Optimizing the Line-Up in Soccer with Real-Life Data

This thesis treats the optimization of the line-up in soccer with real life data. The real life data used in this thesis consists of 40 matches played by Ajax in 2011-2012. The data are provided by Ortec TSS. With the data we created three profiles that will be matched to determine the optimal line-up for Ajax. A player profile containing the grades for each skill, position profile containing the grades of the skills required to play at a position and a team profile which adds additional sets of constraints to the Integer Programming model based on the team statistics. The team profile in particular is unique in the optimization of the line-up. These three profiles are matched with an IP model to optimize the line-up. The results obtained with the model are almost equal to the preferred line-up of the coach in 2011-2012. The only disadvantage of the methods used is that some players are overvalued or undervalued at some positions. This can be caused by the fact that the data we possess does not contain the physical skills of the players, only technical skills are included.