The market consistently backs horses at the wrong distance. Talented animals are campaigned over trips that don’t suit their biomechanics. Punters follow form without questioning whether a horse is even racing at their optimal distance.
Horse racing stride analysis reveals what traditional form analysis misses: whether a horse is physically suited to the trip they’re running. More importantly, it identifies which horses will improve at different distances—before form proves it and prices move.
That was the question I set out to answer. What began as a straightforward attempt to forecast racing distance evolved into something more significant—a method for finding value before the market catches on.
The foundation was already being laid. Total Performance Data pioneered stride capture technology. Horse racing stride analysis was moving from theoretical to practical. Simon Rowlands brought expert analysis to the mainstream. Kevin Blake famously predicted the first three home in the 2025 Epsom Derby using stride analysis.
But publicly available distance prediction models? The last credible attempt was the Dosage Index—developed by Dr. Steven Roman in the 1980s, based on earlier work by Lt. Col. J.J. Vuillier. It used pedigree data to forecast distance aptitude. It’s dated. Yet it’s still used today.
Could stride data provide a better way forward?
The short answer: yes.
The evidence suggests horse racing stride analysis remains significantly undervalued by the market. Horses are regularly campaigned at unsuitable distances while their stride profiles indicate they’d be more effective elsewhere. The market hasn’t caught up yet.
Understanding Horse Racing Stride Analysis: SPS and Distance
The basic premise is straightforward: horses with lower stride frequency stay further. Those with higher stride frequency suit shorter distances.
Stride frequency is measured in strides per second (SPS). Specifically, minimum stride frequency—the slowest rate at which a horse turns their legs over during a race—is the key indicator.
Here’s what the data shows across 348 horses:
| Distance Range | Typical Min SPS | Median |
|---|---|---|
| Sprint (5f→7f) | 2.28 - 2.37 | 2.33 |
| Miler Plus (7.1f→10f) | 2.18 - 2.28 | 2.23 |
| Middle Distance (10.1f→13f) | 2.10 - 2.17 | 2.13 |
| Staying (13f+) | 2.07 - 2.13 | 2.09 |
The pattern is clear and consistent. Lower minimum stride frequency correlates with increased stamina.
For horses running 12 furlongs, you want minimum SPS below 2.17—ideally 2.13 or lower. Horses above 2.20 are operating in miler plus territory, not genuine staying ground.
The Challenge: Early Models Didn’t Work
Initial attempts to model the relationship between stride frequency and stride length produced poor results. Using just these two variables yielded an R² of less than 20%.
In statistical terms, that means less than 20% of a horse’s optimal distance could be explained by stride data alone. The remaining 80%+ was noise. Unusable for prediction purposes.
Something was missing.
The Breakthrough: Race Pace Changes Everything
The single biggest improvement in accuracy came from introducing pace into the equation.
I’ve long believed that after raw ability, race pace has the greatest impact on outcomes. Efficiently run races—where energy is distributed logically—produce reliable results. Slow-then-sprint races, or races where horses empty the tank too early, create chaos.
The same principle applies to stride analysis.
What happens to stride data when a race is run inefficiently?
Stride cadence and stride length distort. They no longer reflect a horse’s true biomechanical profile. You’re measuring race tactics, not physiology.
Case Study: Delacroix and The Cashel Palace Trial
A high-profile example: Delacroix in the Cashel Palace Hotel Derby Trial Stakes at Leopardstown—his final prep run before the 2024 Derby.
In that race, his predicted optimal distance jumped to 11.5 furlongs, putting him on the cusp of getting the Derby trip.
But here’s his full profile from debut through to the trial:
| Date | Course | Distance | Predicted Optimal at 3yo |
|---|---|---|---|
| 25 Jul 2024 | Leopardstown | 8f | 9.6f |
| 10 Aug 2024 | Curragh | 7f | 8.3f |
| 14 Sep 2024 | Leopardstown | 8f | 9.1f |
| 12 Oct 2024 | Newmarket | 8f | 9.8f |
| 26 Oct 2024 | Doncaster | 8f | 8.9f |
| 30 Mar 2025 | Leopardstown | 10f | 9.5f |
| 11 May 2025 | Leopardstown | 10f | **11.5f** |
Every previous run clustered between 8.3f and 9.8f—classic mile to 10f territory. Then one outlier at 11.5f.
What happened at Leopardstown?
Watch the race back. The on-screen speed data tells the story. They crawled through the early stages, then sprinted home from three furlongs out. Delacroix’s stride analysis in that race didn’t reflect his true biomechanics—it reflected race tactics.
