Bayer Leverkusen

The Correlation Between Age of a Football Squad and its Success

There has always been a controversy between those who claim that age of a football squad holds a significant part in the long-term success of a club and those who insist that age is just a number with no effects in the in-game characteristics.

Therefore, in order to give an answer to the above mentioned argument, the example of Bayer 04 Leverkusen (abbreviated as Bayer from now on) is examined. Bayer is a German team which heavily relies on its youth system as it is assumed that this is an efficient way of developing a sustainable talent network of young players that would eventually step in the first team, hence enhancing the income and the sustainability of the club from possible transfers.

The starting point of this research would be the age profile of Bayer over the last decade; data is shown from season 2009/2010. It has to be mentioned that all data and performance statistics used concern exclusively league statistics and European matches as well as DFB-Pokal competition (German Cup) were not taken into account. Furthermore, each season consists of two calendar years but in this research they are expressed with only the first one (i.e. season 2015/2016 is just notated as 2015).

It is observed that age of the squad in the region of 22.8 and 23.9 is closely linked to the performance of Bayer throughout an entire season, expressed in its final league position. In this regime, having a relatively young aged squad enhanced the overall performance of the team.

Evidentially, in 2011 Bayer’s squad with an average age of 23.3 manages to finish as 5th in Bundesliga, while one year later they finished as 3rd having at the same time the youngest squad of this 8-year period (22.8 years old). On the other hand, big derivations in the average age are controversial. Remarkably, in 2010 Bayer finished as runners-up in Bundesliga while having the oldest squad (24.1 years old), in comparison with 2016 season that they finished 10 positions lower, with an average age squad of 22.8, even though other reasons may explain this downfall. Generally, it seems that old squads have an effect on Bayer’s performance as the younger the football team is the better the position at the end of the league.

To continue, as football becomes more and more fast paced, fatigue and its correlation with age should be considered as well. The approach to determine this statistical relationship is based on the amount of goals that Bayer had scored in the last minutes of the game (between 75’-90’) as a proportion of their total goals. The reason is that goals scored in this period are usually game changers and require plenty of energy deposits, both mental and physical.

With the exception of season 2016/2017, the results obtained are surprising. The average age of Bayer’s squad follows the same trend with the number of goals scored in the late parts of the game. For instance, in 2013 Bayer scored 17% of their total goals over the season with an average age about 23.5 years old, while one year later the team (average aged at 23.9) managed to score almost 1 out of 5 of their total league goals between 75-90 min., indicating that young aged players struggled more to put the ball in the net at the final stage of the match. A possible explanation of the behavior observed in season 2016/2017 is that only 29% of (home) exhibitions were actually over before the 75th minute as opposed to 2009 and 2012, seasons where the majority of matches was long before decided and the team was not pursuing a goal at the end.

The former result may indicate lack of maturity and mental attributes, characteristics that are linked with experience and are acquired through time. For this reason, the ability to overcome difficult situations should be examined with respect to age. And is there a more stressing situation than playing a home game in front of your fans and already have conceded a goal? Below the correlation between average age and points acquired from overturns is plotted.

It is again concluded that the average age of Bayer’s team follows the same trend with the amount of points taken after been behind at score. Older squads tend to be capable of reversing – partially and in some occasions totally – such situation and eventually get something out of the game. Indicatively, in 2010 Bundesliga runners – up managed to get 13 points with an average age of 24.1 (the highest in this 8 – year era) as opposed to Bayer’s team in 2014 where they got less than 10 points, finishing in the 4th place of the German league. In 2011, Bayer only managed to get a single point from overturning a home game and as a result finished 5th in the league, the second worst position of the team in this 8-year period. Another interesting point is that 5 years later they managed to get away with no less than 6 points despite the fact that the team’s average age was smaller. This could be attributed to the fact that only in a small proportion of home games (6 out of 17) throughout 2011/2012 Bayer actually fell behind in the scoreline; in 2016 the games were almost doubled (10 out of 17).

Furthermore, aggression regarding the average age is explained. Aggression and enthusiasm is a common characteristic of a young squad and can be expressed via the number of yellow cards per game. It should be noted however, that fouls and cards are directly linked with possession, as a team that holds the ball more is unlikely to commit many fouls, thus receiving cards and penalties.

To examine if changes in ball possession over the years are of significant importance, standard deviation (σ) and average value of possession (μ) are calculated. The value of μ is 51.98 while σ is 1.33. When the confidence level is chosen to be 99.9 % , using the t-distribution (with 7 degrees of freedom) the confidence interval is given by:

49.44<μ<54.52

As a result, it can be assumed that ball possession finds itself in a steady state over time. In this way, there is no need of adjusting the, obtained from literature, number of yellow cards per game.

The results indicate that a youngest squad has smaller probability of getting yellow cards throughout a game, as it could be seen from the similar trends that these two factors follow between 2010 and 2016. This behavior could be explained if one could understand that the majority of fouls committed and result in yellow cards in modern football is for the sake of stopping counter attacks that may prove crucial to the final outcome of the game. Lack of experience along with better physical characteristics (acceleration, pace) could constitute an explanation of the above mentioned state. It has to be mentioned however, that Bayer’s oldest squad which managed to finish runners-up in 2010 had a rather small proportion of fines per game and this could highlight even more the overperformance of the team this season.

To sum up, average age of Bayer’s squad was found to have some importance over the annual performance of the team, even though some controversial results are exported and further investigation should be carried out. On the one hand, older squads were capable of overturning a game that did not started as desired by also having the mental balance and the patience to score in its final stages. On the other hand, a young squad reclaims its enthusiasm to avoid some in-game mistakes expressed in the lower amount of yellow cards, and overall, Bayer had achieved higher positions in the league when young talents were used. However, this is not always the case as the best position of the club between 2009 and 2016 was achieved with an average age of 24.1; 0.7 points more than the mean value of these 8 years.


Bibliography

Pastemagazine.com. (2017). exclusive: How Bayer Leverkusen Outsmart their Wealthier Bundesliga Competitors. [online] Available at: https://goo.gl/1TwBEA [Accessed 9 Dec. 2017].

Transfermarkt.com. (2017). Football transfers, rumours, market values, news and statistics. [online] Available at: https://goo.gl/f3UHhz [Accessed 9 Dec. 2017].

Ltd, F. (2017). Footstats – Cards. [online] Footstats.co.uk. Available at: https://goo.gl/o1C6sY [Accessed 9 Dec. 2017].

Soper, D. (2017). Free Student t-Value Calculator – Free Statistics Calculators. [online] Danielsoper.com. Available at: https://goo.gl/bnT49R [Accessed 10 Dec. 2017].

As an engineer I have learned that it is not only about deriving the result but also explaining it. In Statathlon I found the unique opportunity to combine mathematics with my other passion, football, and strive to describe the phenomena that rule it.

Site Footer