Moneyball- What's the football equivalent?

So watched this Moneyball movie which the idea being statistical analysis can identify useful players by emphasising things that are important for a team to win overall. There's a book too, but haven't read it.

In that movie the golden stat seems to be that a player can 'get on base' in the case of the Oakland 'A's from 2003.

Gareth seemed keen to use this as an analogy for how to buy a team that is cheaper than other teams, but still have success. For the record the 'A's haven't won the World series since this development though they did go on a record winning streak.

For the train spotters- what would be the football equivalent of a player who can 'get on base'? Is there one?

Isn't that what Liverpool's owners were trying to do with their transfer policy last season. Didn't work out too well for them. I wonder what Alan Pardew call his approach? It seemed to work better anyway...

Both Mark Hughes and Sam Allardyce are big fans of Prozone, and used that to try and get results with bargin basement buys at Rovers. That is probably the closest thing in football to Sabermetrics.

Ok its the Daily Mail, but this is a decent read about the EPL and its use of Sabermetrics.

http://www.dailymail.co.uk/sport/football/article-2063956/Premier-League-statistics-Is-Wayne-Rooney-better-Mark-Gower.html

http://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics

There is no football equivalent of money ball. What he said made no sense whatsoever.. Other than for it to be a nice way of saying "we're going to be cheap,and just aim for the playoffs and hope for the best from there"

I think you could use stats for successful recruitment, but the golden stat would depend on positions and on intended tactics and playing philosophy.

As a side note, Liverpool didnt seem to do the moneyball thing right at all. Spent way too much on young players or guys who were average

Agreed, Liverpool were the anti-Moneyball in almost every way

I heard Ricki was really into statistics.

That's one reason why he prefers previous A-league players to players without A-league experience.

Seemed like they started with a Moneyball strategy, then adapted it slightly to favour British players, then adapted it slightly to centre the squad around certain players, then adapted it slightly around getting rid of Woy's players, and so on and so on until their adapted version of Moneyball was actually anything but.

isnt arsene wenger the oakland guy's number one idol? read that somewhere

A tactics book from the FA in 1990 by Charles Hughes "The Winning Formula" used stats to demonstrate that long ball is the most successful tactic in football. Ricki has probably read it already.

Thats a fascinating statistic.

The 'gets on base' equivilant is probably something like shots on target, crosses, tackles and interceptions depending on which postion you play.

Gets on base basically meant that you at least could get runs, which is how you win games. In football a shot on target is whats needed to win games, as well as tackles at the back with interceptions and crosses for the midfield..

Just hypothesising here...

Hughes was sort of plagirising Charles Reep's work: http://en.wikipedia.org/wiki/Charles_Reep

Just reading about this in Jonathan Wilson's Inverting the Pyramid. 

To be fair to Gareth right from day one he has said that their aim is to get the club running so it does not make a loss. Buying cheap may be one factor....but stats have shown that that quite a large proportion of "big name" purchases fail to deliver. As Gareth said its a "pig in the poke" buying big and they won't do it. You could get dispondedent about this as everyone likes the idea of a class player to excit the fans.

However to give the Nix 9 their dues they have another strategy....one that has a proven track record internationally. That is youth development. There are plenty of examples of this internationally....the famous Barca camp especially. But there are others. Man U were knocked out tof the Champs League by a Basel side fielding 6 players developed from their Academy. Montpellier just won the French League fieldling 8 players from their Academy....their budget was 20% of Paris St Germain who blew 100 million euros+ on players pre season.

I have absolutely no doubt that the Academy program will reap rich rewards for the Nix if done properly. Unfortunately it will take time.....in the mean time if the club continues to "buy small" on the transfer market to balance the books then we will just have to live with it. Buying smart is probably more important than any scale of the cost.

You been reading the book soccernomics? I think it's in there.

The main reason that there is no equivalent is that the games have a very different structure.  Baseball is a series of set pieces, which makes it fairly easy to break down into statistical series.  While there are set pieces in football, the main part of the game is based on structured team play and combinations.

^ good point

same goes for NFL

Football notoriously difficult to get more meaningful statistics out of. Even the fundamental ones like possession percentage don't tell a story in a lot of cases. Sometimes possession is 50/50 after 60 minutes, but the team leading will sit deep to see out the game, ending the 90 minutes with perhaps as low as 30-40%. If you read that without seeing the game you'd have an incorrect view of what took place.

# km run per match - hugely useful stat.

My guess ... km run /match at top pace  stat with Brockie and Totori on the field this year could be up there among the best in the League....and if Ifill's drawing 2+players out of position stat score/game remains at previous levels he'll create the gaps for either of them to exploit too. Ha! its football..but not as we know it ;-)

GPS personal player statistics are a whole different kettle of fish...

I was wondering where to link this article and here seemed like the most relevant thread:

http://www.newscientist.com/article/dn22000-ai-fantasy-football-manager-picks-a-winning-team.html

"Sarvapali Ramchurn, a computer scientist at the University of Southampton, UK, and colleagues have developed an artificial soccer manager that ranks in the top one per cent of the 2.5 million players on the official English Premier League fantasy football game.

The AI manager evaluates players' previous performance statistics to predict how many points they will receive during the forthcoming week. It then picks a squad that will maximise the expected fantasy football score while adhering to the rules of the game, such as how much it costs to buy or exchange a player."

Next stop Skynet...

Sack Ricki.

Hire Sarvapali Ramchurn.

The difference with fantasy football is that there are a few key stats that are considered very important in this simplified fantasy world, and thus earn points. These are minutes played, yellow/red cards, goals and assists for all players, goals conceded for defenders and goalkeepers, and saves made for goalkeepers. If you use those stats to select a team to play a game of real football (Moneyball style), I doubt your that team would do very well.

I think the major problems I have with the stats-based approach are twofold: 1)it doesn't account for tactics and team play- how a player is utilised in a formation and the players around him. If you play with Jan Koller as a striker you're gonna have way more successful longball attempts than if you play with Messi as a striker, to make just a really simple example. 2) it doesn't account for personality/psychology. players are assumed to play the same regardless of who they are playing with and for but that clearly isn't true. Some players respond better to certain coaches or environments than others do, take Shevchenko as an example here - compare him at Milan with him at Chelsea.

Baseball, like cricket, is a sport which is really more individuals playing in teams than a team sport - pitcher v batter/bowler v batsman - whereas in football, except in a few rare cases, it is hard to separate the statistical aspects of a players performance from team factors. Shots on target might depend on where your teammates are playing the ball to you for instance. Pass completion relies not just on your passing accuracy but on teammates to be in positions to pass to. Crosses completed depends on the aerial ability of your stikers. Successful dribble attempts might be the result of other players drawing defenders away from you with well-timed runs. And so on...

In my opinion, the major difference to inplementing the "Moneyball" philosophy in football is that there are hundreds of times the amount of potential footballers to recruit from spread across a vast amount of leagues with varying skill levels making it extremely impossible to implement. The talent pool in Baseball is much smaller.

This system worked perfectly for the Boston Red Sox (owned by Fenway Sports who own Liverpool) 2 years after it was tried by the Oakland A's in 2002.