Data analysis in the world of sports
One could think that quantitative analysis has matured in the world of sports but that would be false, we are only at the beginning of what analytics can do for athletes and sports teams. We are currently gathering and analyzing only a small portion of the data that is available and we are looking at it in one dimension.
Data, data, always more data!
We are currently focusing our attention on the here and now of the performance of an athlete or a team. It’s a good start, coaches and sports organizations must be able to detect any change in a player’s performance, even those of their opponents so they can optimise the team’s overall performance. Analytics can play a role both before and after an athlete’s and a team’s performance.
As we can see on all sports media, there are a multitude of statistics on individuals and teams “describing” past results in various conditions: 2 games in 3 days, after travel time greater than 5 hours, an afternoon game, after a loss, etc. and that looks only at the team. Now multiply the amount of data by the number of players on a team and then add the data for the opposing teams. After all, the coaching staff wants to use every ounce of information in order to get a competitive edge.
Player statistics are compiled under every condition and their performance level is evaluated for each one of them. Analytical and predictive models can now be used by coaches and general managers to help them decide which team to field on the soccer field, the basketball court or on the ice.
Interpreting statistics to determine a plan of action
The advantage of “data” is not so much in how much you can collect but how well coaches can interpret the various statistics and KPI’s. Video analysis, RF captured data on individual player performance and game play statistics that are available in real time allow you to measure the performance of your players and those of opposing players.
They now have the ability to do real time analysis and, depending on their predictive models, modify which players to use and modify their tactics based on the weaknesses that they observed in the opposition’s players and team performance.
Analyzing optimal recuperation cycle for players and the best preparation approach will now be done on an individual basis and use individual data, all this to optimize the team’s overall performance level.
This is only the beginning of sports analytics. Just like the first chess programs that were limited by the amount of memory and CPU speed, we can imagine a sport team intelligence department that will be able to analyse the performance of a whole season offering different scenarios based on who they will face and any changes in a teams player personnel. For example:
- In case of injuries
- Modifications in player training approach to address below par performance
- Adapting their line-up based on their calendar
- Identify player trades and acquisitions based on predicted performance and team budget
So, will the next coach be called Watson, HAL or Claude?