MV, MPV or PV: which speed metric to use in VBT

Updated on 26 de June de 2026 by Adrián Escobar Morales

When you start using a sports encoder, one of the first questions that appears is this: what speed metric should I use? Most devices offer three options — mean velocity (MV), mean propulsive velocity (MPV), and peak velocity (PV) — and it is not always clear which is most appropriate for each objective.

The choice is not trivial. Depending on the metric you use, the estimate of %1RM may vary, the reliability of your measurements changes, and the conclusions you draw about the athlete’s condition may be different. A study by García-Ramos et al. published in the Journal of Strength and Conditioning Research in 2018 systematically compared the three metrics in the bench press and offers concrete answers.

What each metric measures: precise definitions

Before comparing, it is important to be clear about what exactly each variable calculates:

• Average speed (MV): average speed from the beginning of the concentric phase until the bar reaches maximum height. It includes the entire concentric phase, both the propulsive phase and the final braking phase.
• Average propulsive speed (MPV): average velocity during the propulsive phase only — from the beginning of the concentric phase to the point where the acceleration of the rod falls below gravity (-9.81 m/s²). Excludes the final deceleration phase.
• Peak Velocity (PV): the maximum instantaneous value of velocity reached during the entire concentric phase.

The difference between MV and MPV is especially relevant for light loads. At 40% of 1RM, the end-of-stroke braking phase can represent a significant portion of the movement — including it makes MV lower than MPV. As the load increases and the braking phase is reduced, MV and MPV converge and from 70-75% of 1RM they practically coincide.

The study: what was compared and how

García-Ramos et al. evaluated 30 men with experience in strength training performing two variants of the bench press throw: concentric only and eccentric-concentric. In each session they measured the three speed metrics in loads from 20% to 100% of 1RM, comparing three key aspects:

• The linearity of the load-speed relationship for each metric.
• The precision of the general equations to predict %1RM from speed.
• The reliability between sessions of the speed obtained at each %1RM.

Finding 1: MV produces the most linear load-velocity relationship

All three metrics showed strong, linear load-velocity relationships — all with R² above 0.95. But the VM was the one that produced the greatest linearity in both variants of the exercise:

• MV: Median R² = 0.989 (concentric only) and 0.993 (eccentric-concentric)
• MPV: Median R² = 0.983 and 0.980
• PV: Median R² = 0.974 and 0.969

Greater linearity means that the relationship between load and speed is more predictable and consistent across the entire spectrum of loads — from 20% to 100% of 1RM. This makes it easier to estimate 1RM from submaximal loads.

Finding 2: VM predicts %1RM more accurately

The accuracy of the general equations for predicting %1RM — measured as standard error of the estimate (SEE) — was better for MV in both variants:

In practical terms: if the actual 1RM is 100 kg, the MV estimates the %1RM with an average error of about 4-5 kg, while the MPV and PV have errors of 5-6 kg. It’s not a huge difference, but in contexts where precision matters — load planning in performance athletes, tracking over the course of a season — that difference adds up.

Finding 3: PV is the most reliable between sessions, but with important nuances

This is the most counterintuitive finding of the study. PV showed the lowest coefficient of variation between sessions:

• PV: CV = 3.50-3.87%
• MV: CV = 4.05-4.93%
• MPV: CV = 5.11-6.03%

This means that if you measure the peak velocity of the same athlete in the same exercise and the same load in two different sessions, the value varies less with PV than with MV or MPV. Apparently, PV would be the most consistent metric.

However, the authors point out a fundamental technical nuance: the lower CV of the PV is partly explained by its higher absolute values. The coefficient of variation is calculated by dividing the measurement error by the average value — and since PV is always the highest value of the three metrics, the ratio is smaller even if the absolute error is similar or greater. When the authors analyze the standard error of measurement instead of the CV, the advantage of PV disappears.

Furthermore, PV had the worst linearity and lowest precision of the overall equation. A very reliable metric between sessions but with less predictive capacity for %1RM has limited usefulness for the main objective of VBT.

Why was MPV historically the most recommended?

González-Badillo and Sánchez-Medina, the historical references of VBT in Spanish, recommended for years the use of the MPV, arguing that it better reflects the real neuromuscular potential of the athlete — especially in light loads, where the braking phase at the end of the stroke “contaminates” the MV with a deceleration that does not reflect the ability to produce force.

