Analysis of headers in high-performance football: evidence from the English Premier League

ABSTRACT

This study aimed to gather information about game situations where headers in high-performance football were performed and to characterise how headers were executed in different game situations. A multidimensional observational system was designed to characterise the header situation and technical execution. A sample of 920 headers were randomly collected from the English Premier League 2017–2018 season, using InStat Scout® website platform. Frequency analysis of headers showed that (1) most led to ball losses; (2) almost half was performed during set plays, especially in goal kicks and throw-ins; (3) a substantial amount was executed to prevent spatial progress of opponents when employing direct attacks with long passes. Chi-square analyses revealed significant associations (p < 0.05) between the following variables: header purpose and player position (large effect size – ES), movement and jump type (medium ES), game state and player movement, game state and jump type, space of occupation and opponent players as well as header type, header purpose and jump type, header purpose and opponent players, pass and jump type, pass and opponent players, and pass and player position (small ES). Based on the current findings, coaches are strongly encouraged to design representative training environments for headers considering position-specific needs, and to promote practice tasks and game strategies for maintaining/regaining ball possession after a player’s intentional header.

KEYWORDS:

• Soccer
• performance analysis
• heading
• situational variables
• technical component

Acknowledgments

The authors would like to thank InStatSport for providing the video material for the study.

Disclosure statement

The authors have no potential conflict of interest and no funding was obtained for the preparation of this article.

Notes

1. Seabra and Dantas (2006) use the term Space of Defensive Occupation (SDO), which in our study was replaced by Space of Occupation (SO).

2. Adapted from Seabra and Dantas (2006). Zone B (area behind the defensive sector) united with D since counted only 0.6% of all the headers.

3. P1: Area of the small box. P2: Area from the edge of the small box to the penalty spot. Width of the small box. P3: Area from the penalty spot to the penalty area line. Width of the small box. Both defensive and attacking headers in the centre and the sides of the penalty area were united because of low frequencies.

4. Control class included in Head pass since counted only 2.6% of all the headers. Shot class kept due to its importance. Diving header included in Standing class since counted only 1.1% of all the headers.

5. Shot class was not calculated since there were no shots in the reliability test sample.

6. In variable movement, classes step(s) sideways and step(s) backwards were united to achieve strong agreement and because step(s) backwards appeared only eight and six times in first and second coding, respectively.

7. Degrees of freedom (DF) for Cramer´s V were counted as (R-1) or (C-1) (R = Rows, C = Columns), where DF was always the smallest number (R-1) or (C-1) (Gravetter & Wallnau, 2013).