The purpose of this study was to determine the effects of model recordings on undergraduate conductors' tempo decisions for selected wind band excerpts. Participants (N = 42) studied three music score excerpts for 5 minutes each either (a) with an increased-tempo recording, (b) with a decreased-tempo recording, or (c) with a metronome set to the faster, printed tempo in the score, followed by a tapping task to demonstrate their preferred tempo. Participants returned 48 hours after their score study session for retention testing in the form of a second tapping task. Results indicated that for two of the three pieces, participants tapped faster in response to the metronome compared to both the decreased- and increased-tempo model recordings. Participants' tempo choices at the first testing session were most often not significantly different than the tempos they heard during study. However, we found that participants' preferred tempos did not vary as a function of study conditions during the retention testing session that occurred 48 hours later, where tempos trended toward moderation across all three musical selections. We speculate that pedagogical recommendations to avoid using model recordings may have limited empirical support, at least among novice conductors and at least in terms of possible long-term effects from short-term use.