Microsoft Trials Tech That Might Explain Why Australian Cricket Sucks

Given how badly the tests against South Africa have gone this season, having Shane Watson back in the side might not be a bad thing. But if that doesn’t correct the atrocious performance of our nation’s cricketers, some new technology might.

If you’re a bit confused, let me explain. South Africa has been touring the cricket pitches of Australia for a little while, and they’ve been belting the living snot out of us. The Australian captain said he was embarrassed to front the media. There’s rumours that as many as five or six players might get the boot for the third Test. And we’ll probably still lose that anyway.

So yeah, things are good. But Microsoft might be about to come to the rescue.

Earlier this week, Cricket Australia announced they would be trialling new technology from Microsoft amongst their coaching and support teams at the state and national levels. The tech is largely data-focused, relying on predictive analytics and machine learning to identify practical measures players can take to improve on their performances and prepare for bigger matches.

It’s not known whether the system has metrics detailing why Australian batsmen are so bad at dealing with reverse swing, although at this point fans will take any number that doesn’t read 8/32 or 10/85.

“Our sports science guys have been doing some very interesting things trying to understand probability of injury, and things like that,” Michael Osborne, head of technology for Cricket Australia, told the SMH. “But what we’re finding is, for a lot of our frontline coaching staff, there’s almost too much data, and to get in and understand what the data is saying is a daunting task.”

Players can provide a daily update on their overall health and the amount of sleep they got through a mobile app, which is then fed into a database that also collates performance metrics and specialist player tracking technology. That allows the coaching and support staff to have a much more current picture of a player’s fitness and readiness for a match, and can also gauge their past performance and updated condition against the live conditions and historical information of a particular ground (like, say, Adelaide Oval or the SCG). The system can even theoretically select the best possible team for the occasion, although Cricket Australia stressed it would not be abandoning human selectors.

(The Marsh brothers will be a permanent fixture of the Australian squad for the next three years, then.)

The machine learning part is handled through Microsoft’s Azure data centres, and Microsoft said they would be able to apply the lessons learnt from the trial to other sporting codes. It’s also not a stretch to imagine that a similar system of similar metrics or may one day become available for video games, provided developers were happy to track and record all of the information necessary for advanced player analytics.