The aims of this study were to determine the effect of different compositions in
transition games (TGs) on the load of soccer players and to evaluate their
performance in physical tests. Using a GPS system, 18 players were monitored
during: 3vs2, 2vs1 and 1vs1. Distance covered (DC), DC
18–20.9 km·h-1, 21–23.9 km·h-1,>24 km·h-1,
peak speed, accelerations (Acc) and decelerations (Dec)>1.0 m·s-2
and>2.5 m·s-2 and rate of perceived exertion (RPE) were
recorded. Before and after each TG, countermovement-jump (CMJ), 15-
(S15) and 30 m (S30) speed tests were assessed.
TG3vs2 showed greater DC and Dec>1.0 m·s-2 than
TG2vs1, and DC, DC 18.0–23.9 km·h-1,
Acc>1.0 m·s-2 and Dec>2.5 m·s-2 than
TG1vs1 (p<0.01). TG2vs1 achieved higher DC,
DC 18.0–23.9 km·h-1, and Acc>2.5 m·s-2
(p<0.01) but lower peak speed (p=0.02) and RPE (p=0.02)
than TG1vs1. Post-intervention, TG1vs1 showed lower CMJ
and higher S15 (p=0.02), while TG3vs2, showed
improvements in CMJ (p<0.01). The three tasks showed large variations
for DC>24 km·h-1, Acc>1.0 m·s-2,
Dec>1.0 m·s-2 and Dec>2.5 m·s-2. The load of TGs
is sensitive to their player composition.