As Bill Shankly told us, ‘Football is a simple game based on the giving and taking of passes, of controlling the ball and of making yourself available to receive a pass. It is terribly simple.’ I have been wondering what the great man would make of the number-led revolution that is sweeping through football. Instead of concentrating solely on playing games on the training ground, modern managers, coaches, directors of football, journalists and fans are having to embrace the reams of new information that are being provided by statisticians.
Using statistics to analyse football is providing insight into the fitness, work rate, accuracy and effectiveness of passes and shots, how many tackles are made and missed, defensive blocks and clearances, how many kilometres per game a player runs and much more. As Simon Kuper says, football is becoming clever. He quotes Chelsea’s performance director, Mike Forde as saying they have collected around 32 million data points over 13,000 games.
Even fans now are having to change the way they think about the game. Some of the best football analysis on the internet comes from places like the eplindex, anfieldindex, infostrada, and the godfathers of football statistics, opta’s joe, jack, paolo, franz, jose, donny, danny, joey, jordan and jonny.
In some ways football is just beginning on this path of using statistical analysis to understand what is happening on the pitch. Arsene Wenger was one of the first to bring statistical analysis of players to the English game. However using stats in American sports has been commonplace for some time now. One reason given for football’s hesitancy is that unlike baseball and American football, football is not characterised by stops and starts. The action usually flows, if you’re not watching England. However football managers are realising that much information, such as that listed above, can be collected, and used to their advantage. As collection techniques become more sophisticated, more and more data is able to be analysed. As John W Henry, the top dog of Liverpool’s new owners Fenway Sports Group, recently tweeted football has a way to go to catch up to the level of application of statistical information in baseball, but technology is helping.
Michael Lewis’ book Moneyball details the efforts of the Oakland A’s manager Billy Beane as he tries to find value in areas that others are unable to see. By using statistical analysis, while others were relying on things like ‘gut feeling’ the A’s were able to compete against teams with much bigger payrolls. The message of the Moneyball is that the richest teams will always win in the long run because they can afford to make mistakes, but also that less wealthy teams can be competitive through innovation and entrepreneurship.
How this translates to football is interesting. At Liverpool FC Henry seems to be following a pattern of working with football experts (Dalglish and Comolli) to identify where the team needs to be strengthened. After the gaps are identified they have spent big on young, talented players to fill those holes. The signings of Luis Suarez, Andy Carroll, Jordan Henderson, Charlie Adam and Stewart Downing are all young, creative players with high-work rates. The other side of this is the promotion of youth players from Liverpool’s academy that has really taken off under Dalglish. John Flanagan, Jack Robinson, Martin Kelly and Jay Spearing all played parts in Liverpool’s 2010-11 season, and this link from the first team to the academy is something that will continue to develop, with Conor Coady and Raheem Sterling looking like the next to make the step up.
At Liverpool both sides of this change are focused on youth, showing that identifying potential is paramount. Statistical analysis is playing a huge part in that. Director of football Comolli said in an interview after the Downing deal was completed that they analysed the statistics and are convinced that Downing is a terrific asset for Liverpool. This is despite many observers and fans, even Liverpool fans, thinking that he isn’t good enough, probably because he isn’t called Juan Mata. Particularly interesting was Comolli’s insistence that Downing is a player who is undervalued in the English market. This is important, many believe that Liverpool have paid too much for Downing (and Carroll, Henderson and Adam), yet Comolli believes he is actually worth far more than what they paid for him. This highlights an area of conflict that is gaining pace, that between the statistical analysis advocates and the traditional ‘gut-feeling’ people. Kuper alludes that Comolli himself was a victim of this when he worked at Tottenham, and that eventually resistance from Spurs’ ‘traditional’ manager meant that Comolli had to leave, taking his spreadsheets with him.
Identifying where a team needs to improve and replace players is one area that clubs use stats, another is to identify weaknesses in opposing teams. The incoming Chelsea manager Andre Villas Boas used to scout opposition teams for the then Chelsea manager Jose Mourinho. His tactical descriptions are purported to contain the strengths, weaknesses and tendencies of each player that he saw. This information was then used to determine the best course of attack in the upcoming game.
Shankly would have appreciated this kind of information, and no doubt used modern technology to gain any competitive advantage he could. As Jonathan Wilson recounts in Inverting the Pyramid Liverpool had been so dominant domestically in the 1960s, yet had struggled in European competition. Shankly made a number of innovations, the beginning of the ‘boot-room’ culture where teams, players and tactics were discussed in an atmosphere conducive to free-speech. Changing to use more ‘continental’ tactics is another. Liverpool went from having a traditional ‘hard man’ at the back in defence, to using players who were more comfortable in possession. Phil Thompson replaced Larry Lloyd in the centre of defence, forming a partnership with Emlyn Hughes. This set the standard for Liverpool’s ball-playing centre-halves for years to come. Making better use of the ball and finding space on the field to receive a pass represented tactical innovations that gave Liverpool a competitive advantage in English competition but also got them onto the same page as their continental opposition.
After World War II American scientists began to rely heavily on science and statistical models to try to gain an advantage over their Soviet rivals. Politics became a ‘science’ and technology was used to map out every possible outcome of every possible action. Think-tanks like the RAND corporation were developed to use computers to run models and simulate the moves of other nations. Through these ‘war-games’ politicians hoped to be able to make better decisions, and out-think their rivals. There is an obvious parallel to make with football currently.
Technology is here to stay, and the winners will be the ones who can use it most effectively. The statistical models are becoming just like the traffic cone and five-a-side pitch, another tool of the training field. As a master of using psychology, deception and innovation to gain an advantage, Shankly would have understand that and embraced it wholeheartedly.
See the other posts in this series here: Football & IR Theory