Sport is an industry of gradual, painstaking development. While received wisdom is passed down from coach to athlete from a young age, true development comes from not just talent but hard work and long-term commitment to improvement. At the top level, though, it’s only in individual sports that the obsession with incremental personal development seems to endure. If a tennis player improves their first serve percentage, they will see a reduction in fouls. If a footballer manages to shave a split second from their 100m time, though, the team won’t necessarily see any uptick in results - there are simply too many other factors at play.
This is not to say that team sportspeople disregard their personal performance statistics, there has just not traditionally been as much emphasis put on them. Rather, players have been talked about with regard to their ‘effect’ on games and in more vague terms like their ‘technique’ and ‘understanding.’ They will be intensely interested in their own numbers, for bettering them is how they can improve as players. But, with distractions like wider team tactics, on-field relationships, and the success of the team as a whole, among others, there is often little time for such close personal scrutiny.
It has long been an issue in sports analytics that, though there is keen buy-in among the coaching and sports sciences teams, encouraging the athletes themselves to engage with the figures isn’t easy. Sportspeople tend to learn through doing and watching rather than interpreting, and on the field of play there is often far more to think about than squeezing an extra point out of one particular physical metric. And so it is the job of coaching teams to present the data collected in a digestible, actionable way. In the same way that a group of players will switch off if asked to sit through hours of video footage of previous games, pushing a sheet of numbers in front of an athlete’s face isn’t likely to elicit passionate interest.
There are some sports for which data has long been a marker of quality, particularly those that have emphasis on isolated incidents rather than free-flowing action - think Baseball or Cricket compared to Soccer or Basketball. Ask a batter how good they are and they’ll cite their batting average. Ask how good a cricket bowler is and people will quote their bowling average. These figures might be basic, but they represent a fascination with numbers in sports that goes back decades, far further than the concept of big data. And now, with data so pervasive in transfer dealings, in punditry, in broadcasting, players would be wise to keep a close eye on theirs.
It’s the job of coaching and analytics teams to help them do this. At the Sports Analytics Innovation Summit in San Francisco last August, Carmelita Jeter spoke about the effect of data and particularly data visualization on athlete performance. Jeter holds the second-fastest 100m sprint time ever recorded by a woman, and got there by mimicking what she describes as a ‘stick figure’ on trainer Ralph Mann’s computer. Rather than presenting Jeter with comparably vague commands like raising the right knee slightly higher in her stride, for example, Mann was able to visualize that command and give it context. The simulation had been created to run the time of 10.6 that Jeter and her team were so desperate to hit, so by improving tiny parts of her sprint - angles, breathing, etc. - she would be able to match the data-calculated avatar.
Personal data becomes all the more interesting when you consider that it’s not just the physical that sports scientists measure. Psychological data is also measured to build a more complete picture of athletes. Psychological data is sensitive, though, so it’s important that sports science teams approach the topic carefully. The most comprehensive sports science teams will measure everything from mood, to learning styles, to whether the athlete is introverted or extroverted. When presented with the cause and effect of these factors in a digestible way, coaching teams can alter their training plans accordingly and even their match day tactics. It’s in this dissemination of information that analytics teams have the most important role. The athlete doesn’t necessarily need to know the reasons behind the change and, in the delicate case of psychological analysis, awareness might actually complicate the matter.
Another example is that sports science teams know that an extended warm up can often lead to improved motor learning for the future acquisition of new skills. This won’t apply to every athlete to the same degree, but sports science teams will know their athletes well enough to know if it is necessary. This is another case in which the athlete doesn’t need to know the science or the numbers behind the change in their routine, they just need to know the outcome and have faith that the sports scientists are acting in their best interests. In doing this, sports science teams can connect athletes with their data without bogging them down with it.