Rotation practices at the 2012 Africa Cup of Nations

Continuing my statistical review of the 2012 Africa Cup of Nations (as well as continuing my development and testing of my FMRD software library), i start a multi-part study of the rotations made in the starting lineup by the managers in the competition.  Squad rotation is a necessity for managers in the modern game as squads expand, the number of matches inflates, and the time between matches shrinks.  It can be enforced by injury or suspension, and it can also be part of a systematic strategy by the manager.  (Of course, fans will tell you that some changes are not necessarily systematic!)

I am going to start with a simple analysis — how many players in a team’s starting lineup have been changed from one match to the previous one?  I’m putting aside the issue of substitutions, injuries, or suspensions — FMRD does not capture injury events nor suspensions (but you can derive suspensions using the disciplinary records and some competition-specific rules).  Every team in ACON is limited to 23 players, but I decided to include the number of players who started at least one match in the competition, which gives a proxy measure of team depth but is less precise than other metrics.

Below is a table that summarizes squad rotations of the AFCON finalists:


# Changes in Starting Lineup from Previous Match
Group Match 2 Group Match 3 QF SF Final/3rd Total Actives
Côte d’Ivoire 1 9 6 3 0 19 20
Tunisia 2 8 4 14 20
Ghana 3 2 2 3 4 14 18
Mali 1 3 2 2 4 12 18
Morocco 4 7 11 20
Senegal 3 6 9 19
Gabon 0 5 3 8 16
Equatorial Guinea 0 3 4 7 15
Burkina Faso 3 4 7 16
Botswana 3 4 7 17
Zambia 0 1 0 2 2 5 13
Niger 5 0 5 16
Angola 2 3 5 14
Libya 2 2 4 14
Sudan 2 1 0 3 14
Guinea 2 0 2 13

As you might expect, the higher-profile national teams in Africa employ a more active squad rotation policy and have a larger pool of active starting players to choose from.  There appears to be a noticeable relationship between the size of the active squad and a national team’s FIFA ranking (in this case, the January 2012 rank before the continental tournament):

AFCON 2012 Squad Rotation Versus FIFA Rank

There was also an inflated number of changes made in the third group stage game by teams that were already eliminated from the competition or qualified for the knockout stage.  Of those teams that made significant changes on the third matchday, almost all of them made at least three or more changes for the quarterfinal match.

Of the four semifinalists, Zambia’s rotation policy stands out.  Basically, they didn’t have one!  Hervé Renard’s team was kept almost constant from game to game and used just 13 different starting players, tied with Guinea for the fewest in the competition.  It was a strategy that left Zambia with little margin in case of injury or suspension, but it did illustrate the fact that Zambia’s side was one that was built over a long period of time.

I haven’t accounted for substitutions or minutes played by all participants in the competition, who could yield some more understanding of each finalists’ rotation policy.  But this initial post does make for a good start.

(UPDATE: I included a plot in the post that I wasn’t able to insert last night due to technical reasons.  It should illustrate the relationship between national team rank and active squad size.)