Skip to main content

The analysis of playing styles across different developmental stages of football.

Dayus, J., 2019. The analysis of playing styles across different developmental stages of football. Doctoral Thesis (Doctoral). Bournemouth University.

Full text available as:

DAYUS, Jacob_Ph.D._2019.pdf



Until recently, playing styles within football have been assessed on subjective perceptions of performance, with limited performance indicators implemented to quantify these tactical aspects of performance. Furthermore, notational analysis research to date has often overlooked Youth football, predominantly examining First teams in elite competitions. As a result, the current study aims to identify key performance indicators that effectively describe the tactical patterns of play and compare these components of performance across critical stages of footballing development (Under 16s, Under 18s and First teams). Three thousand, three hundred and eleven final third entries were analysed over the course of 45 matches, equally distributed across three age-groups, during the 2018-2019 season. The sample consisted of 10 Under 16 teams, 16 Under 18 teams and 16 First teams Both univariate (Kruskal-Wallis H tests & Mann-Whitney U tests) and multivariate data analysis (Principle Component Analysis) methodologies were employed to effectively compare playing styles across the age groups. Results found statistically significant differences across a variety of performance indicators, compared with the younger age groups, the First team were observed to display a significantly greater number of; final third entries (H2= 7.242, p=0.27), passes (H2=7.371, p=0.025), crosses (H2=18.880, p<0.01), forward-diagonal actions (H2=6.392, p=0.041), ball recoveries via a loose-ball (H2=9.906, p<0.01) and build-up play in ‘Wide Areas’ (H2=16.475, p<0.00). In addition to this, crossed assists were more apparent in First team performances (H2=9.328, p<0.01), along with assists from ‘wide areas’ (H2=7.701, p=0.021) and possessions that were interrupted or lost (H2=10.025, p<0.01). Interestingly, Under 18s were seen to bridge the gap in many of the measured metrics, including final Third entries, number of crosses, backward and forward actions, ball recoveries via tackles and loose- balls, dribbled assists, possession outcome. It was however identified that Under 18s exhibited significantly more regains in the midfield zone (H2=11.512, p<0.01), than the other age groups. Under 16s on the other hand, were observed to be inferior to the older ages groups in many of the performance indicators, however it was found that they displayed a significantly greater number of dribbles (H2=11.221, p<0.01), backward actions (H2=8.172, p=0.017), forward actions (H2=6.576, p=0.037), ball recoveries via tackles (H2=9.420, p<0.01), build-up actions in the defensive zone (H2=8.416, p=0.015), assists via a dribble (H2=8.801, p=0.012) and shooting opportunities created (H2=10.232, p<0.01). Principle component analysis identified 7 different factors that had eigenvalues of over 1.0 and accounted for 79.21% of variance within the dataset. These factors grouped performance indicators into clusters, that described different features of a team’s playing style. These were labelled as possession type (PC1), direction of attack (PC2), ball recovery width (PC3), build-up width (PC4), defensive build-up (PC5), reduced attacking-midfield build-up (PC6) and defensive pressure (PC7). Between age groups, the most apparent differences in playing style were in possession width (PC4), defensive build-up (PC5) and defensive pressure (PC7). The study found that there were significant differences in playing styles across the age groups, with First Teams deploying various methods of attack to penetrate the oppositions backline. This suggests that First teams tended to display more elaborate-based football that demonstrated their tactical prowess, utilising the width of the pitch in both offensive and defensive phases of the game. Under 16s on the other hand, were more aggressive in their time spent in and out of possession and were observed to be more reliant on individual ability, displayed by the significantly greater percentage of dribbles used and self- assisted goals. Under 18s were often seen to bridge the gap, demonstrated in build-up width (PC4) and defensive build-up (PC5) and were seen to utilise a more passive defensive approach that relied upon ball recoveries in the midfield zone. Ultimately, the differences in the patterns of play across the age groups, suggests that younger players are more independently focused when attempting to breakdown their opposition’s defence. Furthermore, younger age groups were not observed to utilise the entire space available which can be found in the width of the pitch, suggesting that younger teams have a weakened ability to tactically influence the game, compared to their First team counterparts and are more reliant on an individual’s attributes. Therefore, it is suggested that coaches consider supplementing training methods to facilitate the education of player’s tactical understanding and ability.

Item Type:Thesis (Doctoral)
Additional Information:If you feel that this work infringes your copyright please contact the BURO Manager.
Uncontrolled Keywords:performance analysis; football; principle component analysis
Group:Bournemouth University Business School
ID Code:33041
Deposited By: Symplectic RT2
Deposited On:14 Nov 2019 13:56
Last Modified:14 Mar 2022 14:18


Downloads per month over past year

More statistics for this item...
Repository Staff Only -