Thursday, 5 September 2013

Moneyball, FPL and the Premier League

What is Moneyball?

In a nutshell, Money ball is a bio-pic about a US baseball team that uses statistical theory in the form of sabermetrics to recruit players (based on a shoestring budget). This analytical approach was revolutionary as the general manager and his assistant assembled a team of under-rated players that on face value seemed average, yet had certain strong attributes which could be optimized. The statistical approach initially doesn't fare well, but the team eventually defies the odds and goes on a 20-match winning streak.  We later learn that sabermetrics is eventually accepted as a mainstream baseball theory and is eventually adopted by another team who win the World Series for the first time in 84 years.

Is Moneyball possible in the Premier League?

It depends how one interprets the theory, but I suppose in some respects it is possible.  From my perspective, picture a PL club with a shoestring budget that is looking to rebuild its team.  All they need is an open-minded chairman and a manager willing to take risks and be driven by well calculated stats for player recruitment.  Imagine them purchasing players in each position, not based on their overall qualities per se, but rather on specific attributes or stats and their potential to excel when gelled with other players of different specialisms.

For instance, a team may purchase a left-winger on a free transfer who has gone under the radar, yet has decent crossing ability and a proven record of assisting from out wide.  Or an attacking midfielder that may not have the best shooting ability, yet has natural pace to accelerate past defenders, etc.  Or more radically, converting the likes of Peter Crouch and Nikola Zigic into centre backs due to their height.  Stats and key attributes will determine what type of player is needed in which area and ideally they don't have to cost that much either.

The problem in the Premier League today is that a manager doesn't have total control of player recruitment.  The only exception to the rule arguably is Arsene Wenger, where as the rest have to rely on scouts or so-called "Directors of Football" to recruit players (or in Joe Kinnear's case make boasts yet spend £0).  Moreover, why would a Premier League club blindly rely on statistics? In addition, they may have a preference to go by certain football agents and their stable or perhaps focus on versatile players that have sufficient all-round attributes for their chosen position.

Having said the above, some may say that a stats-based method has already been used in the PL to recruit players.  Courtesy of FPLH Walt (from across the pond!), he cited an alleged example of this taking place under Liverpool's American owners, i.e. apparently by their 2011 purchases of Adam, Henderson and Downing.  I'm not sure how true that is.  But it's not beyond the realms of possibility.  Whether an adapted version of sabermetrics can be used successfully in the Premier League remains to be seen.

Is Moneyball possible in FPL?

Technically sabermetrics can't be fully integrated into FPL due to the scoring system being based on certain fixed statistics.  However, I believe that the principles which propelled the Moneyball team are significant.  Statistics can be used to help identify under-rated players (in particular differentials) which can help you score better than your FPL rivals.

I have to confess that my GW1 team had a bit of Moneyball about it.  The best example of which was my in defence via right-back Steven Whittaker.  I could have bought various Norwich FC players, but I opted for him specifically.  I hadn't noticed him that much till the England v Scotland match.  I then discovered that he had a fantastic scoring record for Rangers (28 goals in 209 matches) and would have played more for Norwich last season (had it not been for injury).  I also assembled other under-rated Moneyballesque players such as Morrison, Snodgrass and Downing. I chickened out at the last hour and made Whittaker my sub 2.  Fortunately, he had a fantastic GW1 and auto-subbed on for me!

Mish Mash: A Moneyball team with FPL Price caps?

All the above begs the question, could I assemble a Money ball inspired team with FPL price caps? i.e. a team of under-rated players that stand out on an individual basis but play for ordinary teams.  Here are the parameters I set myself as a guideline:

  • Attack-minded players (ideally).
  • Under-rated yet possess good attributes/stats.
  • Newly promoted or finished 6th-17th last season.
  • Goalkeepers maximum value: £4.5M.
  • Defenders maximum value: £5.0M.
  • Midfielders maximum value: £6.0M.
  • Forwards maximim value: £7.0M.

My Moneyball team:

I could justify all of my selections; the most random of which is Gerhard Tremmel.  He isn't even Swansea's first choice goalkeeper.  However, when you look at his saves statistics in the PL you wonder why hasn't he been given a better run between the sticks!  Anyway...

My fellow writers also came up with their own Moneyball teams, albeit with a bit more flexibility on the parameters.

Matej's Moneyball team

Walt's Moneyball team

What are your thoughts on Moneyball principles being used in the EPL & FPL?  How would your Moneyball team look like? Tweet me on @FPLHints or comment on this post.  Cheers!