With the second international break well underway, we are now at a point where we have a big enough sample of games for the 2021-22 season that we can start highlighting early trends. Below is a scatter graph detailing the average number of changes made to the starting line-up per game for Premier League teams so far this season. On the x-axis is the average number based on all competitions, while the y-axis has the average number for league games only.

What is shows is the differing approach of teams to rotation early on in a season. For example, Tottenham have comfortably made the highest number of changes to their XI in all competitions, while Crystal Palace have made the least. Alternatively, Thomas Tuchel’s Chelsea has changed its starting line-up the most in the league so far, while David Moyes’s West Ham has remained practically unchanged.
When we break down the axes individually, we can see more clearly: (1) the relative position of teams in comparison to one another; and (2) how far or below the overall average they were. Here is the breakdown for average number of changes per game based on all competitions:

And here is the breakdown of average number of changes per game for league games only:

There are some surprising trends emerging already. The most notable is how middle-of-the-pack Manchester City have been rotation-wise early on. Manager Pep Guardiola is famous for his prolific tinkering, causing much disdain for fantasy footballers attempting to predict who will start each game. However, he has been uncharacteristically restrained in the early period this year, with his team sitting 16th for average in all competitions and 10th for average in the league. This is despite having to negotiate seven games in three weeks during the period between the first and second international break.
In fact, the majority of the teams playing in European competitions this season have comparable averages to those that are not. The exception to this is Tottenham (8.45), who rotated heavily in the early qualifying rounds of the European Conference League. Excluding Nuno EspĂrito Santo’s side, those teams playing in Europe – Chelsea, Leicester City, Liverpool, Manchester City, Manchester United and West Ham – have made an average of 4.61 changes per game in all competitions, whereas the thirteen teams not in Europe have an average of 4.63. Whether this continues as the gap between the number of games played by the two groups grows is worth keeping an eye on.
But what it does do is highlight how rotation is not solely a concern for the big teams. Every manager has a squad that must be utilised effectively across a season. If you can keep key players fresh and minimise injuries in the first team, you increase your chances of winning games and being successful. This is as relevant for a club aiming for mid-table and a good cup run as it is for a club tipped to win multiple trophies.