The aim of this study was to compare the running demands of transition games (TGs) and official matches, analysing their requirements according to the performance of each position. An observational design was used to examine the activity of 20 soccer players during official matches and TGs. GPS technology was used to monitor the total distance covered (DC), distance at speeds between 14–17.9 km·h−1, 18–21 km·h−1, and above 21 km·h−1, peak speed, accelerations and decelerations above 2.5 m·s−2, and Player Load for both activities. All players were assigned to groups: centre-backs (CBs), fullbacks (FBs), defensive midfielders (DMFs), offensive midfielders (OMFs), wide midfielders (WMFs) and strikers (Ss). TGs showed greater total DC, DC 14–17.9 km·h−1, DC 18–21 km·h−1, DC >21 km·h−1, accelerations and decelerations >2.5 m·s−2, and Player Load (p < 0.01). CBs, FBs and Ss showed more DC, DC 14–17.9 km·h−1, DC 18–21 km·h−1, DC >21 km·h−1, accelerations and decelerations >2.5 m·s−2 and Player Load in TGs (p < 0.01). In the midfielder positions, transition game players showed greater DC 18–21 km·h−1, DC >21 km·h−1, accelerations and decelerations >2.5 m·s−2 than in matches (p < 0.05). DMFs showed higher total DC (p < 0.05) and WMFs greater DC and DC 14–17.9 km·h−1 (p < 0.01) in these drills. During transition games CBs showed greater DC 14–17.9 km·h−1 than FBs, and greater DC than Ss (p < 0.05). FBs performed more decelerations >2.5 m·s−2 than DMFs and OMFs (p < 0.05). TGs produced a homogenized load in soccer players, independent of their position, which exceeded the external load of official matches.