Tag Archive for: horse racing stride length

New Data: Weird Experiment

Harsh Reality

There was a big horses in training sale at Newmarket this week. Late August seems a perfect time to acquire a three-year-old off the flat with a view to going hurdling and so I set to work on a mission to find such a horse.

Sadly, though obviously not surprisingly, I was not the only one at that public auction with that intention: the place was crawling with the great and good of the bloodstock agent and training ranks all seeking the pick of the sophomores to embark on a change of subject.

The reality of racehorse syndication is that there will always be people with deeper pockets; actually that's just a reality of life which is also reflected in the microcosm of the sales ring. But, as in life, so it is the case that some smaller owners are consistently 'lucky'. They make their limited budgets stretch far beyond wealthier purchasers for all that the very top prizes are still almost always out of reach.

Getting 'Lucky'

In recent times we've been 'lucky'. Luck, as Gary Player famously once articulated, is primarily a function of effort and persistence. So I try hard to find a route in which, if not exclusive to me (there's very little new under the sun, especially where such sums of money are concerned), is generally under-utilised. And then, importantly, I rely heavily on my data-driven research being underpinned by expert conformation eyes. When the data agrees with the eyes of experts, we want to play. Naturally, so do many others.

But sometimes horses slip through the net.

We bought Coquelicot, who was at the time already a half-sister to an Ebor winner and Melbourne Cup second and by a rising star of the stallion ranks (Soldier Of Fortune), for €26,000. She was a three-time bumper winner, most recently in Listed grade, in her first season.

We bought Windswept Girl, a 13-length winner of her only start to date, for £20,000. Both are exciting mares' novice hurdle prospects for this term and, because of their pedigrees, have long-range broodmare possibilities, too. We recently bought a Kayf Tara full sister to a Grade 2 winner (and twice G1 fourth) for £22,000. She is showing early promise and has a similar blueprint at a similar price point to her pair of proven predecessors in the geegeez livery.

Previously I've bought two three-year-olds to go hurdling. Both were sourced for me, though they had a similar style: not small, recent winners on the flat, likely or certain to stay, proven on softer turf. In the world of juvenile hurdlers, where so many are either too small or can't see out the trip, that's enough to win races; at what level is the remaining question.

Of that pair, Oxford Blu won a Fakenham juvenile hurdle on his debut for us by 20 lengths before running in the Fred Winter at the Cheltenham Festival. Swaffham Bulbeck was very consistent but didn't quite see out two miles and/or wanted softer ground than he typically got. He won twice, though, on Gold Cup day this year and last, both at Fakenham and both decent prize money. Importantly, they both gave us a lot of fun. That is pivotal when seeking a syndicate horse.

And so to Project Three-Year-Old Hurdler...

As ever, I was looking for an edge, and one thing which I think is not (yet) subsumed into this market is stride length and cadence. This is a new dataset that has emerged from the Total Performance Data sectional output. We store this data in our database but do not yet publish it. Attheraces already do publish it.

I'll not talk much about stride length and cadence because that master of matters sectional and striding, Simon Rowlands, has written some excellent introductory pieces, which are linked to from here.

The crux is that a longer stride covers more ground (duh!) and implies a bigger horse; a lower cadence (speed of stride) implies an ability to switch off - not over-race - and suggests stamina: it is very difficult to stride often and for a sustained period. Try sprinting 600 metres!

My theory, then, mindful of not having endless resources, was to find a biggish horse capable of relaxing in its races and therefore having the best chance of getting home when upped to two miles.

The Research, Part 1: What's Flat Got To Do With It?

One thing which seems fairly unclear at this stage is whether there is a correlation in National Hunt racing between stride data and performance. There may or may not be: hurdlers stride shorter and turn over their stride less frequently, due to the longer trip, the often softer ground, and the need to conserve energy to leap an octet and more of obstacles. Moreover, a greater proportion of jumps races are not run at an end-to-end gallop.

So what's the point in measuring strides?

Crucially, I was interested in the relationship between a horse's stride data on the flat and its subsequent hurdling ability. In other words, does a certain stride/cadence configuration imply a greater chance of success in the winter sphere?

Step one was fag packet research. Or, more correctly, back of an envelope research. I first looked at Class 1 and 2 UK juvenile hurdle winners.

C1/2 UK Hurdle Winner Flat Stride Lengths

C1/2 UK Hurdle Winner Flat Stride Lengths


I found that, for those which had raced on the flat at 10f+ (and for which stride data was available), they had generally achieved a 24ft+ peak stride length and somewhere between 2.2 and 2.3 strides per second during the flat race in question.

I then grabbed another envelope, a slightly bigger one, to review UK Class 4 and 5 juvenile hurdle winners' stride data from their previous flat careers. This is what that recycled paper looked like after I'd data vomited across it:

UK Class 4/5 juvenile hurdlers: flat race stride data

UK Class 4/5 juvenile hurdlers: flat race stride data


Here, I found that their cadence remained quite consistent at 2.2-2.3 strides per second; but their peak stride length was shorter. That might simply be as a result of these being smaller horses: regardless, it does seem at least a touch material.

