Favouritism in the allocation of additional time: The case of the English Premier League


Media and spectator attention in English Premier League football games has never been as intense. Constant improvements in technology have facilitated deep analysis of all aspects of the game. This is certainly true of referee decision-making, which has become the subject of much debate in both academic and media circles. The examination of referee decision-making has ranged from the likelihood of a referee awarding a penalty to the visiting team, to the probability of a referee being more susceptible to influence from home supporters located in closer proximity to the playing pitch.

This article examines the impartiality of the referee when applying ‘Law 7’ of the Laws of the Game, the allocation of additional time, at the end of the second half of matches in the English Premier League during the 2009/10 season. The research tests whether referees respond differently under changing conditions and demonstrate favouritism towards particular teams. Two forms of favouritism are examined: home favouritism and ‘big’ club favouritism.

Literature on referee decision-making

Mainstream agency theory is concerned with individual behaviour in light of incentive structures. From an economic perspective, referees act as agents. Decision-making is their most important job (Helsen & Bultynck 2004; Sutter & Kocher 2004), and this task has become more challenging over recent decades as the pace of professional football has increased dramatically (De Oliveir et al. 2011). This has required referees to be physically, technically and psychologically adept in order to meet the demands of professional football and administer the rules of the game effectively. According to Helsen & Bultynck (2004) referees in top European matches make 137 observations on average per game. While some decisions are clear-cut, others may be guided by subjective judgement (Dawson et al. 2007; Buaraimo et al. 2007; Boyko et al. 2007; Dawson & Dobson 2010).1


This analysis covers a total of 380 English Premier League games from August 2009 until May 2010 using data collected from the British Broadcasting Corporation (BBC). Data is collected for the total number of seconds of additional time played at the end of the second half in each game; the venue and score at the end of ‘normal’ time; the number of goals between teams at end of ‘normal’ time; the score and outcome at full time; and the number of goals scored in the second half.


Table 1 presents the mean number of minutes and seconds of additional time, play during the second half for the 2009/10 season, when the home team is winning, losing or drawing.

Table 1 Mean additional time for all 20 clubs

Home team winning Home team losing Home team drawing
4 minutes and 22 seconds 4 minutes and 27 seconds 4 minutes and 31 seconds

These findings suggest very little, if any evidence, of home bias in the allocation of additional time. Further investigation is required to consider if a home bias exists. It is expected that closers games, i.e. games that are level at the end of normal time or games where one goal can alter the result of the game, would be more intense at the end of normal time than games where the outcome is almost a foregone conclusion. This could result in referees behaving differently if they are influenced by stakeholders such as players, managers or supporters.

Figure 1 presents data on closer games, where the home team is lead or losing by two goals or less at the end of normal time (-2 to +2 margin).

Figure 1 Mean second half additional time by score margin

Again, no bias is evident in the allocation of additional time towards home teams when only one goal separates teams at the end of normal time. As expected, home teams do have to play less time when winning by 2 goals at the end of normal time and are granted more time when losing by the same margin.

In order to test whether a ‘big’ club bias exists, a definition of what constitutes ‘big’ is required. For the purposes of this research a club can be regarded as ‘big’ if they have played in every Premier League season since 1992 and have won at least one European club trophy, or are listed in the world top 20 football clubs for total revenue in the 2010 Deloitte Football Money League and have won at least one European Club Trophy. Eight Premier League clubs meet one or both of these criteria and are hereafter known as big clubs.2 The mean number of seconds played when winning, drawing and losing, for each of the twenty English Premier League clubs is displayed in Tables 2, 3 and 4.3

Table 2 Mean 2nd half additional time in seconds when winning at home

Team Winning
Stoke City 305
Sunderland 297
Wigan Athletic 284
Birmingham City 282
Burnley 282
Tottenham Hotspurs* 281
West Ham United 281
Wolverhamption Wanderers 280
Everton* 277
Portsmouth 273
Liverpool* 272
Arsenal* 268
Manchester City* 255
Bolton Wanderers 254
Hull City 252
Blackburn Rovers 251
Aston Villa* 249
Manchester United* 236
Chelsea* 234
Fulham 277

