Algorithm uses data from top tennis players to predict next shot

Nadal vs Federer at the 2012 Australian Open men's final. Data from this match helped inform the algorithm. (Credit: FlashStudio / Shutterstock.com)

Ever wondered what the dream STEM job might be for a tennis obsessive? If you’re already suffering Australian Open withdrawals, then you might be envious to hear a team of QUT researchers have been spending their time analysing thousands of shots by top male tennis players, including Djokovic, Nadal and Federer. All in the name of science of course!

By studying data from the 2012 Australian Open Hawk-Eye (the computer system that visually tracks ball trajectory during matches), the researchers developed an artifical intelligence system to predict a tennis player’s next shot based on the player’s style, and the point in the game.

The researchers analysed more than 3400 shots for Djokovic, 3500 shots for Nadal and 1900 shots by Federer, adding context for each shot such as whether it was a return, a winner or an error.

“After about 1000 shots, the model has a pretty good idea of what is going on,” said Dr Simon Denman, a Senior Research Fellow with the Speech, Audio, Image and Video Technology Laboratory at QUT.

“It needs about three matches to figure out a player’s style. Once it’s got those three matches it’s pretty solid.”

The machine learning system can predict about 1000 shots in 30 seconds. But the hardest player for the AI to predict? That would be the one and only Roger Federer.

“It was least accurate for Federer, who is perhaps the most versatile. It struggled the most to predict him. He can do anything, so the model was more often wrong about him,” said Dr Denman.

Dr Denman said he thinks in less than 10 years top-level players will be able to use this sort of technique in studying the game of an upcoming opponent. It could also lead to new virtual reality games offering the chance to go head-to-head with AI versions of the world’s top players.

If this sounds like your dream STEM job, you might want to consider a career in sports science.

You can read more about this study over on QUT’s website, or watch their video below.

Gemma Chilton

Author: Gemma Chilton

Gemma has a degree in journalism from the University of Technology, Sydney and spent a semester studying environmental journalism in Denmark. She has been writing about science and engineering for over a decade.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.