Objective: A limited number of studies have explored the external load experienced in indoor sports such as ice hockey, and few the link between training and match performance. As a paucity exists within this topic, this study explored whether a simulated match design (i.e., scrimmage) could be representative of official match demands and elicit similar external loads as in official matches in a group of elite youth male ice hockey players.

Methods: A total of 26 players were monitored during eight official and four simulation matches using a Local Positioning System. Total distance, max velocity, slow (0–10.9 km/h), moderate (11–16.9 km/h), high (17.0–23.9 km/h), and sprint (>24 km/h) speed skating distance, distance per min, PlayerLoad^TM, PlayerLoad^TM per min, high-intensity events (HIEs) (>2.5 m/s⁻²), acceleration (ACCs), decelerations (DECs), and change of directions (CODs) were extracted from the tracking devices. A two-level regression analysis was conducted to compare the difference between match types when controlling for time on ice, match day, and position.

Results: Between match-type results showed a credible difference in all variables except max velocity and ACCs. Distance per min was 27.3% higher during simulation matches and was explained by a 21.3, 24.1, and 14.8% higher distance in sprint-, high-, and moderate speed skating distance, while slow speed-skating distance was 49.2% lower and total distance only trivially different from official to simulation matches. Total PlayerLoad^TM was 11.2% lower, while PlayerLoad^TM per min was 8.5% higher during simulation matches. HIEs, CODs, and DECs were 10.0, 11.9, and 22.3% higher during simulation matches.

Conclusion: The simulated match design is related to official match demands with comparable match-time, playing time, number of shifts, and shift duration. However, simulation matches provoked a higher external load output compared with official matches, possibly explained by a more continuous movement design. A game-based simulation match design can therefore be utilized when match-related actions at high intensity are warranted. © 2022 Byrkjedal, Luteberge, Bjørnsen, Ivarsson and Spencer.