A few years back I tracked every bet I placed for six months and discovered something humbling: my most profitable bets came not from horses I picked for form reasons but from a simple rule I had adopted almost by accident — always back a specific trainer’s runners at Cheltenham when the ground is soft. The trainer’s Cheltenham strike rate on soft going was 28%, versus 14% across all conditions. That one data point, applied mechanically, outperformed all my other analysis. Britain had 15,070 horses in training at the start of 2025, spread across hundreds of yards, and the people who train and ride those horses are not interchangeable — they carry statistical profiles that are genuinely predictive if you know where to look.
This article covers the specific trainer and jockey metrics that matter for betting, where to find them, and — importantly — when they mislead.
Key Trainer Statistics and Where to Find Them
A trainer’s overall strike rate — winners divided by total runners — is the starting point but never the whole picture. A yard with a 16% strike rate might look solid until you realise it is inflated by a few well-backed favourites while the rest of the string runs poorly. Context turns a number into intelligence.
The metrics that matter most for betting are: strike rate by race class, strike rate at specific courses, performance with specific jockeys, seasonal patterns, and results with first-time equipment changes (blinkers, tongue ties, cheekpieces). A trainer who runs at 22% in Class 4 handicaps but 9% in Class 2 tells you something about the level at which the operation is competitive. The 15,070 horses in training are not evenly distributed — larger operations with 100+ horses generate more data points, making their statistics more reliable, while smaller yards with 15-20 horses produce stats that are subject to far more randomness.
Course-specific data is where the real edges live. Some trainers target particular tracks with specific types of horses. A jumps trainer who sends only his best chasers to Sandown might show a 30% strike rate there against 12% nationally. The gap is not coincidence — it reflects deliberate planning and course knowledge. These patterns persist over years and are among the most exploitable statistical signals in racing.
Where to find this data: the Racing Post website provides free trainer statistics filterable by course, distance, going, class, and time period. At The Races and Timeform offer similar breakdowns. Most bookmaker apps now include a basic trainer strike-rate figure within the racecard, though the level of detail varies. For the deepest analysis, specialist databases like Proform and Raceform Interactive provide granular data that general sites do not.
Jockey Metrics That Matter for Betting
Jockeys influence race outcomes in ways that are harder to quantify than trainer statistics. A trainer prepares the horse; the jockey executes on the day. The two skill sets are different, and the statistical signatures are different too.
Overall strike rate is a useful quality indicator but varies enormously by the type of rides a jockey gets. A top flat jockey who rides primarily for a powerful stable might show a 20%+ strike rate because the horses are better, not because the jockey is dramatically more skilled than one showing 12%. The most informative jockey stat is how often they outperform the market — do their mounts finish ahead of what the odds implied? That metric, sometimes called impact value or adjusted performance, strips out horse quality and isolates jockey contribution.
Course-specific jockey data matters. Some jockeys excel at tight tracks (Chester, Pontefract) where tactical skill and track knowledge make a bigger difference. Others are strongest on galloping tracks (Newmarket, Ascot) where pace judgement in a straight-line finish is the key variable. A jockey booking at a course where that rider historically outperforms is a positive signal that the racecard alone will not tell you.
In National Hunt, jump-jockey statistics carry additional weight because jumping ability — both the horse’s and the jockey’s influence on it — is a genuine factor. A jockey who consistently delivers fewer falls and unseats per ride is not just safer to follow; they are systematically improving the probability that the horse completes the race, which has a direct effect on place percentages and each-way value.
Trainer-Jockey Combinations and Course Specialists
Certain trainer-jockey partnerships produce results that exceed the sum of their individual statistics. This happens when a jockey understands a particular trainer’s horses — their quirks, their preferred running styles, their optimal distance. The combination generates a win rate that reflects accumulated private knowledge rather than just public form.
Tracking combinations is straightforward with the right tools. Filter a trainer’s results by jockey, and compare the strike rate with that jockey versus the trainer’s overall rate. A 5-10% difference in strike rate is significant and suggests a genuine edge. I maintain a shortlist of about fifteen trainer-jockey combinations that I monitor throughout the season, and when one of them lines up at a course where both trainer and jockey have strong individual records, the bet becomes close to automatic.
Course specialists — trainers or jockeys who perform disproportionately well at a specific venue — are another exploitable angle. A trainer who sends 30 runners to Wetherby each season and wins with 25% of them (against a national average of 14%) has course-specific expertise that the market sometimes undervalues. The market tends to price horses based on overall form and market signals; course-specific trainer data is a niche that gets overlooked by casual punters and underpowered algorithms.
The booking of a top jockey for a horse at a minor midweek meeting is always worth noting. Top riders can choose their rides, and a champion jockey travelling to a Monday meeting at Wolverhampton when they could be riding at a more prestigious venue is a signal that the horse is expected to run well. It does not guarantee a winner, but it tells you the connections are confident enough to secure the best available rider.
The Limits of Statistics: Sample Size and Context
Every statistical edge in horse racing comes with a health warning: sample size. A trainer’s 40% strike rate at a particular course sounds impressive until you learn it is based on five runners. Two winners from five is random noise, not a pattern. The threshold for a meaningful trainer or jockey stat is roughly 30-50 data points — anything less should be treated as suggestive rather than definitive.
Context matters equally. A trainer’s course strike rate might reflect the quality of horses sent there, not the trainer’s skill at that venue. If the trainer only sends their best horses to a particular track, the strike rate will be inflated. Check whether the trainer’s runners at that course were well-backed in the market — if they were all short-priced, the high strike rate is simply a reflection of quality rather than an exploitable angle.
Recency also applies. A jockey who rode brilliantly five years ago might have declined. A trainer whose record was stellar in 2020 might be going through a quiet patch in 2026. Always weight recent data more heavily than historical figures, and be prepared to update your assessment as new information accumulates.
The most dangerous use of trainer-jockey stats is as a sole decision-making tool. Statistics should inform your overall picture — alongside form, going, class, and draw — not replace it. A horse with perfect trainer-jockey data but terrible recent form on unsuitable ground is not a good bet. The stats add value when they confirm or question what the form already tells you.
