Statistical innovation helped to quickly turn the Red Sox into champions, writes Ciaran Cronin, and crunching numbers may aid a Liverpool revival
If there was any doubt as to whether John Henry and his NESV consortium were going to model their ownership at Liverpool on the structures they put in place at the Boston Red Sox, they evaporated on Wednesday afternoon. In appointing Damien Comolli, Tottenham's former director of football, as the club's first director of football strategy – different title, same job – it was entirely reminiscent of the day, in November 2002, when the Red Sox hired 28-year-old Theo Epstein as general manager. Epstein had no formal qualifications in the practical side of baseball but was brimful of knowledge on the theory. Much like Comolli, in respect of football. The American went on to bring two World Series titles to Boston in the space of his first four seasons in the job, not by paying the biggest wages, or signing the most talented players, but through a clear devotion to the principles of sabermetrics. Comolli does not have a body of knowledge to guide him, like young Epstein did when he was appointed, but Henry clearly believes the studious-looking Frenchman can develop it.
Sabermetrics, you ask? "The search for objective knowledge about baseball," is how Bill James, the man credited with developing the theory, describes it. Essentially, it involves the rigorous analysis of the reams of statistics baseball throws up as a matter of course but, crucially, seeks to analyse them in a different way than convention had previously dictated. The best, and indeed easiest, example concerns "batting average", a statistic which goes back to baseball's inception. It is calculated, quite simply, by dividing the number of hits a batter manages by the number of times he is at the plate. For decades, it was the primary method for judging how successful a batter was but the disciples of sabermetrics deem it an unsatisfactory measure of a batter's worth. Instead, they believe a more useful gauge of precisely that is his ability to help his team score runs, which is why they came up with the formula for "on-base percentage". That is but one small example of how sabermetric practitioners have, over the past 20 years, turned the analysis of baseball on its head.
Henry and NESV, it would seem, are attempting to do something similar in the domain of football. A key component of NESV's ownership strategy with the Red Sox was, and still is, to eke out every possible value on the playing side of the financial equation. In Boston, Epstein used sabermetric principles to identify, and in turn recruit, players that few other franchises had even thought of signing. In essence, the Red Sox became successful because of how efficiently they used the limited cash their owners provided them with. They exploited the inefficiencies in the player recruitment market and reaped the benefits.
Producing a similar value for money seems to be high on Henry's plans for Liverpool, even if football has no direct equivalent to sabermetrics. It's fairly obvious why none exists – there are fewer "data points" to glean statistics from in football because, as a sport, it is less start-stop in nature than baseball – but Henry seems to believe that recruitment by numbers in football is not necessarily impossible, particularly as the statistics accompanying the game get better and better. He will, no doubt, be fascinated by the realm of data provided by Prozone, a service that Roy Hodgson insisted Liverpool sign up to on his appointment this summer. While many have been writing the 63-year-old's Anfield obituary these past few days, Hodgson's belief in statistical worth may be regarded as especially useful under the new regime.
Arsene Wenger, as you'd expect, was one of the first managers to subscribe to Prozone's fledgling services a decade ago, which are entirely confidential to the club who pays for them. The common perception is that the statistics only tell how far a player has run and how many passes he has completed but in truth, those are the numbers that are put up for public consumption; in private, they go far deeper than that, as Wenger explains. "If I know that the passing ability of a player is averaging 3.2 seconds to receive the ball and pass it, and suddenly he goes up to 4.5, I can say to him, 'Listen, you keep the ball too much, we need you to pass it quicker'. If he says 'no', I can say look at the last three games: 2.9 seconds, 3.1, 3.2, 4.5." Word has it that Gilberto Silva, Arsenal's Brazilian international midfielder, was shipped out of the Emirates precisely because he was dwelling on the ball too long. It might not be the type of observation that Wenger would have missed without the data at his fingertips, but it must help to have some affirmation all the same.
Prozone, however, is primarily concerned with performance of current players and while it has recently initiated a service for tracking future signings – which Everton have become the first Premier League club to sign up to – Henry, NESV and now Comolli might be more interested in the transfer theories of Simon Kuper and Stefan Szymanski. The pair's book Soccernomics (or Why England Lose: And other curious phenomena explained as it was called in its first edition) has become a bible for people who want to turn football convention on its head, just like Michael Lewis's Moneyball was for sabermetrics and baseball.