Table 3 Mean 2nd half additional time in seconds when losing at home

Team Losing
Arsenal* 319
Hull City 312
Tottenham Hotspurs* 311
Chelsea* 308
Manchester City* 286
Liverpool* 285
Everton* 282
Wigan Athletic 275
Manchester United* 267
Blackburn Rovers 264
Burnley 261
Portsmouth 256
West Ham United 254
Birmingham City 252
Bolton Wanderers 249
Sunderland 248
Aston Villa* 241
Fulham 233
Wolverhampton Wanderers 224
Stoke City 216

Table 4 Mean 2nd half additional time in seconds when drawing at home

Team Winning
Arsenal* 351
Manchester United* 317
Manchester City* 307
Chelsea* 305
Tottenham Hotspurs* 296
Portsmouth 290
Birmingham City 287
Stoke City 281
Hull City 275
Wigan Athletic 273
Everton* 271
Sunderland 264
Bolton Wanderers 260
Aston Villa* 257
West Ham United 257
Wolverhampton Wanderers 254
Liverpool* 242
Fulham 231
Blackburn Rovers 218
Burnley 206

An examination of Tables 2, 3 and 4 provides descriptive evidence that a big club bias may exist. When winning at home, smaller clubs tend to play more additional time, with Stoke, Sunderland, Wigan Birmingham and Burnley all at the top of the list. When losing at home, the bigger clubs get to play more time, with Arsenal, Tottenham, Chelsea, Manchester City, Everton and Liverpool occupying six of the top seven places on the list. This trend continues when drawn matches are considered; the top five places are all occupied by ‘big’ clubs.

Further analysis is presented in Table 5, which displays the mean difference between additional time in seconds when drawing and losing at the end of normal time minus additional time in seconds when winning a game. Larger positive scores are more favourable than larger negative scores. Again, the bigger clubs tend to do better – Chelsea, Arsenal, Manchester United, Manchester City and Tottenham occupy five of the top six places on the list. The bottom eight places are taken up by eight of the Premier League’s smaller clubs.

Table 5 Mean difference in added time when losing or drawing minus when winning

Team Difference
Chelsea* +72.5
Arsenal* +67
Manchester United* +51
Manchester City* +46.5
Hull City +41.5
Tottenham Hotspurs* +22.5
Fulham +5
Bolton Wanderers +0.5
Portsmouth 0
Aston Villa* 0
Everton* -0.5
Liverpool* -8.5
Wigan Athletic -10
Blackburn Rovers -10
Birmingham City -12.5
West Ham United -25.5
Sunderland -41
Wolverhampton Wanderers -41
Burnley -48.5
Stoke City -56.5

Given the reported findings in Tables 2 to 5, there is evidence of favouritism shown to bigger clubs, particularly when drawing and losing at the end of normal time, further examination of the big club small club bias is required.


Given the data presented in Tables 2 to 5, an alteration is perhaps required to Law 7 of FIFA’s The Law of the Game so that greater transparency exists regarding the allocation of additional time. The law states that “an allowance [for additional time] is to be made only when these [stoppages] delays are excessive” (FIFA 2009: 104). This explanation is too vague and requires tightening. Public display of the official time could be made compulsory at all Premier League games so players, managers and spectators know when stoppages have occurred and how much normal time remains.  Should additional stoppages occur in the allocated additional time, the countdown of additional time could simply be stopped if required.

Additionally, there is no reason why the referee should continue to act as timekeeper. This occurrence is an accident of history. Allowing referees to officiate without the pressure of timekeeping or allocating additional time should reduce the pressure they are under from players, managers and spectators. An anonymous timekeeper could appointed by the Premier League for each game and control timekeeping without having pressure exerted on them by external stakeholders, reducing the incidence of any bias that may arise.