The study by García-Ramos et al. qualifies that recommendation: although the theoretical argument makes sense, in practice MV produces a more linear load-velocity relationship and more accurate %1RM prediction equations. The difference is not dramatic, but it is consistent across all comparisons in the study.

The authors add another practical argument: MV is easier to calculate with most devices — linear encoders, mobile apps, accelerometers — because it does not require detecting the exact point where the acceleration falls below gravity, which is a more complex calculation and subject to technical errors in some systems.

Summary table: when to use each metric

What this means for using the ADR Encoder

He ADR Encoder records all three velocity metrics in real time during the concentric phase — average speed (MV), average propulsive speed (MPV) and maximum speed (PV) — and shows them simultaneously in the app ADR System for each repetition. This gives you the flexibility to use the most appropriate metric according to the objective of each session or exercise, without having to choose beforehand.

Recommended product

ADR Encoder

Speed, strength and power in real time · Scientifically validated · Free app · €249

See product →

In practice, for estimating daily 1RM and monitoring intra-set velocity loss, MV is the primary metric — supported by the evidence from this study. To evaluate explosive performance or power in ballistic exercises, PV is the natural complement. And for coaches working with protocols and references from the classic González-Badillo literature, the MPV is still available to maintain consistency with that data.

Having all three metrics available on the same device is no small detail — it allows you to compare your data to any published study or reference, regardless of which metric each author used.

Conclusion

All three velocity metrics — MV, MPV, and PV — are valid and produce strong, linear load-velocity relationships. But they are not equivalent. The study by García-Ramos et al. (2018) shows that the average velocity (MV) is the most appropriate to estimate %1RM with greater linearity, greater precision and greater reliability between sessions than the MPV. The PV has better apparent reliability between sessions but lower predictive capacity, which limits its usefulness for the main objective of VBT.

The next time your encoder shows you three different metrics for the same repetition, you know which one to prioritize — and why.

Frequently asked questions

What is mean propulsive speed (MPV) in VBT?

The MPV is the average velocity during the propulsive phase of the movement — from the beginning of the concentric phase until the acceleration of the bar drops below gravity (-9.81 m/s²). It excludes the final deceleration phase and was for years the most recommended metric in VBT, although recent evidence indicates that the average velocity (MV) is more accurate for estimating 1RM.

What is the difference between MV and MPV in an encoder?

The MV includes the entire concentric phase until the bar reaches the maximum height, including the final braking phase. The MPV only includes the part where the bar is accelerating. In light loads the difference is notable; in heavy loads (above 70-75% of 1RM) both converge and practically coincide.

What speed metrics does the ADR Encoder offer?

The ADR Encoder and the ADR System app display all three metrics simultaneously for each repetition: average velocity (MV), average propulsive velocity (MPV), and maximum or peak velocity (PV), all calculated during the concentric phase. For the estimation of 1RM and the control of velocity loss, MV is the main metric recommended by scientific evidence. MPV and PV are available for those working with specific protocols or wanting to evaluate explosive performance.

Is it better to use MPV or MV for VBT?

To estimate %1RM and daily 1RM, the evidence from the study by García-Ramos et al. (2018) indicates that MV is more accurate—it produces a more linear load-velocity relationship and prediction equations with lower error. The MPV can be useful to evaluate performance at light loads where the braking phase is relevant.

Literature

1. García-Ramos, A., Pestaña-Melero, F.L., Pérez-Castilla, A., Rojas, F.J. & Haff, G.G. (2018). Mean velocity vs. mean propulsive velocity vs. peak velocity: which variable determines bench press relative load with higher reliability? Journal of Strength and Conditioning Research, 32(5), 1273–1279. See study →
2. González-Badillo, J.J. & Sánchez-Medina, L. (2010). Movement velocity as a measure of loading intensity in resistance training. International Journal of Sports Medicine, 31(5), 347–352. See in PubMed →
3. Sánchez-Medina, L., Pérez, C.E. & González-Badillo, J.J. (2010). Importance of the propulsive phase in strength assessment. International Journal of Sports Medicine, 31(2), 123–129. See in PubMed →

Adrián Escobar Morales