He went off 2/1 favourite at Epsom and was beaten 16 lengths.
He subsequently won the Eclipse Stakes—another slowly-run race where his explosive late speed was decisive. He followed up in the Irish Champion Stakes, again thriving on tactical pace and a blistering turn of foot.
Delacroix is a brilliant horse. But he was never a mile-and-a-half horse. His stride profile made that clear. The Cashel Palace trial created a false signal.
The Solution: Quality Data Only
Using stride data exclusively from efficiently run races transformed the stride analysis models. R² jumped to 75%+.
In statistical terms, that means the model now explains over three-quarters of the variance in optimal distance—a substantial improvement and a strong foundation for biomechanical prediction.
The key learning for stride analysis: Input quality determines output reliability. Stride data from truly-run races dramatically improves prediction accuracy. Data from tactically distorted races should be treated with caution.
A useful byproduct: The range between minimum and maximum stride length within a race serves as a pace indicator. Competitive horses showing a wide range—low minimum, high maximum—signal slow-early, sprint-late tactics. Delacroix’s Cashel Palace race showed a stride length range of 1.38 metres, a clear red flag.
Later iterations of the model incorporated a confidence factor, flagging predictions that warranted further scrutiny beyond the raw output.
Bottom line: treat inefficiently run races with scepticism.
The 2-Year-Old Challenge: Racing in The Wrong Context
Modelling 2-year-old development presented unique difficulties.
The problem: Physically immature 2-year-olds often race at distances that don’t showcase their natural stride characteristics.
Why this happens:
- Limited race opportunities – Relatively few 2yo races exist beyond 8 furlongs
- Trainer strategy – Getting three quick runs for a handicap mark, often at suboptimal distances
- Market demands – Early-season sprints dominate the 2yo calendar
- Physical immaturity – Stamina horses aren’t yet ready for longer distances
Example: A future 10-furlong 3-year-old forced to race at 6 furlongs as a 2-year-old. Their stride data at 6f won’t reflect their true biomechanics. They’re racing in a sprint context when they’re built for stamina.
The solution: Don’t rely on stride data alone.
Predicting 3yo optimal distance from 2yo biomechanics requires combining:
- Stride frequency and length
- Race pace analysis
- The distance they actually raced at as a 2yo
This combined approach accounts for horses “racing in the wrong context.” The model recognises that a stamina horse showing certain biomechanics while racing short distances will likely want significantly more trip at 3yo.
This approach improved the 2yo model to 78.7% R²—the most accurate iteration.
Maturity Changes Everything: The 7-Furlong Breakpoint
Conventional racing wisdom says: “This horse will want further with another year on its back.”
The assumption is simple: maturity equals increased stamina.
The data revealed something more nuanced.
Physically, horses grow stronger and more athletic as they mature. But this growth doesn’t affect all horses uniformly. Maturity manifests differently depending on a horse’s biomechanical profile.
Enter the 7-furlong breakpoint. Horse racing stride analysis revealed this critical insight through data from hundreds of horses.
Horses can be broadly divided into two groups based on their optimal distance:
Sprinters (<7f): Primarily anaerobic energy users
Middle-distance and stamina horses (>7f): Increasing reliance on aerobic energy systems
As horses mature from 2yo to 3yo and beyond, the direction of development splits:
- Sprinters (<7f) Maturity translates into speed.
- Stamina horses (>7f) Maturity translates into increased stamina.
The 7-furlong mark sits roughly at the transition point between these developmental paths. This pattern, validated across 500+ horses, reveals why some 2-year-olds develop stamina while others develop speed—read the full analysis of the 7-furlong breakpoint.
Why Does This Happen? Energy Systems Explained
Horses produce energy through two distinct systems:
Anaerobic Energy (Speed)
- Burns fuel without oxygen
- Creates explosive power
- Depletes quickly—think sprinting
- Dominant in races under ~7 furlongs
Aerobic Energy (Stamina)
- Burns fuel with oxygen
- Sustainable over distance
- Slower but efficient
- Dominant in races beyond ~7 furlongs
As horses mature, they don’t develop uniformly across both systems. Physical development tends to favour one energy system over the other:
Speed horses optimise their anaerobic capacity. They become faster sprinters, developing explosive power over shorter distances.
Stamina horses develop their aerobic systems. They sustain pace over longer distances, building cardiovascular efficiency.
The 7-furlong mark represents the rough transition between these energy systems. Horses naturally gravitate toward optimising one system as they physically mature.
