יום חמישי, 24 בנובמבר 2011

How statistics shaped a Hollywood blockbuster

Baseball, statistics and Brad Pitt are the unusual ingredients of Hollywood's new release.
Moneyball tells the compelling true story of a coach who pioneered a radical use of data to turn around the fortunes of his side.

Based on the book by Michael Lewis, the film centres on three baseball players being plucked out of relative obscurity by Oakland Athletics's general manager Billy Beane.

Using a complex statistics system, he drafted in a player with a strange pitching action, one who was seen as over the hill and another who was renowned for his off-field antics rather than his aptitude on the park.

They were seen as cast-offs by other teams but Beane, played by Brad Pitt, used the sabermetrics theories documented in a series of books by Bill James to capitalise on values other franchises ignored.

In essence, he focused on on-base percentage - the number of times the player makes it to first base by whatever means - instead of more traditional or eye-catching batting statistics like home runs or batting average.

 

And in doing so he cast aside traditional scouting opinions, causing controversy, but it was a success as Beane's side closed the gap between his cash-strapped outfit and richer franchises.

Beane embarked on his radical approach in 2002. Since then the use of statistics has become commonplace across sport. However that does not mean the Moneyball approach can be easily applied outside baseball.

Brad Pitt plays general manager Billy Beane in Moneyball. Photo: Getty

In football terms, it is like valuing shots on target or cross completion over headline statistics like goals scored and assists.

Yet there are several problems in trying to distil all football's permutations into one or two key statistics. The game's fluidity and the fact teams score their goals in different ways are just two stumbling blocks.

What works for Arsenal does not necessarily work for Stoke, but that does not stop all 20 Premier League clubs and 23 out of 24 Championship sides trying to gain an advantage by using data from performance analysts Prozone and Amisco.

While yet to crystallise a team's fortunes to the extent of the Oakland As, it is clear that a reliance on physical and technical data now plays an integral part in many professional clubs, from signings right down to the youth teams.

 And football has identified what it believes are some key indicators of successful performance:

- The number of touches
- Goals and, importantly, goal difference
- Shots and, importantly, shots on target
- Total passes and, importantly, forwards' passes percentage success
- High-intensity* distance covered without possession *(5.5 m/s)

Surprising insights from last season's data show Wigan's Hugo Rodallega hit the second most shots on target, while Leighton Baines was the Premier League's most successful crosser.

So what is the true value to football of statistics?

Prozone and Amisco arranged a viewing of Moneyball this week to highlight how similar principles worked in football.

Former England boss Steve McClaren was among the audience.

"For me it is the best coaching tool that I have in my bag," said McClaren, who was one of the first to use Prozone's services when he was a coach at Derby under manager Jim Smith and took it with him when he joined Manchester United in 1999.

"Performance analysis allows me to get as much information as I can on that player in that game or in training, and it helps me to coach and develop players. So from that perspective it has been very important.

"The only thing I would say is that there is so much information available now it's really deciphering what is relevant. The key thing for me is: what actually wins football matches?

"Give me the statistics that win football matches, whether that's shots on goal, outrunning opponents, especially in high-intensity sprints, or whether that's winning challenges."

Throughout Moneyball, Beane has little contact with his players and avoids watching games in person. But after a horrid start to the 2002 season, where the staff and players openly question his methods, the team's fortunes begin to improve once he starts speaking to his valued assets.

McClaren said: "What showed in the film was that he took all that information, he brought the players in on that information but at the end of the day he still had to coach.

"You see the turnaround, compared to when there was a resistance from the manager Art Howe which obviously transmitted to the players. That's why they didn't perform.

"Once Beane interacted with the players and started coaching, the relevancy of the statistics became apparent and then the team started to improve.

"You can get bogged down by the information and just having that information is not enough. You still have to coach and still have to have the conviction in your ability."

Liverpool are an interesting case in all of this, with the club's director of football Damien Comolli known to be an advocate of using statistics to back up his choices in the transfer market.


 And owner John W Henry, who also owns the Boston Red Sox, was clearly a fan of Beane's methods as he offered him a record-breaking contract, which he turned down. James is a senior advisor at Fenway Park too.

Of the many signings that Comolli oversaw in 2011, it is the contrasting fortunes of £22.8m Uruguayan striker Luis Suarez and £35m English forward Andy Carroll that have drawn most scrutiny.

It shows that, for all the statistics used to bring a successful team together, there are just as many social and psychological factors which play their part.

You can find out more about Moneyball in a special show on BBC Radio 5 live on Thursday, 24 November from 1900 GMT.

Source: http://www.bbc.co.uk/blogs/thefootballtacticsblog/2011/11/how_statistics_shaped_a_hollyw.html

Gerard Pique Rio Ferdinand John Terry Karim Benzema

אין תגובות:

הוסף רשומת תגובה