In Matthew McConaughey’s film, Two For The Money, McConaughey plays a sports consultant who makes it big predicting NFL winners for people to bet on. Things go wrong for him when he stops doing his research and starts relying exclusively on his intuition, and he steadily loses his fortune.
Predicting the outcome of sports matches has funded the careers of such so-called experts for years, many of them charlatans with slick hair and premium rate phone lines. But analytics could be putting an end to the days of half-baked guesswork based on hunches and supposed expertise, and turning the tide against bookmakers.
A number of firms, such as Swish Analytics and numberFire, are applying predictive algorithms across professional sports. Analytics usage in the arena now widespread, driven by pioneers like Billy Beane. Teams are subsequently now producing more data than ever, and much of it is readily available to fans. And sports betting analytics is set to keep growing. Currently worth $125 million, it is set to reach a valuation of $4.7 billion by 2021.
All of this additional information means that gamblers - and anyone looking to help them get an edge - are shifting away from the qualitative data that was traditionally used to predict outcomes, and they’re utilizing quantitive data instead. Swish Analytics, for one, is scraping every piece of information they can from every corner of the internet, in particular Twitter, and feeding it into prediction models to get a better idea of the likelihood of a result. They add to this real-time data, like the weather and umpires. By combining mathematically derived metrics with advanced algorithms, they find patterns that change the unstructured, and often misleading, data around sports into extremely accurate stats and predictions.
The use of analytics methods to gain an advantage is easier in some sports than others. American Football is still notoriously difficult to gain an edge in, having such a high number of variables that too much of it comes down to chance. With an oval ball, you can’t even bet on which direction it’s going to bounce in. NFL teams also only play 16 games a season, which means its sample size is smaller and it’s harder to find patterns in the data than in, say, baseball and basketball, which have longer seasons and far more games.
It is also important to remember that watching sport is still an important part of betting. What you see with your eyes is also data, and relying solely on analytics as a means of predicting an outcome, yet having no knowledge of the game, is likely to put you at a significant disadvantage over someone who does. Bookmakers are used to making money, and they are not likely to give this up in a hurry. They have access to the same predictions and metrics, and can lengthen and shorten odds accordingly to negate a better’s advantage. However, the number of games and the number of sports that they have to cover means that they are spread a lot thinner, and for a devoted shark, analytics can certainly give them the edge.