The Transfer Price Index 2011/12 Premier League Predictions

By Dan Kennett, with input from the rest of the TPI team.

  • Definitions Used
  • CTPP = Current Transfer Purchase Price.  The player’s original transfer value adjusted for football inflation using the same method as the Retail Price Index. (Andriy Shevchenko remains the highest CTPP, his £30.8m transfer in July 2006 now worth £68.2m)
  • Sq£ = The cost of a club’s squad for a season in CTPP
  • £XI = The average cost (in CTPP) of the starting XI’s in Premier League matches that season
  • MSq£ = Multiple of average Sq£. Not how much a squad costs in CTPP, but how much more (or less) it costs versus the average squad that season.

The original Pay As You Play book focused on £XI and demonstrated that as the Premier League era has progressed, the final positions in the league table have become ever more correlated to £XI.  As a retrospective measure £XI is unrivalled in showing how close the relationship is between money and success.

Since publication of PAYP in October 2010, subsequent analysis of the Transfer Price Index database has shown how MSq£ can be used as a predictive measure and with significant[1] correlation to final league positions. The new season represents the first opportunity to use MSq£ as a predictive model before the season starts.

Important Note: The cut-off point for the initial analysis was 11th August, some 3 weeks before the end of the transfer window.  Final Sq£ for a season can only be calculated after all transfers for a season are complete (i.e. 1st February).  This article will be updated after 1st September and 1st February with the latest Sq£ data fed into the model.  The original predictions will be tracked and any changes clearly identified.

As we head into this coming season there is unprecedented imbalance in the distribution of the cost of squads in the league[2]. Over the history of the Premier League, teams that have the average squad cost for that season (MSq£ = 1) have typically finished in 11th place.  As of 11th August 2011, only the top 6 had MSq£ > 1 and this is extremely unlikely to change.

The model clearly divides the PL into a number of bands.  In our opinion, the only real scope for variation is the positioning within a band.  This is based on trends in the Premier League era and some of these are detailed in the band narrative below.  However, where the absolute MSq£ rankings and squad costs are very closely grouped, the subjective predictions of the TPI team have been used to rank the teams.


  • 1. The “big 3”: Chelsea, Man City and Man Utd

Between them, Chelsea and United have 5 of the top 7 biggest CTPP transfers in their squads

Man City have the largest number of players in the top 10% of Premier League CTPP

It’s difficult to convey just how far these 3 are ahead of the rest in the MSq£ model. However, there is only a 2.5% chance of Man United finishing outside the top 3. For Chelsea and Man City it’s less than 1%

  • 2 “Champions League aspirants” (4th to 6th):,Liverpool, Spurs and Arsenal

The model has shown that if a club wants a better than 50% chance of getting a Champions League spot it needs MSq£ of at least 1.98.  Neither Spurs, Liverpool or Arsenal are at that figure so the race for 4th could be extremely close. (Note: Arsenal tend to overachieve in the model. But with more clubs investing in players, and Arsenal losing some key ones, it may now be harder than before.)

  • 3. “Europa League aspirants” (7th to 9th): Aston Villa, Sunderland and Everton

The model substantiates the perceived “glass ceiling” of the Premier League in more places than one.  We are 95% certain that based on this seasons squad costs, none of Aston Villa, Everton and Sunderland will finish in the top 6.

It will certainly be interesting to see the fortunes of Everton vs Sunderland.  Everton have a more expensive first XI with very little depth versus Sunderland with a less expensive first XI but a deeper squad.

  • 4. “Mid-Table security” (10th and 11th): Newcastle and Fulham

We are 95% certain that there are only 12 teams that can possibly finish in the top 10.

The teams from these top 4 bands are joined only by Wolves.

  • 5. “Survival as priority” (12th-16th):  Wolves, Wigan and Stoke then Blackburn and Bolton

Five clubs covered by just £14m CTPP.  We are 95% confident that four of them cannot finish in the top 10 and all have MSq£ close enough to the promoted sides to make a season long relegation battle likely.  Lady Luck and injuries will dictate their finishing position as much as the MSq£ model.

The interesting team to watch here will be Stoke who look closer to the Soccernomics wages model rather than MSq£.  Can they follow in the footsteps of Sam Allardyce’s Bolton team of the early noughties?

[PT: Their style of football is incredibly effective at overachieving, but is one that will surely need refining for loftier ambitions. But as with the Oakland A’s in Moneyball, expecting any more than healthy overachievement is unfair; winning titles is not realistic, but doing much better than they should be certainly is.]

  • 6. “Relegation Favourites” (17th to 20th): West Brom, QPR, Norwich and Swansea

If a club wants to have a better than even chance of avoiding relegation it needs MSq£ of at least 0.2.  Currently West Brom are fractionally over this threshold at 0.23 and QPR are the highest of the promoted clubs at 0.13

As it stands, QPR need to sign about £10m worth of players to reach this level, Norwich £12m and Swansea £14m.  However due to the way MSq£ works as an average of all squad costs in the league, it’s more likely that even if one of the promoted clubs did go on a spending spree, at least one other team in the league would do the same and they still wouldn’t be over the 0.2 threshold.

Regarding the subjective predictions of the TPI Team, we have 15 teams either the same as their MSq£ rank or within 1 place

The only exceptions are:

  • West Brom We suggest 14th instead of 17th
  • Man Utd We think 1st instead of 3rd
  • Fulham We think 9th instead of 11th
  • Chelsea We think 3rd instead of 1st
  • Newcastle We think 12th instead of 10th

It will be interesting to see if we outperform the mode or prove horribly mistaken.

1 All of the last 9 seasons have been significant to α = 0.05, with 5 of the 9 being significant to α = 0.025
2 As of 8/8/11, the standard deviation of MSq£ is highest it’s been in Premier League history

Trackbacks/Pingbacks

  1. TPI Premier League Predictions « Joshua Kearney - August 12, 2011

    [...] The Transfer Price Index 2011/12 Premier League Predictions [...]

  2. Arsenal’s net spending of £48m in a decade puts them 9th in Premier League « Sporting Intelligence - August 12, 2011

    [...] economic-based predictions for the coming season have been published on the eve of the season. The Transfer Price Index prediction puts Chelsea at No1, with City No2 and United No3 – with QPR, Swansea and Norwich [...]

  3. Soccer: The Transfer Price Index 2011/12 Premier League Predictions » Stathead » Blog Archive - August 12, 2011

    [...] The Transfer Price Index 2011/12 Premier League Predictions: At Pay As You Play, we see what the 2012 EPL season will have in store for each team (according to Zach Slaton’s predictive model). [...]

  4. Predicting A Predictable Premiership | Sports Illustrated - August 17, 2011

    [...] these hardworking writers have done exactly that. The Transfer Price Index (TPI™) approach to predicting the league table uses squad value as a [...]

  5. The Rise and Fall of Arsene: An Analysis of Wenger’s Transfers and Results | Pay As You Play - September 7, 2011

    [...] of the horrible start to the season for the club, but I would stick by my assessment as part of the overall TPI pre-season predictions that Arsenal would finish 5th.  Sure, they’ve lost Nasri, Clichy, and Fabregas – three [...]

  6. The 2011/12 TPI Predictions: Post Summer Transfer Window Update | Pay As You Play - September 8, 2011

    [...] initial Transfer Price Index 2011/12 Premier League Predictions were published using squad data as of 11th August, some 3 weeks before the end of the transfer [...]

  7. Bang for buck: how a tech investor would view the Premier League table | Not 100% Accurate - October 3, 2011

    [...] Squad cost data by Transfer Price Index [...]

Leave a Reply