The Research, Part 2: Projecting Forward

Mindful that a little knowledge can be a dangerous thing, I set to work in applying this half-cooked quackery to the upcoming - now just passed - sale. Armed with the timeform sales guide, a snip at £50 (well, if you're about to spend £30k on a horse...), the ATR striding data, and a spreadsheet, I listed every three-year-old that had achieved a TF rating of 75+, plus a couple of vaguely interesting extras.

[I eliminated some trainers from whom I would never buy because, for whatever reasons, they leave very little with which to work]

And then I added in some breeding intel, some ratings intel, and some stride intel.

Then, just for fun, I created a composite figure for each horse's official and Timeform ratings (added the two together and divided by 20).

And I created a second composite from stride length and cadence (stride length plus [[10 - cadence] x 2]) divided by five. Examples will help.

Camouflaged had an OR of 76 and a TF rating of 83. His ORTF composite, then, is (76+ 83) / 20 = 159/20 = 7.95

He had a stride length of 24.93 and a cadence of 2.21. His StrCad, then, is

(24.93 + ((10-2.21) x 2)) / 5 = (24.93 + (7.79 x 2)) / 5 = (24.93 + 15.58) / 5 = 40.51 / 5 = 8.102

Camouflaged's total was the sum of those two numbers, 16.05


These formulae are obviously sub-optimal. They're a stab in the dark on the basis of a tenuous going in position and, as such, they may be worthless. But this is what the embryo of progress looks like: data fertilised by an idea, loads of chucked away doodles and partially credible beginnings eventually whittled down to something which may form a basis for more robust and rigorous analysis.

To this crackpot code I added some delicious conditional formatting and it all looked a bit like this:

Then I Watched Some Races

The numbers, good or bad, will only take us so far. We must subsequently trust our peepers and our people. Eyes first.

I watched a lot of races involving the horses on my shortlist. I was looking for nods towards stamina, or horses being asked to do the wrong job, or racing with the wrong run style.

The challenge with those at the front is always whether they got outpaced or were simply not good enough. Honestly, I can't usually tell the difference. With those finishing off their races from further back, it's easy. Or at least easier.

To my spreadsheet I added some comments, which have not been sanitised below, so apologies for any offence caused:

The Sale Context

And then the sale began. I was buying this horse to be trained by Olly Murphy, and I relied heavily on his eye for conformation in conjunction with those on my list I considered more likely. That meant the most obvious horses - which would go for considerably north of my top budget - were not inspected.

The ones listed in green on my sheet were those I felt we had a chance of getting. As it turned out, they either went for plenty more or had underlying issues which meant they might be hard to train. Some will take a punt on such horses, and a subset of those punters will get lucky: consider Hilltop Racing's £2,500 purchase of Sceptical. He was a horse who hadn't stood training but managed to win a bunch of races and place in Group 1's in what turned out to be a short but illustrious career. He has since sadly perished, fatally injured while galloping.

Hilltop's modus operandi is thus - horses with great pedigrees, latent ability, but physical weaknesses - and it must be terribly attritional. That is their business, of course.

Looking for a biggish horse with a clean bill of health on a middling budget is tough. Here is my spreadie with the hammer prices and purchasers appended. I've removed some comments about physical issues observed.

What Did I Buy?

There was little of interest to me on the first two days, though I felt Far Rockaway and Chankaya might have been good. I wish their new connections luck but had reservations which precluded a bid.

The horse I liked on Tuesday made bundles - £82,000 - and might be decent. He kind of wants to be for that sum.

There was most to go at on Wednesday but, when my marker horse - Camouflaged - was knocked down for £90,000, I was resigned to the game being up.

The one that might have got away is Just The Ticket. It's possible, maybe even probable, I overrated his form a touch; and, if I didn't, his hammer price implies a potential physical issue. The agent who bought him buys for owners in the Arab states so I guess at least I'll not be frustrated to see him win the Triumph Hurdle..!

The horse we eventually bought was...

a four-year-old!

Makthecat was not what I went looking for at all; indeed he was already bought by Olly by the time I looked at his form. Almost all of his track form - a mile and a half clunk on the Southwell fibresand aside - has been at ten furlongs or less and ridden prominently or on the lead.

But... hark back to his career debut and his only run on turf beyond a mile and a quarter... where he finished second in a junior bumper... where he had a Listed bumper winner in front of him... and another one directly behind him! Six lengths back to the fourth.

Makthecat handles soft ground, looks likely to stay, and has a good size about him. He was bought for 22,000 guineas, and I am syndicating him right now. Current syndicate members get first dibs as always.

Oh, and he has a stride length and cadence 'in the zone': quite what that is worth, time will tell. I've added all the spreadsheet horses to a QT Angle so I can see how they go.


What About You?

This exercise took time, and it may ultimately have no nutritional value. But the process is important. Asking questions is important. Seeking candidate solutions is important. Trusting yourself to look for hints, clues, answers is important.

Some of you will not be curious, most will not trust themselves, or simply will not have time or inclination for such a project. Fair enough. But if you are curious, do trust yourself, and have both the time and the inclination to look inside the box, there are mysteries to unravel and there is value to be had, be it at the sales, in your punting, or in another walk of life entirely.

And, crucially, the process itself was great fun for me regardless of how things pan out. Life, as far as I can tell, is a series of journeys where the destination is often irrelevant.

Thanks as always for reading.


p.s. here's the new lad 🙂