Several researchers have examined the effect of particular training interventions on various physical capacities in football players. However, the presence of confounding factors can influence the results of the training study, for example obscuring important differences during comparative studies. Confounding occurs when one or more separate factors (that are of no interest to the research) influence the outcome variable that the intervention is designed to moderate. However, if known, potentially confounding variables can be controlled using experimental or statistical techniques. The purpose of this study was to examine the factors influencing the individual training response in football players. METHODS. 31 players [age 23(4)yrs, body mass 72(7)kg, height 178(5)cm] out of 46 assessed for eligibility completed the Yo Yo Intermittent Recovery (YYIRT) and a VO2max test at the end of the competitive season (end), at the beginning of the following season (pre), and after 7 weeks of training (post). During the training period heart rate and rating of perceived exertion (RPE) were measured. A hierarchical multiple regression model was applied with absolute pre-post change scores as dependent variable. All change scores entered in the models were corrected for the regression towards the mean (RTM) to control for statistical artifacts. The entering order was established based on collection feasibility of the variables: step 1) age; 2) pre-training values; 3) training load (TL) quantified with heart rate (TRIMP) or session-RPE relative to physical, technical training sessions and matches; step 4) changes in YYIRT or VO2max between the end of the previous competitive season and the start of the training. RESULTS. Age explained 6% of variance in YYIRT changes (P=0.172). The absolute pre-training values explained only 2% (P=0.410) of variance (6% if RTM is not taken into account). TRIMPs explained an additional 22% (P=0.030) and end-pre change scores accounted for a further 14% (P=0.021). The total adjusted R2 of the full model was 0.33. When the session-RPE was entered instead of TRIMPs, the additional explained variance was 44% (P=0.001). The end-pre change scores explained a further 6% (P=0.078). The adjusted R2 of the full model was 0.48. Similar results were found for VO2max (data not presented). DISCUSSION. The findings of this study indicate that TL accounted for a greater proportion of variance in pre-post changes of YYIRT and VO2max. The TL induced by technical-tactical sessions did not significantly influence the training outcomes (P>0.176). Furthermore, the additional proportion of variance explained by end-pre change scores suggests that the pre-experimental training level (and not the starting absolute value especially if corrected for the RTM) is a factor that may influence the individual training responses. These findings should be taken into consideration when designing and analyzing training studies for football players.
© Copyright 2009 14th annual Congress of the European College of Sport Science, Oslo/Norway, June 24-27, 2009, Book of Abstracts. Julkaistu Tekijä The Norwegian School of Sport Sciences. Kaikki oikeudet pidätetään. / Kääntänyt https://www.DeepL.com/Translator