In its pages, Kuper and Szymanski outline, on the back of their research, a number of faults that exist in the transfer market and suggest where real value might actually lie. They offer a list of 12 transfer principles. Some – that stars of recent major tournaments are overvalued, that older players are overvalued and that you should buy players in their 20s – are pretty obvious but there are some suggestions that NESV might want to consider.
For example, the sabermetric belief that too many "sight-based prejudices" exist in recruitment is backed up by Kuper and Szymanski. They cite a survey conducted on football scouts which found that they regularly throw up the names of blond-haired players more than any other, simply because their hair catches the eye. The pair also make the observation that players of certain nationalities are overvalued – the example is given of the agent who finds it much easier to sell a crap Brazilian than a brilliant Mexican – and that clubs need to put more resources into helping new signings relocate.
There is an interesting tranche of beliefs surrounding the sale of players. Kuper and Szymanski recommend that you should always sell a player if another club offers more than he's worth and that sentimentality should never come into it. They also float the notion that a good way of getting a bargain is to sign a player with personal problems and then help him solve them. :angel:
Arguably the most fundamental point that the authors make, however, is that transfer decisions should never be made solely by somebody more interested in the short term. Which is precisely what Henry has done in appointing Comolli to recruit for them. The 38-year-old may have left Tottenham effectively a failure after Juande Ramos was sacked in October 2008 but there is an argument to suggest his White Hart Lane tenure wasn't a complete disappointment. There were as many failures as there were successes – Gareth Bale and Luka Modric on the one side for example; Didier Zokora and David Bentley on the other – when he was responsible for recruitment at the north London club but what's not up for debate is the obtuse manner in which Comolli went about his business.
The Frenchman tended to sign either players in their mid-20s from teams in the top half of European Leagues, or young Englishmen with potential at smaller clubs than Tottenham, rather than established stars. If he wasn't an unqualified success, at least his approach was innovative.
It's something we should expect in Liverpool's transfer dealings from now on.
If there was any doubt as to whether John Henry and his NESV consortium were going to model their ownership at Liverpool on the structures they put in place at the Boston Red Sox, they evaporated on Wednesday afternoon. In appointing Damien Comolli, Tottenham's former director of football, as the club's first director of football strategy – different title, same job – it was entirely reminiscent of the day, in November 2002, when the Red Sox hired 28-year-old Theo Epstein as general manager. Epstein had no formal qualifications in the practical side of baseball but was brimful of knowledge on the theory. Much like Comolli, in respect of football. The American went on to bring two World Series titles to Boston in the space of his first four seasons in the job, not by paying the biggest wages, or signing the most talented players, but through a clear devotion to the principles of sabermetrics. Comolli does not have a body of knowledge to guide him, like young Epstein did when he was appointed, but Henry clearly believes the studious-looking Frenchman can develop it.
Sabermetrics, you ask? "The search for objective knowledge about baseball," is how Bill James, the man credited with developing the theory, describes it. Essentially, it involves the rigorous analysis of the reams of statistics baseball throws up as a matter of course but, crucially, seeks to analyse them in a different way than convention had previously dictated. The best, and indeed easiest, example concerns "batting average", a statistic which goes back to baseball's inception. It is calculated, quite simply, by dividing the number of hits a batter manages by the number of times he is at the plate. For decades, it was the primary method for judging how successful a batter was but the disciples of sabermetrics deem it an unsatisfactory measure of a batter's worth. Instead, they believe a more useful gauge of precisely that is his ability to help his team score runs, which is why they came up with the formula for "on-base percentage". That is but one small example of how sabermetric practitioners have, over the past 20 years, turned the analysis of baseball on its head.
Henry and NESV, it would seem, are attempting to do something similar in the domain of football. A key component of NESV's ownership strategy with the Red Sox was, and still is, to eke out every possible value on the playing side of the financial equation. In Boston, Epstein used sabermetric principles to identify, and in turn recruit, players that few other franchises had even thought of signing. In essence, the Red Sox became successful because of how efficiently they used the limited cash their owners provided them with. They exploited the inefficiencies in the player recruitment market and reaped the benefits.
