The FIFA Virus is a term which readers who primarily follow the Premier League might not be too familiar with, but at the start of this decade, it was a popular theory in Spain’s La Liga. The theory is that the first league game after an international break is the best time to face a ‘big’ club, and on face-value, it appears entirely logical. The bigger clubs tend to lose more of their players, both in percentage and absolute terms, for international duty than smaller clubs do. Those players not only suffer the fatigue of additional matches, but also the effects of clocking up hundreds or even thousands of air miles, potentially crossing several time zones in the process. Those who travel the furthest may only be able to train once with their club teammates before the next league fixture, reducing the available time to specifically prepare for the opposition. If that opposition have few international players, they will arrive into the match fresh, perhaps even enjoying a few days on holiday before an extended period to prepare. An ideal set of ingredients for an upset, or at least, that’s how the theory goes.
There are unfortunately a couple of challenges that affect the statistical analysis of the FIFA Virus theory. One issue is how to you assess which clubs have suffered the most disruption during an international break? Losing 5 regular starters could be deemed a bigger disruption than losing 10 squad players, particularly at clubs who prioritise a consistent starting line up highly. Air miles are not a reliable indicator either, since a 1000 mile flight across multiple time zones will be more arduous than the same distance within a single time zone. The physical impact of the international fixtures will also not be the same. One player may enjoy an easy victory largely played in 2nd gear against an international minnow, whilst another may have to come through an intensely fought, bruising encounter.
Then we come to the issue which so often impacts statistical analysis in Football, sample size. Whilst the international calendar does occasionally change slightly from season to season, in 2018/19 there were 4 international breaks within the Premier League season, and that figure has rarely exceeded 6. Such a small sample size means that results may be unduly schewed by injuries, suspensions or opposition without an opportunity to balance out. Tottenham for example picked up 6 points from their 4 Premier League matches immediately following international breaks, a return comfortably lower than their points per game average across the rest of the season. However, attributing this drop to the FIFA Virus would be inaccurate given the fact that the 2 matches they lost were both against Liverpool, a club with a similar number of international players to Tottenham. Indeed, the fact that the sample size is too small is reflected by the table below, which shows you will find both 4 of last season’s top 6 and 4 of last season’s bottom 5 within the group of 10 clubs to record a better points per game average immediately after an international break than across the rest of the season.
|Club||Points per game average following international break||Points per game average across the rest of the season|
|West Ham United||0.75||1.44|
|Brighton and Hove Albion||1.25||0.91|
The easiest way to increase the sample size would be to investigate the FIFA Virus over multiple seasons, those this is not without issue. The random nature of the fixture list means that imbalances in the strength of the opposition faced immediately after an international break will not necessarily be resolved over multiple seasons. The table below showing Manchester City’s fixtures after the international break this decade show that whilst there are clubs who they have faced multiple times, there are also other clubs such as Chelsea they have never faced. Manchester City were chosen due to the fact that they are not only the most successful Premier League club this decade, but also the most consistent. Nevertheless, the composition of their squad has changed over the decade, as has the composition and relative squad strength of each of their opponents, which must be considered when analysing their results.
|Season||Opposition for league fixture following international breaks||Points per game average following international break||Points per game average across the rest of the season|
|2018/19||Fulham (H), Burnley (H), West Ham United (A), Fulham (A)||3.00||2.53|
|2017/18||Liverpool (H), Stoke City (H), Leicester City (A), Everton (A)||3.00||2.59|
|2016/17||Manchester United (A), Everton (H), Crystal Palace (A), Arsenal (A)||2.00||2.06|
|2015/16||Crystal Palace (A), A.F.C. Bournemouth (H), Liverpool (H), A.F.C. Bournemouth (A)||2.25||1.67|
|2014/15||Arsenal (A), Tottenham Hotspur (H), Swansea City (H), Crystal Palace (A)||1.75||2.11|
|2013/14||Newcastle United (H), Stoke City (A), West Ham United (A), Tottenham Hotspur (H)||2.50||2.24|
|2012/13||Southampton (H), Stoke City (A), West Bromwich Albion (A), Aston Villa (H), Southampton (A), Newcastle United (H)||2.16||2.03|
|2011/12||Wigan Athletic (H), Aston Villa (H), Newcastle United (H), Bolton Wanderers (H)||3.00||2.26|
|2010/11||Tottenham Hotspur (A), Blackburn Rovers (H), Blackpool (A), Fulham (A), Manchester United (A), Sunderland (H)||1.83||1.88|
The analysis of Manchester City’s results certainly do not support the idea of a FIFA Virus. Indeed, their record is remarkably strong immediately after an international break, with a superior points per game average than they achieved across their season during 6 of the 9 seasons. Amongst the 3 seasons when their return was worse after international break, the difference was minimal during 2010/11 and 2016/17. The 2014/15 season was the only season where the average was significantly lower immediately after an international break, a consequence of dropped points on the road against Arsenal and Crystal Palace. Across the 9 seasons, we have a sample size of 40 Premier League fixtures immediately after an international break to work with, and City have amassed 94 points across those fixtures at a points per game average comfortably higher than across the other 302 fixtures played during those seasons.
Obviously this is a study of a single club. Yet whilst the results could be considerably different at other big clubs, Manchester City are exactly the type of club you would expect to be affected by the FIFA Virus if such a thing exists. The results are therefore surprising given the fact that the theory sounds logical, and is a reminder of the danger of highlighting individual results rather than longer trends to test a theory.
Fantasy Premier League Tip – Matthew Lowton
Burnley are exactly the type of club one would expect to benefit from a FIFA Virus, and they travel to high-flying Leicester City this weekend. A Burnley win could only be considered a minor upset given they are currently 7th, 2 points behind Leicester, and the consistency of Sean Dyche’s team selections are always welcome for a Fantasy Premier League Manager. At £4.5 million, Matthew Lowton is the cheapest of Burnley’s regular back four, and should consistently pick up appearance and clean sheet points across the season.