Gender | Runs | Wins | Win% | P/L(SP) | Place% | ROI(SP) | P/L(BF) | ROI(BF) | A/E |
Colt | 169 | 23 | 13.6 | -74.0 | 33.1 | -43.8 | -62.6 | -37.0 | 0.7 |
Gelding | 2 | 1 | 50.0 | 6.5 | 50.0 | 325.0 | 11.2 | 559.0 | 2.7 |
Filly | 137 | 34 | 24.8 | 64.8 | 43.8 | 47.3 | 102.3 | 74.7 | 1.3 |
In anyone’s book that’s quite a variance between colts and fillies (despite the 325% ROI I think with a sample size of two, we can ignore the geldings). A debut MJ filly wins a quarter of the total runs and returns nearly 50% on SP, and that’s very much backing blind territory if you are that way inclined. Certainly such types are worthy of marking up when you’re betting in a race containing a Johnston debut juvenile filly.
Digging deeper, backing those fillies from May to August improves the picture, the runners early in the season generally underperforming, as do many fillies at that time of year.
Mark Johnston 2yo first time out fillies running May-August between 5f-6.5f in distance
Year | Runs | Wins | Win% | P/L(SP) | Place% | ROI(SP) | P/L(BF) | ROI(BF) | A/E |
ALL | 98 | 26 | 26.5 | 79.3 | 45.9 | 81.0 | 115.5 | 117.8 | 1.5 |
2018 | 18 | 6 | 33.3 | 5.8 | 66.7 | 32.2 | 7.3 | 40.6 | 1.3 |
2017 | 20 | 5 | 25.0 | 39.0 | 35.0 | 195.0 | 60.0 | 299.8 | 1.7 |
2016 | 24 | 4 | 16.7 | 5.8 | 33.3 | 24.3 | 10.3 | 42.7 | 1.1 |
2015 | 17 | 6 | 35.3 | 18.5 | 52.9 | 108.5 | 22.3 | 130.9 | 2.1 |
2014 | 10 | 4 | 40.0 | 16.5 | 70.0 | 165.0 | 22.0 | 219.6 | 2.4 |
2013 | 9 | 1 | 11.1 | -6.3 | 22.2 | -69.4 | -6.3 | -69.8 | 0.7 |
The same underperformance applies to horses making their first strides in the autumn, which may be being geared up for 3YO handicaps the following season and potentially running to get a mark (more on those unexposed types later).
So, by backing Johnston first time out fillies over 5-6.5f from May to August a return on investment of 81% would have been achieved; just for clarity this covers both Maiden and Novice races.
Looking beyond the first-time out angle, considering novice races only (Johnston’s record in nursery handicaps is par at best so ignored for the purpose of this article) there is more of potential interest.
Since the expanded novice programme was introduced in 2016 to encourage more sightings of maiden race winners, it is worth a check to see if Johnston is making use of these races by evaluating the performance of his previous winning horses. The table below shows those runners in novices that have got their noses in front in their fledgling two-year-old careers.
Mark Johnston novice runners from 2016 to date with a previous career win
Runs | Wins | Win% | P/L(SP) | Place% | ROI(SP) | P/L(BF) | ROI(BF) | A/E |
92 | 36 | 39.13 | 18.4 | 66.3 | 20.0 | 26.9 | 29.2 | 1.13 |
Three-year-olds
Considering the Classic generation, it was actually quite hard to find a robust angle but there is potentially something in evaluating unexposed horses in handicaps with very little in the way of solid form. Taking all of Johnston’s 3yo runners in 3yo only handicaps we have the following:
Mark Johnston 3YO runners in 3YO only handicaps from 2013 to date
Runs | Wins | Win% | P/L(SP) | Place% | ROI(SP) | P/L(BF) | ROI(BF) | A/E |
1736 | 262 | 15.09 | -316.7 | 32.4 | -18.2 | -154.5 | -8.9 | 0.9 |
That overarching dataset is miles away from a profitable or even remotely interesting angle; however, if we look for potential lurkers, horses that the market may overlook due to previous underwhelming performance there might just be a sliver of light to work with.
