One rationale for using simulation when learning real life complex skills (e.g., surgery, fire fighting) is that in the simulated environment the learner has the opportunity to generate errors without the risks associated with real life performance. However, within the motor skill learning literature there have been conflicting viewpoints on the role of errors in practice (see Lee, Eliasz, Gonzalez, Alguire, Ding & Dhaliwal, 2015 for review). Sport provides a unique opportunity to address the role of errors during simulated games (practice) and as the high stakes game-play. The relationship between errors made during practice and game-play was of interest in the present study. Five types of errors (serve, serve-receive, attack, block, and defense) were recorded for simulated (in practice) and actual game-play for fifteen varsity volleyball players. Segments of time where the ball was in play, time played for each individual, and instances of errors were identified and organized using Studiocode (Newark, DE; Digital Tec Solutions). Results revealed that players with high error scores during practice were more likely to produce high errors scores in game-play ρ=.80, p<.001., even when player skill, player position, and the sex of the player were controlled for, (partial correlation ρ =.78, p= .003). Additionally, the total number of errors produced did not differ significantly between practice and games (T= 25, p=.79). These results provide information about errors in ecologically valid, complex, skills and situations as well as for skilled performers.