Exploring Different Dimensions of Analytics: SPORT
Today we are talking about not the most obvious dimension of analytics — sports analytics.
Analytics help athletes score goals, coaches develop winning strategies, clubs build better squads, and federations run profitable competitions.
Of course, such a complex multi-level system cannot be based on simple statistics of goals, wins and losses. Expensive data is collected from a variety of sources, including wearable trackers, sensors in balls, photos and videos of matches, and medical records of athletes.
This data allows us to solve many different problems — below we have collected some striking examples.
1. ANALYSIS OF THE OPPONENT
Analysts gather information about the style of play, schemes and tactics of a future opponent and predict his expected behavior during a match or duel. Coaches use this information to develop the optimal strategy.
«Is it possible to predict the direction of a penalty kick taken by a particular player from the run-up before the kick? With what probability?» — analysts not only give an answer, but can also correct it right during the match, focusing on the incoming data.
2. PLAYER ESTIMATION
National teams constantly evaluate the performance of their athletes — and, of course, a simple number of goals and assists is not enough. To calculate complex indicators like the sequence of actions during the game, the level of efficiency and motivation, you need analytics.
«How do bonus payouts affect an athlete’s effort? What is the best payout amount for the club in the long run?» — simulation of various scenarios allows finding a balanced solution, avoiding real risks.
3. TRANSFERS
Analysts help clubs with recruiting and talent scouting. Based on the data, it is possible to hunt athletes who do not just show abstract good results, but are able to meet the needs of a particular team.
«Having a player with what characteristics would increase the team’s chances of winning the last N lost matches?» — information that is not so easy to extract from a bunch of unstructured data, but which can save the club millions of dollars.
4. INJURY MONITORING
By tracking athletes’ health and performance data, predictive analytics can help predict and even prevent injuries, as well as assess recovery time after injury.
«What is the probability that a football player with X screening results and Y style of play will be seriously injured before the end of the season? How to reduce this probability?» — It is interesting that even if the injury cannot be prevented, the information is still critical in terms of match planning.
5. COMPETITION PLANNING
Based on data on audience engagement and player motivation, sports leagues and federations develop balanced competitions: gambling for participants, interesting for spectators, and therefore profitable.
«Which matches cause the least interest among the audience? How to minimize the number of such matches?» — another task to search for causes / effects and predictive modeling.
6. STAND AGAINST DOPING AND CORRUPTION
Analysts identify the use of doping in situations where it is impossible either to detect it directly in the analyzes, or to catch the athlete in the act. The method is based on the search for abnormal deviations in medical indicators and sports results. By analogy, you can find potential fixed matches.
«Is everything clean here?» — Analysts cannot definitely answer such a question, but they can point out suspicious outliers in the data, thereby setting the direction for the investigation.
7. BOOKIES
Analytics for bookmakers is of particular meaning in this conversation, but it is impossible not to say about it. All in all, in betting companies, analysts solve the same task over and over again — they predict the results of the game and set odds for bets.
We are talking about predictive analytics based on a huge amount of accumulated information + dynamic data that appears during the game. Moreover, in the case of a bookmaker, it is important to exactly calculate not only the probabilities of outcomes themselves, but also the margin in order to win in any case.
Betting margins are the difference between the odds (an implied probability) the customer is offered to bet at, and the true probability of the outcome.
What does it take to become a successful and in-demand sports analyst?
In terms of the required competencies, sports analytics is no different from other industries: you need to be well versed in statistics, be able to work with Big Data and ML models — the rest depends on place of work.
Of course, it is also important to enjoy the sport and navigate it perfectly. And also understand that some vacancies involve working “live”, which means an irregular schedule.
Before you leave:
If you enjoy reading stories like these and want to support me, don’t forget to give me a few claps 👏, follow me ✅ and thus receive all updates 🔄 about new publications.
