Tom Seymour is the convenor of equine innovation platform Equate
The Pattern of Racing has always been a thorny issue. Every few months it resurfaces, and with it the same debate: how do we ensure a program that balances quality, commercial reality, respects traditional values and the long-term health of the industry?
The current focus on aligning the Pattern to ratings has merit as ratings provide a useful measure of performance. But on their own, they risk becoming too blunt an instrument. If prizemoney-driven ratings are the only determinant, we will tilt towards sprints and short-term commercial outcomes at the expense of balance. Other critical factors — age, sex, distance, breeding impact, wagering turnover, and the scheduling of lead-ups into any Black-Type races - all deserve equal consideration.
This is where Machine Learning (ML) could provide genuine value. ML has the power to analyse decades of historical data and model the cause-and-effect of different programming decisions:
> Wagering turnover trends across race types and distances.
> Breeding outcomes, showing how certain Pattern structures influence foal crop diversity and genetic balance.
> Scheduling effectiveness, testing how well Listed, G3, and G2 races prepare horses and optimise depth of pinnacle G1 events.
> Participation metrics, linking programming choices to new ownership, sales results, crowd attendance, media interest and field sizes as examples.
By running simulations, ML can surface insights that would otherwise remain hidden. For example: what happens to foal crops if the Pattern tilts further toward more BT sprint races? Or how do wagering returns shift if certain lead-up races are poorly placed in the calendar? ML can’t replace human judgement, but it can sharpen it - giving decision-makers a transparent, evidence-based framework to explain and defend their choices.
Beyond the Pattern
The Pattern isn’t the only area where ML could help. Welfare benchmarking, ownership retention, workforce planning, and cost of participation are all complex challenges where data-driven modelling could support better outcomes. ML can highlight unintended consequences before decisions are made, helping the industry to act proactively rather than reactively. More on this later ; )
Communication is Still Key
Of course, data alone isn’t enough. Even the best ML framework won’t work if decisions are still made behind closed doors. Transparency and dialogue remain the missing ingredients.
But imagine this: a round table where stakeholders are presented with scenarios, backed by ML-driven evidence. The debate shifts from “why didn’t we have a say?” to “which of these options is best for the industry?” That kind of process would build goodwill, consensus, and adoption - can you imagine!
The Opportunity
It’s easy to criticise. But the real opportunity is to modernise how we make decisions. Racing doesn’t lack smart people — it lacks modern tools and inclusive processes. The Pattern of Racing debate is just the latest reminder.
We don’t have to keep repeating the same arguments. With ML, decision-makers could test reforms, communicate them clearly, and build industry-wide support.
The choice isn’t between ratings or committees. It’s about combining data and dialogue to deliver reforms the whole industry can trust. We just have to live in hope someone will listen to us one day...
Tom Seymour is the convenor of equine innovation platform Equate