Producing a similar value for money seems to be high on Henry's plans for Liverpool, even if football has no direct equivalent to sabermetrics. It's fairly obvious why none exists – there are fewer "data points" to glean statistics from in football because, as a sport, it is less start-stop in nature than baseball – but Henry seems to believe that recruitment by numbers in football is not necessarily impossible, particularly as the statistics accompanying the game get better and better. He will, no doubt, be fascinated by the realm of data provided by Prozone, a service that Roy Hodgson insisted Liverpool sign up to on his appointment this summer. While many have been writing the 63-year-old's Anfield obituary these past few days, Hodgson's belief in statistical worth may be regarded as especially useful under the new regime.
Arsene Wenger, as you'd expect, was one of the first managers to subscribe to Prozone's fledgling services a decade ago, which are entirely confidential to the club who pays for them. The common perception is that the statistics only tell how far a player has run and how many passes he has completed but in truth, those are the numbers that are put up for public consumption; in private, they go far deeper than that, as Wenger explains. "If I know that the passing ability of a player is averaging 3.2 seconds to receive the ball and pass it, and suddenly he goes up to 4.5, I can say to him, 'Listen, you keep the ball too much, we need you to pass it quicker'. If he says 'no', I can say look at the last three games: 2.9 seconds, 3.1, 3.2, 4.5." Word has it that Gilberto Silva, Arsenal's Brazilian international midfielder, was shipped out of the Emirates precisely because he was dwelling on the ball too long. It might not be the type of observation that Wenger would have missed without the data at his fingertips, but it must help to have some affirmation all the same.
Prozone, however, is primarily concerned with performance of current players and while it has recently initiated a service for tracking future signings – which Everton have become the first Premier League club to sign up to – Henry, NESV and now Comolli might be more interested in the transfer theories of Simon Kuper and Stefan Szymanski. The pair's book Soccernomics (or Why England Lose: And other curious phenomena explained as it was called in its first edition) has become a bible for people who want to turn football convention on its head, just like Michael Lewis's Moneyball was for sabermetrics and baseball.
In its pages, Kuper and Szymanski outline, on the back of their research, a number of faults that exist in the transfer market and suggest where real value might actually lie. They offer a list of 12 transfer principles. Some – that stars of recent major tournaments are overvalued, that older players are overvalued and that you should buy players in their 20s – are pretty obvious but there are some suggestions that NESV might want to consider.
For example, the sabermetric belief that too many "sight-based prejudices" exist in recruitment is backed up by Kuper and Szymanski. They cite a survey conducted on football scouts which found that they regularly throw up the names of blond-haired players more than any other, simply because their hair catches the eye. The pair also make the observation that players of certain nationalities are overvalued – the example is given of the agent who finds it much easier to sell a crap Brazilian than a brilliant Mexican – and that clubs need to put more resources into helping new signings relocate.
There is an interesting tranche of beliefs surrounding the sale of players. Kuper and Szymanski recommend that you should always sell a player if another club offers more than he's worth and that sentimentality should never come into it. They also float the notion that a good way of getting a bargain is to sign a player with personal problems and then help him solve them. :angel:
Arguably the most fundamental point that the authors make, however, is that transfer decisions should never be made solely by somebody more interested in the short term. Which is precisely what Henry has done in appointing Comolli to recruit for them. The 38-year-old may have left Tottenham effectively a failure after Juande Ramos was sacked in October 2008 but there is an argument to suggest his White Hart Lane tenure wasn't a complete disappointment. There were as many failures as there were successes – Gareth Bale and Luka Modric on the one side for example; Didier Zokora and David Bentley on the other – when he was responsible for recruitment at the north London club but what's not up for debate is the obtuse manner in which Comolli went about his business.
The Frenchman tended to sign either players in their mid-20s from teams in the top half of European Leagues, or young Englishmen with potential at smaller clubs than Tottenham, rather than established stars. If he wasn't an unqualified success, at least his approach was innovative.
It's something we should expect in Liverpool's transfer dealings from now on.
