Analysis backs long kicking
Stephen Rielly | May 23, 2007 | The Age
LAST Wednesday evening, AFL recruiters and football department heads from around the country gathered in a small, nondescript room on the 28th floor of Commerce House, at the top end of Collins Street.
They did so to hear Dr Mark Stewart, of the RMIT School of Economics, Finance and Marketing, reveal a secret. To be initiated.
Stewart and two colleagues, Heather Mitchell and Constantino Stavros, are the researchers who, a little over 12 months ago, sought to apply the principles unearthed in Michael Lewis' groundbreaking and best-selling book, Moneyball, to the Australian game at the highest level.
Lewis, once a Wall Street analyst, dumbfounded many inside the American professional baseball community on the publication of his book in 2003, by producing an answer to a simple question: how is it that the Oakland Athletics, with a relatively low payroll of $40 million in 2002, could win as many games in that season as the New York Yankees, a team with a payroll more than three times greater at $126 million?
The answer was a discovery, made by Oakland and its general manager, Billy Beane, of a new statistical method of analysing the game that not only found folly in a number of the traditions in which baseball is steeped but allowed Beane and the A's to recruit players in whom others could not see value. Moneyball was an inspiration to the little guys of professional sport the world over and no less an encouragement to statisticians, economists and sports-minded thinkers everywhere.
Which is why every AFL club was listening to Stewart last week, as he delivered the earliest conclusions to identify and measure what is truly important to winning games of football.
Stewart and his team analysed 738 of the 740 AFL games played between 2002 and 2006. In doing so, they found 21 statistics critical to success or not and then measured how important or costly each was.
So what did all of those AFL ears hear? According to the research, each accurate long kick is worth almost one point (.99) to a team's winning margin. A kick to the opposition subtracts .62 of a point.
They heard that players who bounce the ball or, in other words, take the game to the opposition are almost as valuable. Each bounce adds .54 of a point to a team's score.
"This form of advancement is particularly valuable in that it is a relatively safe way to move closer to a scoring opportunity as the ball is not required to be transferred to a teammate, eliminating the possibility of an interception," the Stewart, Mitchell and Stavros findings reported.
They heard that decisions made by the men in white, green or orange have next to no bearing one way or another upon results. Ditto hitouts, which are devalued by the analysis and, therefore, question the worth of the game's biggest men.
"Coaches, media and fans often blame umpires for losses and similarly lament that their ruckman was unable to win enough contests in hitting the ball out following stoppages in play. What this study indicates is that who hits the ball out in a ruck contest is not as important as who is able to effectively clear the ball from such a stoppage," the report said.
A single centre bounce clearance is worth .51 of a point which, of course, makes a player who can win the football, sprint away bouncing it and then kick it accurately the most valuable footballer of all.
Which is precisely what Chris Judd is.
Surprises? The knock-on is accorded a relatively high value. Each is worth .35 of a point, according to the analysis, considerably more than many statistics that have conventional importance attached to them, such as tackles.
What matters most:Accurate long-kicking, bounces, centre clearances, knock-ons.
What is not important:Umpiring decisions, ruck contests, tackling.
http://www.realfooty.com.au/news/news/analysis-backs-long-kicking/2007/05/22/1179601410561.html?page=fullpage#contentSwap1
