THE use of data in horse racing and the general reluctance of the sport and its participants to embrace new data is something I’ve visited before, with the decision of the BHA to make breathing operations notifiable at the start of last year inevitably leading to the sort of debate which has me banging my head against a wall.

Many of those against the introduction of new data make a presumption of how that data will be used, which I like to think of as the ‘silver-bullet presumption’.

Just as a silver bullet is the only effective weapon in horror-folklore against werewolves and other monsters, so the term has come to mean a magical solution to seemingly complex problems.

Despite our desire to find such a simple answer to life’s problems, like the werewolf itself, the silver bullet is mythical, and solving problems takes work.

When new data, be it notification of gelding, other surgery, or even the use of a tongue tie – which was unpublished not so long ago – there is a flutter of hope among a small minority that having this information will make all traditional form study obsolete and enable the backer to become fat off his simple betting system.

We all know this simply doesn’t happen, but the reaction of those opposed to the requirement to provide such information is almost always to point out that the data does not constitute a silver bullet, that surgical procedures, like headgear only occasionally produce immediate returns.

Of course, they don’t think that gelding, or wind surgery, or the fitting of blinkers are purely random factors, or they wouldn’t advocate them, and that’s an important thing to bear in mind.

Gullible punters

While the gullible minority of punters who might believe the silver-bullet myth also happen to be the most vocal, that shouldn’t hide the fact that, like the Flat Earth Society who boast of having “members around the world”, they are a tiny minority and best left to their own fantasies.

Most of us, and this doesn’t simply apply to punters, but anyone who takes an analytical interest in horse racing, welcome new data and understand that the story it tells is a complex one, and that individual pieces of data are almost worthless without context.

The key to improving our understanding – of form, of behaviour, or the intricacies of breeding – is to provide more context by concentrating on big data. We are all accustomed to the small data that racing presents, and distance, class, going, ratings etc are all quantifiable factors we use to make selections, with little changing in that regard in the past half century. Big data doesn’t make any promise to relevance in terms of individual items of information, and pretty much none of us would know what to do with the information if horse weights were published in racecards tomorrow, for example.

Big data

The key to big data is in the name, and the value of analysing it is in the size and scope of that information. Sophisticated data-mining techniques can identify patterns and anomalies which would be impossible to spot without specific data which may, on its own, appear of little or no relevance.

In that regard, the relevance of the data is provided by the answers it throws up, and not on a subjective decision on the merit of the raw data. The scope of this goes beyond the information pertinent only to the horses, of course, and racecourses could benefit from data analysis to maximise attendance, crowd spend and betting turnover. Data could potentially identify welfare and safety issues, and much more besides.

Traditionally, we have sought to simplify racing by keeping the data small and specific, but other sports, even the lumbering traditionalist that is association football, are being transformed by an appreciation of big data.

Racing must follow, because pandering to the lowest common denominator in seeking to keep the sport relevant hasn’t worked and won’t work. The complexity of racing as a puzzle, and the inscrutability of the racehorse are two of the most attractive aspects of the sport, and we need to open that up, not shut it down.