Boyko, R, A Boyko & M Boyko (2007), “Referee Bias Contributes to Home Advantage in English Premiership Football”, Journal of Sports Sciences 25, 1185-1194.

British Broadcasting Corporation, (2011), Premier League Football Results [accessed at http://www.bbc.co.uk/sport/football/premier-league/results].

Buraimo, B, D Forrest & R Simmons (2007), “The 12th man? Refereeing Bias in English and German Soccer”, Journal of the Royal Statistical Society: Series A (Statistics in Society) 173(2), 431-449.

Dawson, P & S Dobson (2010), “The Influence of Social pressure and Nationality on Individual Decisions: Evidence from the Behaviour of Referees”, Journal of Economic Psychology 31, 181-191.

Dawson, P, S Dobson, J Goddard & J Wilson (2007), “Are Football Referees Really Biased and Inconsistent? Evidence on the Incidence of Disciplinary Sanction in the English Premier League”, Journal of the Royal Statistical Society: Series A (Statistics in Society) 170, 231-250.

De Oliveria, M, R Orbetelli & T De Barros Neto (2011), “Call Accuracy and Distance from the Play: A Study with Brazilian Soccer Referees”, International Journal of Exercise Science 4(1), 287–295.

Dohmen, T (2005), “Social Pressure Influences Decisions of Individuals: Evidence from the Behaviour of Football Referees”, IZA Discussion Paper No, 1595.

FIFA (2009), The Laws of the Game, Fédération Internationale de Football Association, Zurich, Switzerland.

Garicano, L, I Palacios-Huerta & C Prendergast (2005), “Favouritism Under Social Pressure”, Review of Economics and Statistics 87, 208-216.

Helsen, W & J B Bultynck (2004), “Physical and Perceptual-Cognitive Demands of Top-Class Refereeing in Association Football”, Journal of Sports Sciences 22, 179-189.

Johnston, R (2008), “On Referee Bias, Crowd Size, and Home Advantage in the English Soccer Premiership”, Journal of Sports Sciences 26, 563-568.

Morris, D (1981), The Soccer Tribe, London, Cape.

Nevill, A, N Balmer & A Williams (2002), “The Influence of Crowd Noise and Experience Upon Refereeing Decisions in Football”, Psychology of Sport and Exercise 3, 261-272.

Pollard, R (1986), “Home Advantage in Soccer: A Retrospective Analysis”, Journal of Sports Sciences 4, 237.

Pollard, R (2008), “Home Advantage in Football: A Current Review of an Unsolved Puzzle”, The Open Sports Sciences Journal 1, 12-14.

Rickmann, N & R Witt (2005), “Favouritism and Financial Incentives: A Natural Experiment”, Journal of American Statistical Association 89, 1124–1127.

Scoppa, V (2008), “Are Subjective Evaluations Biased by Social Factors or Connections? An Econometric Analysis of Soccer Referee Decisions”, Empirical Economics 35, 123-140.

Sutter, M & M Kocher (2004), “Favouritism of Agents – The Case of Referees’ Home Bias”, Journal of Economic Psychology 25, 461-469.

1 For a detailed analysis of this area readers should see Morris (1981) Pollard (1986), Nevill, et al. (2002), Sutter and Kocher (2004), Dohmen (2005), Garicano, et al. (2005), Rickman & Witt (2005), Boyko, et al. (2007) , Buraimo et al. (2007), Dobson, et al. (2007), Dawson, et al. (2008), (Johnston, 2008), Pollard (2008), Scoppa (2008) and Dawson and Dobson (2010).
2 Arsenal, Aston Villa, Chelsea, Everton, Liverpool, Manchester City, Manchester United and Tottenham Hotspurs
3 * denotes a ‘big’ club, as describe by our definition, in Tables 3, 4, 5 and 6.