Spottern is a low-cost data collection prototype that records accurate data on basketball players during training and matches.

Under certain circumstances, it is possible to analyze and optimize basketball players’ behavior and team’s performance, gathering advanced statistics that allows coaches to detect the team’s weaknesses and work on them on future training sessions. The system used by NBA is quite expensive and it requires a difficult implementation, that’s why Adriá Arbués and Enric Martos have decided to work on the first low-cost prototype analyzer.

Both Audiovisual systems engineering degree students at UPF, have designed a position sensor system of the size of a coin that tracks and records players movements during training and plays.

Employing an artificial intelligence that uses a total of 51 geometric characteristics (like the distance between players or the speed of shots) allows them to detect with a 98% of precision 5 classic basketball plays. This way, they can extract advance statistics, individual or collective.