Mark Johnston 3YO runners in 3YO handicaps by number of career wins from 2013 to date
Career wins | Runs | Wins | Win% | P/L(SP) | Place% | ROI(SP) | P/L(BF) | ROI(BF) | A/E |
0 | 405 | 70 | 17.3 | 11.4 | 35.8 | 2.8 | 46.8 | 11.6 | 1.05 |
1 | 619 | 91 | 14.7 | -184.8 | 31.5 | -29.9 | -121.1 | -19.6 | 0.84 |
2 | 384 | 69 | 18.0 | -31.2 | 35.2 | -8.1 | 8.1 | 2.1 | 1.03 |
3 | 224 | 25 | 11.2 | -54.5 | 28.6 | -24.3 | -33.2 | -14.8 | 0.79 |
4 | 82 | 5 | 6.1 | -44.1 | 23.2 | -53.8 | -42.1 | -51.4 | 0.51 |
5 | 20 | 1 | 5.0 | -16.5 | 20.0 | -82.5 | -15.9 | -79.7 | 0.44 |
6 | 2 | 1 | 50.0 | 3.0 | 50.0 | 150.0 | 3.1 | 154.3 | 2.13 |
7 | 1 | 0 | 0 | -1 | 0 | -100 | -1 | -100 | 0.00 |
The top line is interesting. Horses that have no previous career wins running in a 3YO handicap for Johnston are marginally profitable if you back them all, and they have a respectable 17% strike rate too. There are several ways to potentially sharpen from this starting point. However, to stay with the unexposed theme, evaluating the horses’ total number of career runs might be considered a logical way to delve in a bit deeper; fewer runs should mean less predictable?
Mark Johnston 3YO handicap runners from January 2013 with no career wins, by number of career runs
Career runs | Runs | Wins | Win% | P/L(SP) | Place% | ROI(SP) | P/L(BF) | ROI(BF) | A/E |
3 | 91 | 19 | 20.9 | 8.0 | 37.4 | 8.8 | 16.7 | 18.3 | 1.21 |
4 | 87 | 21 | 24.1 | 45.9 | 40.2 | 52.7 | 56.8 | 65.3 | 1.41 |
5 | 60 | 11 | 18.3 | 16.0 | 36.7 | 26.6 | 26.6 | 44.3 | 1.03 |
6 | 51 | 7 | 13.7 | -15.5 | 27.5 | -30.4 | -12.5 | -24.5 | 0.88 |
7 | 37 | 7 | 18.9 | -0.1 | 37.8 | -0.2 | 1.8 | 5.0 | 1.22 |
8 | 29 | 1 | 3.5 | -27.4 | 34.5 | -94.4 | -27.6 | -95.3 | 0.23 |
Maybe fewer runs = more predictable! No career wins, fewer than 6 career runs, 30% profit thank you very much. Even if ploughing in on them all isn’t for you it’s a smart move to put an unexposed Johnston 3YO on your shortlist or at the very least be wary of them if you have another fancy in the race. This chart shows win strike rate, and ROI (SP/BSP) by number of previous starts for Mark Johnston maiden three-year-olds.
Those with more than eight career runs have a 4/50 win record and represent an incredibly poor way of investing.
Possible angles and betting opportunities:
- Johnston First Time Out 2-year-old fillies over 5-6.5f, especially during the months of May to August
- Previous 2yo winners in novice races if they have had a recent run throughout spring and summer
- Winless 3yo runners with five or fewer career runs in 3yo handicaps
In part two of this article, I look at older horses and a course where the 'Always Trying' bandwagon are particularly potent. You can read that here >
- Jon Shenton
Its also worthlooking at his travellers down to Brighton, Lcr and such like courses. He seems to target the “best” races at these group b courses. Jockey bookings can also help.Franny at GW and Chester seems to pay his way. Also bear in mind that he has 3 or 4 assistant trainers looking after there squads so I havent a clue how that works out
Welcome to the Geegeez blog Jon – a great opening piece. Love reading articles like this, and, (take this as a compliment) I must say you have a certain “Bisogno-esque tone” in the way you deliver your findings 🙂
Thanks again, and I look forward to the next instalment!
Thanks Chud, appreciate that and glad you like it, hopefully more to come!!
Thanks Jon, that’s a brilliant and informative piece of research. Like many others I’m sure, I’ve tried to get a handle on MJ over the years without much success. The nearest I came was when Fanning was riding for him on certain tracks, can’t remember which now, it was profitable for a while but didn’t stand the test of time.
I’m an Annual Member at Ripon so I picked up on this anomaly fairly quickly, from 2009 for six years, his 2yos at the track just didn’t win, or to be honest just one did. 2009-14 he sent 45 2yos to Ripon for just one winner, SR 2.2%. Over the same period his overall 2yo SR was 13.06%, and he sent 75 2yos to Beverley for 18 winners, SR 24%. I was never brave enough to lay them, these things can change quickly, and there didn’t seem an obvious reason for their lack of success, but they were certainly the first off the shortlist, which helped.
It all changed in 2015 when he acquired more and better quality 2yos, they started winning at Ripon again, but whether the previous 6 years were a statistical coincidence or a deliberate strategy by the MJ yard I still have no idea.
Thanks sinkov, appreciated. He certainly seems to change things up, when I was looking at course performance Beverley did pique my interest as he has an overall profitable record there but it’s fallen off a cliff (relatively speaking) over the past 3 summers. Always a challenge trying to recognise whether it’s a bad patch in a trainer based system or a change in trainer behaviour. If I crack it I’ll be a bit richer!!
A well researched piece and a good read (nap).