This is why conventional wisdom—’all horses want further with age’—is incomplete. Only stamina horses want significantly more distance. Speed horses stay at sprint distances and get faster.
Minimum Stride Frequency: The Stamina Indicator
Why does minimum frequency in horse racing stride analysis correlate so strongly with stamina?
Because a horse’s breathing is synchronised to their stride—one breath per stride.
Lower stride frequency means more time per stride. More time per stride means more oxygen intake. More oxygen intake means better aerobic capacity.
Horses racing at low minimum stride frequency are operating aerobically. They’re conserving energy, managing oxygen intake, and preparing for sustained effort over distance.
Horses racing at high minimum stride frequency are operating anaerobically. They’re burning through fuel quickly, producing speed, but with limited endurance.
For stamina horses, the ability to “rev down”—to slow their stride frequency through the middle stages of a race—is essential. It allows them to:
- Maximise oxygen intake
- Reduce lactic acid buildup
- Conserve energy for late acceleration
This is why minimum SPS below 2.17 is important for longer distances. Horses above that threshold can’t slow their stride enough to run aerobically for extended trips. They’ll burn out before the finish.
Speed horses aren’t worse athletes. They’re just optimised for a different energy system. Asking them to stay 12 furlongs is like asking a 100m sprinter to run a marathon. Wrong tool for the job.
Putting It Into Practice
Next time you’re analysing a potential stayer, check their minimum SPS:
- Above 2.20? They’re operating in miler-plus territory. Treat stamina claims with scepticism.
- Between 2.13-2.17? Borderline. Could get middle distances but monitor carefully.
- Below 2.13? Genuine staying profile. These horses are built for longer trips.
Regardless of form, pedigree, or trainer comments—horse racing stride analysis tells you whether the engine is suited to the distance.
This isn’t a replacement for form analysis. It’s a filter. Use it to eliminate horses racing at unsuitable trips before committing to detailed analysis.
Ground Conditions: Less Important Than You’d Think
When developing the horse racing stride analysis model, I started with assumptions. One was that ground conditions would significantly impact predictions—horses stride differently on firm ground versus soft, right?
They do stride differently. But it barely affects optimal distance predictions.
The data:
- 3yo model: 76.6% accuracy on all ground vs 74.9% excluding soft ground
- 4+yo model: 75.9% accuracy on all ground vs 76.1% excluding soft ground
These differences are statistically negligible. The models work equally well regardless of going.
What this suggests: A horse’s fundamental biomechanics transcend surface conditions. Optimal distance is determined by physiology, not footing. A 10-furlong horse on firm ground is still a 10-furlong horse on soft ground.
Putting Stride Analysis Into Practice
Horse racing stride analysis offers predictive insight that the market still undervalues.
For punters: Stride analysis identifies horses racing at unsuitable distances before the market adjusts. It spots genuine stayers at generous prices while the market backs flashy 2yo Group winners who won’t get extended trips.
For trainers and owners: It informs campaign planning, race targeting, and realistic distance expectations before committing to unsuitable engagements.
For analysts: Stride analysis provides a framework for understanding performance that goes beyond traditional form metrics.
The Bottom Line
Markets regularly misprice horses based on distance suitability. Talented animals campaign over trips that don’t suit their biomechanics. Meanwhile, genuine stayers trade at inflated odds because they haven’t “proven it” yet at shorter distances.
Stride analysis isn’t a magic formula. Predictions will be wrong. Horses will surprise. But it’s a significantly more robust approach than waiting for horses to prove their distance the hard way—by which point the market has already adjusted and the value has evaporated.
The models work. Over 75% of distance variance can be explained through biomechanical analysis. The patterns are consistent. The correlations are statistically significant.
The edge exists because most of the market isn’t using it yet.
Traditional punting relies on pedigree, trainer reputation, and performance at different distances—all lagging indicators. By the time a horse “proves” their optimal trip through results, everyone knows. The price has collapsed.
Horse racing stride analysis gives you forward visibility. You can identify stamina horses before they’ve raced beyond a mile. You can spot milers being aimed at unsuitable staying races. You can filter out horses with no biomechanical chance of getting the trip.
That’s the opportunity.
The data is there. The patterns are clear. The question isn’t whether stride analysis works—it’s whether you’ll use it before the rest of the market catches up.
That window won’t stay open forever. The question isn’t whether the market will catch up and prices move—it’s whether you’ll act before it does.
See this methodology in action with our ante-post Derby 2026 and Oaks 2026 predictions, , or read our weekly Stride Watch series for ongoing biomechanical analysis.