How important is a good serve in women’s tennis? By how many percentage points does a player need to improve her serve to climb a position in the WTA ranking? Does it take a minimum number of cross-court slice backhands to beat a specific rival? The answers to these and other questions that anyone, from fans to tennis players and their coaches, may ask are nothing but the result of measurements that are very easy to take. Current Big Data technology is capable of this and much more and, properly applied, could play a decisive role in winning games, changing training plans, or tactics in real time and contributing to decision making processes.
A sentence like ‘the flapping wings of a butterfly can be felt across the world ‘could prove a valid metaphor for this if the tennis-equivalent to a light flutter – say, for example, increasing the annual percentage of ‘breaks’ or first serves won by one percentage point – didn’t require hours and hours of hard work and complex training, and the whole process was a bit more reminiscent of the smoothness of that motion. However, having predictive models at hand will help prioritize certain aspects of the game, based on the surface on which players are going to be competing or depending on the rivals we have to face in a specific tournament.
An anecdotal example may be the best way to shed some light on the world of possibilities that the Big Data technology enables,.
Below we can see how Big Data correlates the WTA top 20 (as of October 26th, 2015) with these other ‘re-rankings” of the top 20 players based on: Good first-serves, points won with first serve, points won with second serve and ‘break points’ saved so far in 2015. (Data taken from the WTA’s website)
The stats that best correlate to the WTA ranking are the percentage of second-serves won and the percentage of break points saved: 7 of the top players ranked in these two categories are ranked in the WTA’s top-10. Based on this, we’ve seen that in just a matter of minutes, one could conclude that those who are able to increase the power of their second serve (players have very similar %), while keeping their percentage of good serves, will have an advantage over their rivals, just as those that work on improving their performance in stressful situations such as when their rivals have the chance to ‘break’ their serve.
But there’s more. Because when combining these classifications, we see that Serena Williams and Maria Sharapova rank in the top-5 in three of them. Analyzing many more data from many more years would certainly help understand fundamental key aspects to decide on the right strategy when preparing a match against a specific rival. But there’s even more. Because this massive amount of data can help to gain a technical interpretation of the reasons behind the ascent of wonderful Garbiñe Muguruza. And her rivals’ coaches should be working on this right now.
In the WTA, unlike in the ATP, the coach is allowed to speak to his/her player between game and game, and therefore a powerful app capable of reading and analyzing the stats of each player in a match in real time would prove an invaluable help for an experienced coach at the time of giving advice to her pupil to beat her opponent, and, subsequently, climb in the WTA ranking. Also, in a future, coaches and players may be allowed to communicate via headset. Real time communication with real time application.
Data analytics, a frequently used tool for sport-teams in the US
Although database analytics are not used so extensively in individual sports, they have been used in team-sports for years. And not only for the sake of churning out numbers alone, but also for using them to develop in tactics and training plans. The US was the first country to do so. The film “Moneyball” did a great job offering a glimpse of how teams have been using data in recent times to broader audiences. The movie is based on a novel and tells the story of a baseball team that starts relying more on stats than on the intuition of the club’s veteran advisers.
But maybe, being the most recent and factual, the best example is the case of the Houston Rockets and the mid-range shots. There it goes: Daryl Morey, the team’s current manager, has been a data analytics diehard for years, but especially since his strategic consultant days at The Parthenon Group. The thing is that last season, data analytics ‘banned’ players to take mid-range shots near the three-point line: They were an extremely poor investment, due to the high percentage of errors! After players took on this mandate, they would either do all they could to take paint shots or take a three-pointer.
The Houston Rockets finished the season setting a new 3-pointer record, netting 894 from January through the team’s end of the season: the Western Conference finals, which they played marred by a plague of injuries.