MICHAEL BLAUVELT — Ever wondered whether winning teams committed fewer fouls than their opponents?  Or whether draws really were as boring as they appear?  I did.  So I compiled a variety of statistics FIFA provides about the 2010 World Cup, put them together into a spreadsheet, and crunched the numbers.  Today, an overview of the stats for shooting, actual playing time, and fouls committed.

Shooting

It should come as no surprise that winning teams put a higher percentage of their total shots on goal (45%) than their opponents (32%).  And when you put a higher percentage of shots onto goal, you increase your percentage of shots that score (winners scored on 13% of their shots and losers on 3%). 

But what may not be so obvious is that both winners and losers at the World Cup averaged about the same number of minutes of Actual Playing Time (time when they were actually in control of the ball and ‘playing’) per shot.  Winners took a shot every 2.5 minutes of APT; Losers ever 2.8 minutes.  But frequent shots aren’t necessarily accurate shots.  It took 10.2 minutes of APT for a loser to put a shot on goal.  Winners took only 5.8 minutes per shot on goal.

Draws are interesting (at least on paper).  Teams that ended a match tied (without going to penalty kicks) took an average of 3.4 minutes per shot and 11 minutes per shot on goal.  That means that during a draw, the teams both attacked less (on average) than the average winner or average loser.  And in a draw, both teams averaged scoring on only 6% of their shots (the historical average for soccer is 10%, though the 2010 World Cup average was 7%).

Actual Playing Time

The 2010 World Cup was the first tournament for which FIFA made “Actual Playing Time” statistics available.  Ball Possession percentages are interesting, but everyone knows that there typically aren’t 90 minutes of actual gameplay in a soccer match.  APT helps account for that.  Add the two teams’ APT together and you have a good sense of exactly how long the teams were “playing.”  It also helps us identify how big an advantage in time one team had over its opponent.

At the 2010 World Cup, winning teams had an average APT of 36.1 and losing teams had an average of 32.8, so roughly 3.3 more minutes of APT for the winner (a 52-48 ball possession split).  It means the winner is probably going to get at least one more shot than its opponent (if shots occur every 2.5-2.8 minutes) and that the winner is likely to be fouled at least once more than the loser (see below).

Fouls Committed

During the 2010 World Cup, winning teams commited an average of two fewer fouls per match than their opponents (14.2 to 16.4).  But more interesting is that both winners and losers committed fouls with almost the same frequency: winners committed one foul for every 2.6 minutes of their opponent’s APT and losers committed one foul for every 2.7 minutes of thier opponent’s APT.  The rate of fouling in draws? 2.1.

ANALYSIS:

While these numbers can’t be used to predict winners, they help validate some impressions we take away from soccer matches:

1. Sometimes teams play for draws.  It took an average of 3.5 minutes for a team playing in a draw to take a shot (longer than either winners or losers).  While the teams weren’t shooting as often, they were fouling even more often: committing 17.2 fouls on average at a rate of one foul every 2.1 minutes of their opponent’s APT.
2. The team with fewer fouls committed wins more often.  You can chalk this up to the better team not needing to foul as much, referee bias, desperation by losers (trying to force something), etc.  But winners averaged fewer fouls than losers.  In the knockout stage, the team that fouled less won or went to a shootout 86% of the time. 

But I think we can make some other observations as well.

1. The frequency of fouls committed during games where there is a winner and loser is remarkably consistent.  It suggests that even the better teams can still only go so long on defense without committing a foul.  If this 2.6 minutes of opponent’s APT per foul is borne out over time, then it could be used to evaluate referee bias in specific matches.  Or make for a not-so-interesting party game in which you correctly guess the winner of the match simply by looking at the number of fouls committed.
2. Research has shown that your chances of scoring a goal are affected by the distance and angle of the shot.  So a team’s G/S ratio could reflect that the team had unnaturally good or bad luck.  Or simply that their opponent’s defense forced them to take low-percentage shots.  Either way, it appears to merit further reasearch.
3. If the average minutes of APT/S are roughly consistent at 2.5-2.8 in matches with a winner and the rate of fouling is consistent in those matches at one foul committed every 2.6 matches, it suggests that teams that win ball possession have a slightly better chance of victory.  It may mean simply one more lottery ticket with a 1-in-10 chance of winning (a 10% historical average of scoring a goal on a shot), but even that makes your odds of winning a little better.

by admin