To determine if generic off-ice physical fitness tests can provide useful predictions of ice hockey players' match performance. Approximately 40 to 60 defenders and 70 to 100 forwards from the Norwegian male upper ice hockey league were tested for strength (1-repetition maximum in squat and bench press), power (40-m sprint and countermovement jump), and endurance (hanging sit-ups, chins, and 3000-m run) annually at the end of every preseason period between 2008 and 2017. Measures of match performance were each player's season mean counts per match of assists, points, goals, penalty minutes, and plus-minus score. Overall, match performance measures displayed trivial to small correlations with the fitness tests. More specifically, points per game had at most small correlations with measures of strength (range, approximately -0.2 to 0.3), speed (approximately -0.2 to 0.3), and endurance (approximately -0.1 to 0.3). After adjustments for age that showed moderate to large correlations with player match performance, multiple-regression analyses of each test measure still provided some predictability among players of the same age. However, players selected for the national team had substantially better mean scores for most tests and match performance measures than those not selected, with a moderate to large difference for age, 1-repetition maximum squat, and 1-repetition maximum bench press. Fitness tests had only marginal utility for predicting match performance in Norwegian hockey players, but those selected into the national team had better general fitness.