Pantelis Theodoros Nikolaidis, Beat Knechtle

Pacing strategies in marathon runners have previously been examined, especially with regard to age and performance level separately. However, less information about the age × performance interaction on pacing in age-group runners exists. The aim of the present study was to examine whether runners with similar race time and at different age differ for pacing. Data (women, n=117,595; men, n=180,487) from the "New York City Marathon" between 2006 and 2016 were analyzed. A between-within subjects analysis of variance showed a large main effect of split on race speed (p<0.001, η(2)=0.538) with the fastest speed in the 5-10 km split and the slowest in the 35-40 km. A small sex × split interaction on race speed was found (p<0.001, η(2)=0.035) with men showing larger increase in speed at 5 km and women at 25 km and 40 km (end spurt). An age-group × performance group interaction on Δspeed was shown for both sexes at 5 km, 10 km, 15 km, 20 km, 25 km, 30 km, 35 km, and 40 km (p<0.001, 0.001≤η(2)≤0.004), where athletes in older age-groups presented a relatively more even pace compared with athletes in younger age-groups, a trend that was more remarkable in the relatively slow performance groups. So far, the present study is the first one to observe an age × performance interaction on pacing; ie, older runners pace differently (smaller changes) than younger runners with similar race time. These findings are of great practical interest for coaches working with marathon runners of different age, but similar race time.