It’s that time of year again when the football calendar starts to get congested and the games come thick and fast. I am of course referring to Christmas, the most wonderful time of the year for football supporters. The narrative for those who follow professional football in England is that the absence of a winter break and the relentless schedule mean that clubs, squads and players are stretched to the limit. While the rest of us are enjoying the festive season, managers are struggling with the impossible task of juggling their playing resources to avoid burnout, injuries, excessive travel and the psychological strain of being away from their friends and family. Squad rotation, we’re told, is the inevitable consequence.
While the headline writers and 24 rolling news outlets enjoy the speculation, the constructed storylines and the discussion in order to create valuable content to fill the airtime and column inches, the Fantasy Premier League (FPL) managers struggle with the task of trying to pre-empt who is going to be rested and when in order to put out a team of 11 players. I myself have been a vocal advocate of squad depth in the festive period, espousing the virtues of a 15 man squad to cope with the inevitable disappointment when a key player is dropped. However, in amongst this received and accepted wisdom, it didn’t occur to me to check whether it was true; is there really greater rotation at Christmas?
Player Rotation, 2014-16
To begin with, I’ll clarify what I mean by the ‘festive period’:
- GW16: 13th – 14th Dec
- GW17: 17th – 19th Dec
- GW18: 26th – 28th Dec
- GW19: 30th Dec – 01st Jan
- GW20: 02nd – 04th Jan
This period looks brutal from an FPL planning perspective, with five gameweeks in little over three weeks and barely any turnaround between most of them. The natural assumption is that managers will rest and drop players accordingly to afford them a break.
To test this assumption, I’ve looked at what happened during the same period over the last two seasons. The charts below show the number of players used by the 20 Premier League clubs over five gameweek ranges throughout the season, with the festive period highlighted in red.
Figure 1: Count of players used, 2014/15 (all clubs)
Figure 2: Count of players used, 2015/16 (all clubs)
What the data shows us is that in the last two seasons there has actually been a decrease in the number of players utilised over the Christmas period compared to earlier in the season. In 2014/15 the total number of players dropped from 390 in the five weeks preceding Christmas (an average of 19.5 per team) to 384 over the holiday season (19.2 per team). In 2015/16, there was a fall from 379 (18.95 per team) to 367 (18.35 per team). It is interesting to note that there were far more players utilized during the 2014/15 season, indicating that rotation during the whole season was more prevalent than it was during the 2015/16 season.
There is a contrasting pattern between the seasons when we strip out cameo appearances. In 2014/15 the Christmas period saw 314 players contribute at least one 60 minute appearance in the five gameweeks, a fall of six against the previous period, whereas in 2015/16 the figure increased by three, from 305 to 308. These figures though are not particularly erratic, and in practical terms there was little change between the two periods in both seasons.
This tells us that the number of players being used by managers is not affected by the Christmas period; there is no evidence that managers are calling up youngsters or reserves to plug the gaps in the squad caused by first team players falling foul of fatigue or injury.
However, the question that FPL managers are interested in is not whether my player will play during the Christmas period, but will he be dropped for any of the games? For that, we need to look at the number of players who played every game during a given period (note that the distribution of double / blank gameweeks from gameweek 26 onwards skews the data in these periods).
Figure 3: Count of players who played every game, 2014/15
Figure 4: Count of players who played every game, 2015/16
We can see here that there is again a contrasting pattern between the seasons. In 2014/15, the number of players playing in all his team’s matches over Christmas was at the lowest rate all season, with only 94 (4.7 per team) contributing at least 60 minutes in each game. However in 2015/16, the Christmas period saw the first teams at their most settled, with 119 (5.95 per team) bettered only by the opening period of the season (120).
It is difficult to draw any conclusions from this data, which in itself is interesting. We can say that teams were a lot more settled in 2015/16 than they were the previous season where more players played but fewer were guaranteed starters, and this is somewhat reflected in the Christmas periods for both seasons. However, the extent of the impact that Christmas leaves is far from conclusive with volumes of players starting all matches during this period not noticeably deviating from the overall trend. Conventional wisdom would suggest that 2014/15 is closer to what we would have expected to see (e.g. fewer guaranteed starters), but the common myth in the minds of FPL managers is that there would be a pronounced dip in this period, whereas the reality is that the average team will go from having 4.95 consistent starters to 4.7, a barely noticeable drop.
Figures 5 and 6 below show that there is no pattern between teams; 45% of teams have more settled teams at Christmas, 45% have less settled teams, whilst 10% won’t change their rotation patterns from that seen in GW11-15.
Figure 5: Change in number of players starting all fixtures in GW16-20 vs. GW11-15, 2014/15
Figure 6: Change in number of players starting all fixtures in GW16-20 vs. GW11-15, 2015/16
The main lesson I’m taking from this research is that the Christmas Rotation is not a phenomenon that is clearly evident. Rotation is present certainly. However, data suggests that over the past two seasons it hasn’t been any more or less pronounced than what would have been expected based on what has gone before in the season. From an FPL perspective, I would not disagree with the old truism; that having a deep squad of 14 or 15 regular starters is important because they are likely to be called on over the Christmas period as Premier League managers look to rotate. The second unspoken part to that approach that is highlighted here is that the risk of rotation is not greater just because Santa Claus is coming to town but is present throughout the season. How much importance you choose to place on the strength of your bench in FPL is a judgment call, but it is important to note that it does not need to be a seasonal concern. The prospect of your players being rotated is for life, not just for Christmas.