Basketball and AI. Positioning
They say that, it is possible to transfer collective behavior of molecules and fruit flies to NBA players.
Density functional theory comes in handy when one talks about the optimal positioning of basketball players in the given situation. This leads to increased chances of successful scoring or defending against the other team.
Boris Barron, presenting his work at the American Physical Society in Las Vegas explained that he utilized in depth data of individual’s positioning from the current NBA season games to create his model.
Barron explained that he could predict the direction where a player could go next, which players tend to be in good or bad positions, predict the chances of success, both offensively and defensively, considering the player’s positioning, and predict the other team’s reaction depending on the player’s move.
One can see where a player should go to be of help to his team and those few feet can make around 3% difference, according to him.
If we deal with high-scoring games, those 3 points out of 100 matter a lot for a player, according to Aris.
The Barron’s uses Nobel Prize methods, created to study large collections of quantum mechanically interacting materia. It also builds on Arias' test, fusion of mathematical concepts and approaches from density-functional fluctuation theory for researching crowd behavior, social phenomena, things like migration and segregation.
Those are appropriate when analyzing basketball, since the peoples’ groups are tough to quantify.
It is important, because one is not looking at players individually, but at their collaboration, he said.
The Barron’s model is beneficial for team sports. Coaches can load the data, team specific or player - specific, for more efficient plays on court.
AI Catalog's chief editor
Share on social networks:
