Training Insights, Publications & Articles

Why Measuring Power Led Me Astray

by Matt Jordan on March 8, 2013 No comments

Article Overview: 1355 words on assessing strength abilities and mechanical power. I focus on the four reasons that early in my career as a young strength coach, quantifying the impact of my programs turned out to be tougher than I thought it would be.

In this article, there are four main lessons for the young strength coach: (1) physiological factors like maximal muscle power and the output we measure from various movements like vertical jump mechanical power aren’t the same thing; (2) mechanical output in the vertical jump can remain depressed even after unloading periods due to the persistence of peripheral fatigue factors – thus, vertical jumping and the various performance variables may be great markers for neuromuscular readiness; (3) vertical jumping is a great test for assessing explosive strength abilities – but it is still a movement that may or may not be related to an athlete’s sport-specific requirements; (4) shoot for gold standard assessments because measurement error and fundamental mistakes in calculating variables of interest can lead you astray.

Early in my career as a strength coach, I learned about the many different strength qualities. I prefer to call them strength abilities though. Qualities are not quantifiable. Strength abilities are quantifiable.

I also learned about the importance of quantifying strength abilities to quantify the impact of my programs. However, I struggled to find consensus on the topic of assessing strength abilities.

First, let’s take a brief dive into some of the relevant strength abilities. These include maximum strength (maximum force produced irrespective of time), explosive strength (the rapid application of force per unit of time or rate of force development), and maximal muscle power (the muscle work rate).

These strength abilities are well-defined in the scientific literature and appear in exercise physiology textbooks. Early research by A.V. Hill in the 1930’s depicted the hyperbolic muscle force velocity relationship, and the parabolic power velocity relationship.

Both maximum strength and maximal muscle power are often discussed in the context of the muscle force velocity relationship. Here, the maximal force producing ability of a single muscle fibre or whole muscle can be quantified across a range of shortening velocities.

It is somewhat natural and logical to extend the work of A.V. Hill to the whole body level. This led to many different methods for evaluating the whole body force velocity or power velocity relationship.

I should note that when we evaluate whole body power, we are now referring to the mechanical power generated by the system, which results from the many muscles working together in a synergistic and coordinated manner. Of course this would be highly related to the maximal muscle power ability.

Now back to the story. As a young strength coach it did not take me long to see the positive association between powerful athletes and sport performance. I use the word power here as a qualitative term. This word is also used interchangeably with explosive. The use of the terms powerful and explosive are often bemoaned by purists in biomechanics but I think you understand what I mean.

The controversy or challenge arose when it came to quantifying mechanical power. I had a mentor who told me not to waste my time trying to evaluate “power”. He encouraged me to focus on assessing muscle mass/body composition and maximal strength. In his opinion, both of these factors were highly trainable, possibly more influential for developing maximal muscle power, and he had never had an experience where assessing “power” enhanced his programming.

On the other hand, I read many scientific papers through the 90’s that encouraged strength coaches to evaluate maximal mechanical power in the vertical jump

My first run at assessing mechanical power led me to purchase a jump mat (contact mat). I found an equation to convert jump height to “power” and I began testing athletes. I then moved onto a position transducer that used the bar velocity and the external load of the system to yield a “power” value.

The long story short is that both methods for evaluating “power” did not seem overly sensitive to the training process. In fact, I would give my athletes a couple of days rest at the end of a training cycle, measure “power” with one of the two methods above, and I would often find that “power” had decreased from the baseline test.

The first problem was that I had underestimated the persistence of peripheral fatigue or low frequency force depression. A couple of days rest was insufficient for a full recovery in the higher end neuromuscular abilities like maximal muscle power and explosive strength.

The second problem was my equipment and my measurement techniques. The linear position transducer was glitchy. My uncertainty was further intensified when I attended the International Conference on Strength Training in Colorado Springs, and listened to a presenter who casually pointed out that many scientific papers had failed to account for the barbell mass and the system mass when calculating this elusive “power” value with the position transducer method.

At this point I was relatively convinced that I was chasing my own tail when it came to measuring “power”. You will also notice that I continue to put “power” in scare quotes as I don’t think any of the above mentioned methods were measuring “power” per se.

At this point, I am now 9 years into my S&C career and I was not about to give up on assessing mechanical power. I upgraded my equipment and purchased a force plate. There was some additional math involved to calculate power. After solving for acceleration from the equation F=ma, time integration of the acceleration vs. time curve yields a velocity vs. time curve. Power is the work rate, and work equals force x distance. So, power equals force x distance divided by time. This can be rearranged as force x distance/time or force x velocity.

So, I finally had my solution or so I thought. I had gold standard equipment, a scientifically supported method for calculating mechanical power, and while there were still the biomechanical purists criticizing this approach, I felt far more confident with my methodology.

But I still struggled to find utility for this metric “power”. Maybe it is because “power” is just a correlate to jump performance, and jump performance is best quantified by calculating an athlete’s takeoff velocity using Newtonian mechanics.

The other issue is that I might have been confusing the importance of maximal muscle power and mechanical power assessed in the vertical jump. While maximal muscle power can be essential for sport performance, the vertical jump movement from which we are obtaining mechanical power may have nothing to do with sport performance.

I went on to measure mechanical power in various forms of jumping from 100’s of athletes in many different sports.  Here’s a breakdown of the relative peak power (W/kg) by sport and sex.


The gist of things is obvious:  the athletes in the more explosive sports generate more mechanical power in jumping.  If I broke the plot down by performance level you would see that power is also related to the level of performance with more elite athletes tending to be more powerful than development level athletes.

As you look at this, let me remind you of the difference between correlation and causality.  A high degree of correlation means that two variables are related, and I have found that peak power correlates extremely well to performance in a variety of sports.

However, correlation does not imply that variable ‘x’, causes variable ‘y’. This might be the main problem with assessing mechanical power in the vertical jump. While correlated to sport performance and jump performance, mechanical power in the vertical jump isn’t the same as maximal muscle power and often fails to discriminate top performers.

Here are a few more thoughts and considerations on the limitations of measuring mechanical power and why it led me astray:

  1. I will say it again – jump performance is heavily affected by fatigue and training stress so it can remain depressed for the majority of a training cycle and even after a few days of rest (this makes it interesting as a parameter to evaluate neuromuscular readiness or for building reaction curves to training).
  2. Peak power and more importantly vertical jump ability often do not improve at the same rate as competitive performance in many sports (i.e. you get a whole lot better in your sport than you do in a test of peak power in a jumping movement).

And finally,

3. Lots of athletes don’t participate in sports that require an optimization of peak muscular power

To this point, peak power in a jump is hit somewhere around 400 ms (at a minimum) and it is more likely a value of 500-700 ms after the onset of the jump. This is an eternity compared to many sport movements.

So, if the sport in question involves speed or acceleration (which most sports do), where contact times for the foot against the ground are 100-300 ms, peak power in a countermovement jump is not overly relevant.

That is my opinion, and this is why measuring peak power led me astray initially in my career. But like most things, it’s rarely good vs. bad and more how a parameter or assessment is used. The key is to determine this in the iterative process of determining what matters, measuring what matters and changing what matters as it relates to sport performance.

Matt JordanWhy Measuring Power Led Me Astray

Join the conversation