Article Review: Considerations for Velocity-Based Training The instruction to move “as fast as possible” is less effective than a target velocity
Read time ~6 minutes
By Ian Kaplan
- This post reviews a recent paper in the journal of strength and conditioning that tested the effect of different instructions on bench press velocity in male powerlifters.
- It compared goal-directed numerical feedback (move the bar 1 m/s vs non-specific intent ie move as fast as possible) in a within-subject crossover design. both groups received knowledge of bar velocity
- The numerical feedback group’s maximum mean velocity per set was significantly higher, suggesting more effort, but there was no significant difference at the end of session repetition to failure test.
- Though the effect was very small, (.02 m/s average difference), it suggests a potentially meaningful effect of an objective velocity target in training design.
Background and study objective
With the advent of numerous commercial devices that track bar speed, significant research into the use of velocity information to inform training (velocity-based training) has been conducted. Once confined to exercise physiology labs and elite sports performance facilities, these measurement devices are now accessible and widely available to the public. However, it’s still unclear how best to use these tools.
On its face, velocity-based training seems to be useful in a variety of situations. Velocity can be used as an indicator of long term fatigue. If velocity across a movement index slows below a certain threshold relative to a baseline, that could set off some warning lights. A coach or athlete would react to velocity data by reducing training volume and/or intensity to recover, especially if velocity training is a priority. Intra-set drops in velocity can reliably indicate enough muscle fatigue and subsequent mechanical tension to stimulate hypertrophy across an enormous range of loading parameters (approximately 30-90% 1RM). (Schoenfeld et al., 2018). Velocity at submax weights may also be used to estimate 1RMs based on an inverse force/velocity relationship.
In this study, Velocity was used as an augmented feedback tool. Augmented feedback is additional information about a task that a performer would not otherwise have known. In this case, knowledge of results (KR) in the form of quantitative information about bar speed. Knowledge of performance (KP) information would have included feedback about the quality of the movement. In practice, KR and KP can be combined, but KR information has been shown to improve performance in high-velocity movements like jumping and sprinting. This study’s authors hypothesised that given KR information about bar speed, a specific velocity goal would yield improvements. The question then followed whether the increased performance would also create more fatigue. The authors hypothesized that higher bar speeds would translate to more fatigue.
This study used a within-subject crossover design, meaning the same subject did both the control and the test intervention. 15 male powerlifters with several years of competition experience were selected for the study and 13 completed it (that level of training experience is not common in a training study). Each participant completed 2 sessions separated by 3-7 days, the session order was randomized to eliminate an ordering effect.
Both sessions consisted of a warm up into a 1 rep max bench press, followed by 4 sets of 5 @45% 1RM for maximum velocity. Feedback about bar velocity was given after every repetition, the only difference was the instruction following the velocity information. The session ended with a 1 set to failure @75%. The authors made an effort to standardize as much of the procedure as reasonably possible.
The velocity-based portion for session A included the goal to move the bar 1.0 m/s. The equivalent portion for session B simply included the instruction to move it as fast as possible. Remember, the order of session A vs B was randomized between subjects.
The average velocity for the target velocity group was (0.84 , standard deviation= 0.10), whereas the control group average was (.82, SD=.09). This difference seems small in the absolute sense but, the statistical effect size is not small (ES=.29), and the p value=.0002 means that the likelihood that these results are due to chance is .02%.
The mean velocity of the fastest rep in each set also increased in both groups from sets 1 to the following sets. This effect was likely not dependent on which group they were in, it was just a function of practice and KR information. People improve speed with practice, especially with relevant feedback.
Interestingly, there were no differences in session repetition max performance between groups. The small increase in velocity might not have created enough latent fatigue to interfere with the RM set. This makes sense since fatigue, influenced by many complex feedback loops, is accelerated more so by sustained contractions. The contractions in the study were all brief, high velocity and low force contractions (Place et al., 2010). A post-activation potentiation (PAP) effect may confound a delayed test of fatigue since muscle force capability may increase following a period of stimulating contractile activity. In principle, small relative increases in power over a few reps should have virtually no impact on fatigue, and any effect may be offset by a PAP effect. Since the subjects in both groups improved performance across sets, fatigue was clearly a nonfactor across the training session. The RM test almost seems redundant.
As with many training studies, the sample size was very small, but the statistical power of the results suggests confidence in the results. The subjects’ high training age means that these results may not apply to populations with lower training ages. Objective KR information might be too distant or abstract for novices to work with if they don’t have not yet mastered the basic movement skill.
It’s also important to note that only one of the subjects in the study actually achieved the goal velocity of 1.0 m/s. And 2 subjects moved slower when presented with that objective goal. Selection of a more achievable, perhaps individualized, goal number may produce better results than the relatively unachievable goal presented in this case.
One significant limitation in the measurement procedure may have overestimated the effect size. This study measured bar velocity by taking only the fastest rep in each set and comparing the A and B sessions. If the average bar velocity for each rep across the entire set was used rather than just the fastest rep, the calculated effect may have been smaller or even nonsignificant. The number of reps that were actually faster in the A group vs the B group remains unknown since the complete data set is not available.
Translating research to practice
As is always the case, there are problems inherent in making predictions about practice based on only one study. It’s impossible to know to what extent a possible .02 m/s difference drives better training adaptations. Its abundantly clear that bar speed at 45% 1RM is not very specific to powerlifting competition. At first glance, velocity based goals are better suited for sports with low load, high-speed movements. Velocity-based training information may have a place in powerlifting as well, just at higher loads than 45% 1RM, and is certainly only part of the picture. Other, more specific, objective KR velocity-based research should be explored. The frequency of KR information should also be explored since it might not be necessary to provide feedback after every lift. The interactions of variables like feedback type, frequency, and measures of short term performance vs long term learning deserves more investigation.
Despite its limitations, this study offers some data to suggest that objective, challenging training targets may enhance performance. It’s easy to implement and doesn’t seem to negatively impact performance or fatigue. The risks involved in setting training targets are minimal if not nonexistent. If anything, it can add some much-needed competition, variety, and “gamification” to a potentially monotonous training program.
Whereas research methods are rigid, practical systems are fluid. If you don’t do so already, feel free to establish different objective targets in your training. They don’t have to be velocity based. They can be based on reps, load, or whatever objective criteria you can imagine. Use the tools you have available. Don’t be afraid to be creative and experiment on yourself. Let us know what you come up with. If you already use this type of strategy in your training, this paper should give you a little more confidence and maybe a few more ideas.
Hirsch, S. M., & Frost, D. M. (2019). Considerations for Velocity-Based Training: the instruction to move “as fast as possible” is less effective than a target velocity. Journal of Strength and Conditioning Research, (July), 1. https://doi.org/10.1519/jsc.0000000000003233
Place, N., Yamada, T., Bruton, J. D., & Westerblad, H. (2010). Muscle fatigue: From observations in humans to underlying mechanisms studied in intact single muscle fibres. European Journal of Applied Physiology, 110(1), 1–15. https://doi.org/10.1007/s00421-010-1480-0Schoenfeld, B. J., Grgic, J., Ogborn, D., & Krieger, J. W. (2017). Strength and Hypertrophy Adaptations Between Low- vs. High-Load Resistance Training. Journal of Strength and Conditioning Research, 31(12), 3508–3523. https://doi.org/10.1519/jsc.0000000000002200