Peak Power or Mean Power

What parameter should I measure and why?

There is no doubt that peak power is a very popular measure in much of the published literature relating to power measurement in athletes. But why is this?

Power is a measure of the rate of doing work as per these equations –

Peak power is the highest rate of doing work over a sample period tsample

Mean power is the average rate of doing work over the concentric lift phase tconc

There is no doubt that peak power is a very popular measure in much of the published literature relating to power measurement in athletes. But why is this? It is certainly a much simpler measurement to take, just read the highest value on the power curve. In fact, this is often the reason quoted in research methodologies(1), it is simpler and quicker than manually marking up the concentric lift phase of each individual rep - but is it the best measure to reflect the performance of the jump?

Often too, only a single rep is studied for the same reason, this is too time-consuming.

This means that the performance of the athlete being studies is being boiled down to a single sample of a single effort - simply because, due to limitations in the measurement technology used, this is the easiest measure to make! The literature actually suggests that mean values of multiple reps are a better measure(2).

GymAware solves the problem of difficult markup by automating it as shown below.

Looking at the graph above, the grey bands show where GymAware has detected the concentric lift phase for two Olympic lifts - Power cleans (above) and two concentric jumps (previous page). The large power spikes in the power clean are caused by the rapid deceleration as the bar hits the floor.

Note: Only the actual effort is detected, not just the min and max displacement - This is GymAware’s unique rep detection algorithm at work. The catchphrase is not included in the concentric measure.

So now researchers have a tool that enables them to study mean power for multiple reps with no additional effort.

References:
(1) Comparison of Four Different Methods to Measure Power Output During the Hang Power Clean and the Weighted Jump Squat. Naruhiro Hori, Robert U. Newton, Warren A. Andrews, Naoki Kawamori, Michael R. McGuigan Journal of strength and conditioning research: the research journal of the NSCA

(2) Comparison of Different Methods of Determining Power Output in Weightlifting Exercises Hori, Naruhiro MS, CSCS, NSCA-CPT; Newton, Robert U. PhD, CSCS; Nosaka, Kazunori PhD; McGuigan, Michael R. PhD Strength & Conditioning Journal.

NOTE:

With mean power we are measuring the power output throughout the complete concentric phase. From the start to the finish.

 

It's typically a good metric to assess performance, If it reads an unusually low value the algorithm likely included bar movements that weren't part of the concentric lift phase.

 

Peak power in our opinion is less repeatable. It's only measured during a single time period within the concentric phase. So it's an instantaneous point in time measurement that can occur anywhere within the lift. For this reason it tends to be a more variable metric to track.