Training load (TL) based on the subjective rating of perceived exertion (RPE) may be a useful athlete monitoring alternative when wearable technology is unavailable. The aim of this study was to examine the validity of RPE-based TL monitoring in elite ice hockey. A male ice hockey team (n = 18) was monitored using a 200-Hz accelerometer, heart rate (HR) and RPE (0-10 scale), throughout a 4-week competitive period (n = 309 individual observations). Session-RPE (RPE × duration) averaged 244.8 ± 135.2 and 728.6 ± 150.9 arbitrary units (AU) during practice sessions and during official games, respectively. The smallest worthwhile change was 19.8 AU. Within-individual correlations between session-RPE and total accelerations >0.5 m·s-2 (Acctot), accelerations >2 m·s-2 (Acc2), total decelerations >-0.5 m·s-2 (Dectot), decelerations < -2 m·s-2 (Dec2), time > 85% maximum HR, Edwards' TL, and modified training impulse were very large (r = 0.70-0.89; p < 0.001). In addition, correlations between RPE and measures of exercise intensity (Acctot per min, Acc2 per min, Dectot per min, mean HR, and peak HR) were small (r = 0.02-0.29; p < 0.05) except for Dec2 being unclear (p = 0.686). Differences in intensity parameters between RPE range (easy to very hard, 2-7 AU) were small (r = 0.22-0.31; p < 0.05). The session-RPE method can be used as a global indicator of TL in ice hockey. Specific ranges of time-motion and HR intensity variables can be demarcated between RPE categories (easy to very hard; 2-7 AU). Accounting for training volume (session-RPE) more accurately reflects objective methods of TL based on accelerative efforts and HR, than the RPE score (based on the perception of the intensity).
© Copyright 2022 The Journal of Strength and Conditioning Research. National Strength & Conditioning Association. Alle Rechte vorbehalten.