Back in February I started a multi-part statistical retrospective of the Africa Cup of Nations which served two purposes: it shed light on the practices of managers of the national teams in the competition, and it provided me with an opportunity to further develop and test the software library for the Football Match Result Database.  The recently completed UEFA Euro 2012 has provided another opportunity to refine the software and demonstrate its capabilities, and I start with a review of squad rotation practices at the tournament.

The analysis follows the same format that I used for the ACON 2012, but I account for substitutions this time.  (I still don’t factor suspensions and injuries as neither event is captured by FMRD.)  The first part of the analysis calculates the change in the number of players in a team’s starting lineup relative to the previous match.  The second part calculates the change in the number of players who made a match appearance relative to the previous match, so not just starters but substitutes as well.

So below is the table that addresses the first part of the analysis:

In contrast to the Africa Cup of Nations 2012, few of the Euro 2012 finalists rotated more than 15 players through the starting lineup.  Only four European teams rotated more than 15 players, while 10 of the 16 African teams exceeded that mark.  Upon inspection of the data there doesn’t appear to be much correlation between (starting) squad rotation and order of finish in the competition; national teams such as Republic of Ireland and Russia who were knocked out at the group stage had the fewest rotations (12), but semifinalists Portugal also had 12 and champions Spain 13.

I am struck by the practices of Spain and Germany, who were widely expected to meet in the final.  Spain arrived at a starting XI and made very few changes to it throughout the tournament, with a total of just four rotations over the entire competition.  Germany manager Joachim Löw also made few changes to their starting XI in the group stage, but made a large number of changes in the Greece and Italy matches.  The other two semifinalists also had divergent squad management practices; Portugal’s rotation matched Spain’s very closely while Italy had just as many rotations as Germany, but those happened in the third group match (against the weakest team in the tournament) and the quarterfinal match against England.

The second part of the analysis looks at the total number of rotations between starters and substitutes, which is displayed in the table below:

When substitutes are included it’s clear that the European finalists are indeed using most of their squad, even if some players only entered via a match substitution.  Spain’s rotation policy becomes clearer.  While Spain used a solid base of 13 starters, Vicente del Bosque used his substitutions to rotate a constant number of players into the squad from match to match.  The result was that 18 players appeared at least once in a match for Spain.  Italy also rotated a large number of players, but they were all within the subset of starting players, as I’ll show later.  Netherlands changed their starting lineup over their three group matches but kept roughly the same core of starting and substitute players over those matches.

The third part of the analysis looks at the size of the ‘bench’ of the Euro 2012 finalists.  ‘Bench’ isn’t an accurate term, and I’ll think of a better one later, but it is a measure of the number of role players who didn’t start yet made an appearance in a Euro match (I didn’t use number of minutes played):

Once again, there doesn’t seem to be much correlation between the size of the bench and the finishing order of the national teams.  All of the players used by Italy made at least one starting appearance, as did almost all of the players used by Germany.  Yet Spain and Germany made heavy use of ‘role’ players, as did Netherlands.  Perhaps number of minutes played by substitutes or ‘role’ players could illuminate the teams’ rotation practices.

So while it’s interesting to look at how certain managers incorporated their squads at the Euro, over this short tournament the numbers don’t reveal much of a correlation with performance.  One would think that if a manager is selecting a 23-man squad that he would want to use as much of it as possible, but in reality they have a starting XI and prefer to stick to it.  That seems to be true in European football judging from these numbers.