Taylor et al. (2005) examined the tactical aspect of soccer performance at playing position level by constructing spatial profiles of on the ball behaviours for different pitch areas which did not include off the ball movements, something Reilly and Thomas (1976) suggested accounts for around 98% of player movements. This paper addressed this issue by examining both on the ball behaviours and off the ball movements. At the simplest level there appeared to be a reasonable relationship between on the ball behaviours and off the ball movements with correlation coefficients ranging between 0.53 and 0.89. However, it was recognised that players did not operate in some areas of the pitch to any great extent i.e. sometimes no on the ball behaviours and a very small proportion of the overall time (<5%) spent in some areas. When these areas were discarded from the analyses the relationship between on and off the ball variables sometimes changed dramatically e.g. the full back`s correlation changed from 0.74 to 0.18. This would suggest that at least for some situations the frequency of on the ball behaviours does not correspond to the amount of time spent in a particular area. Thus a single analysis of either on the ball behaviour or a work rate analysis does not give the full picture. In the example provided here it would seem that the full back only went into some areas to perform on the ball behaviours whereas in other areas a great deal of time was spent in off the ball activity. This type of insight suggests that both on and off the ball behaviours should be analysed to provide details of the interaction between the two types of activity. The extent to which match conditions e.g. match status or quality of the opposition (Taylor et al., 2008) affect this relationship should also be examined.
© Copyright 2008 World Congress of Performance Analysis of Sport VIII. Published by Otto-von-Guericke-Universität, Department of Sports Science. All rights reserved.