In the Tokyo 2020 men’s basketball match between the United States and France, a robot stole the show which – entering the court during the break of the match – put on a show, hitting the basket not “only” from the line and from the line of the 3 points but also from midfield. Created by Toyota engineers, the “Cue” robot basketball player (of which over time have been released updated versions) has already set the Guinness World Record for the number of baskets made in succession: 2020 hits in 6 hours and 35 minutes.
This is the comment of Chris Matthews, known as the “lethal shooter”, among the best basketball players in the world: “Cue shows the impressive technology of Japan. He is a robot that is anything but normal. He doesn’t make mistakes almost never and, if it does, it self-calibrates for the next shot.” This is the hope of Tomohiro Nomi, who leads the development team behind Cue (already in the ranks of the Japanese club Levanga Hokkaidothe robot is used during halftime of games to entertain the public and show the developments of artificial intelligence in basketball): “It would be nice if it could do Michael Jordan’s dunk on a free throw.”
AI applied to basketball
From abroad to Italy, where there are more and more sports clubs that are applying the technology created by TwinPlay. The “made in Italy” startup has in fact developed – the first in our country – a software which, using a camera, exploits computer vision (this area of AI, explains Ibm in one article, “uses machine learning and neural networks to teach computers and systems to derive meaningful information from digital images, videos, and other visual inputs”). Thus, through intelligent algorithms and cameras, coaches and players can process advanced statistics extrapolated during training sessions. Acquiring useful data to optimize the game plans to be implemented during matches.
Davide Todeschini, founder of TwinPlay, explains: “The system is able to track the movement of 8 players at the same time. The parabolas of the basket shots appear graphically on the computer monitor with the calculation of the angles, the release time of the ball , up to the three-dimensional reconstruction of the trajectory taken by the ball”. And again, he continues, “you obtain statistics that range from the percentages of shots taken to the shooting maps to the typology of the shots themselves”. Finally, “the system provides video highlights, an aspect that was previously left to a person who had to watch the entire recording and then choose the most salient moments. Now this task is performed quickly by our AI”.
Effective and versatile system
An advanced technology, also effective for youth sectors (at professional and semi-professional club level, TwinPlay has already supplied the system to Pallacanestro Varese, Olimpia Milano, Orange Bassano, Blu Basket Treviglio, Virtus Padova, Basket Pomezia, Bottanuco, Sermoneta, Piombino, just to name a few), where even more so the final result goes hand in hand with the improvement of one’s performance (individual and team).
The versatility of the system has also allowed it to have been tested, or adopted, by specific training centers for coaches, the Player Development Coaches (whose role, explains former American basketball player David Moss, “is to take young veterans and non-veterans and work on their skills”), including Sharp Shooter Academy, Extrapass, Clash Of Hoops and Bit-Lab which train both male and female Serie A and B players.
Without forgetting that TwinPlay technology is already being exploited by away student-players competing in the “NCAA Division I” basketball championship (i.e. the highest level of collegiate basketball in the United States). Remaining in the USA, it is worth mentioning the NBA Launchpadan initiative born in 2021 with the aim of identifying, evaluating and testing emerging technologies capable of advancing basketball and its related business.
