Power training: Current vs. average power display

Many of my teammates and I mount power meters on our bikes that display our pedaling power during the "intervals" we do during training rides. This is much more useful than using our heart-rate to try to pace ourselves. There are lots of subtle details, though, that affect how we ride during power-based intervals, like what data to display on the meter's LCD.

So, if we're riding, say, a 20-minute interval, we press the "interval mode" start button and then constantly adjust our pedaling effort so that the interval's resulting average power reading (in Watts) is within our 20-minute training power zone. If the reading is too low we pedal harder; if too high we back off a bit. We then dutifully record this reading in our training logs for future reference. Just for an example, here are my training zones (we rarely train below Level 4 unless we're too dog-tired to do any better):
  • Level 4: 246-285W
  • Level 5: 286-332W
  • Level 6: 333-600W
  • Level 7: as high as possible!
It's very hard to pedal 100% smoothly. So most of my teammates who train with power meters set them to display the average power of the interval they're riding. That smoothes out the natural peaks and valleys in the power we produce so the value displayed doesn't swing up and down as wildly. And since we can only record the average power in our logs (not current, of course), displaying the average power seems to make sense. The down-side is that in long intervals the inevitable power variances, below and above our training zones, don't really show on the average display. I've noticed most guys who use average display tend to have huge power spikes on short hills, and very low power on descents. Not a problem on a steady climb (like, say, Bonny Doon Road) but definitely less ideal on a route with rolling hills (like Swanton Road or Highway 1).

Others of us use the current ("instantaneous") power display. It shows the power we're producing at any given moment. I do this to learn how to pace myself as smoothly as possible. The experts agree, as do my legs, that smooth and steady is more efficient. And more efficient is faster and more sustainable over time. Per Matt Russ: "A variable effort is more fatiguing to the body when compared to an even, steady effort." I am pretty much always able to hit my power training zones on every interval, day after day, week after week. I suspect my even pacing gets some credit for that.

So, which is best? Neither; both can work just as well. Try both and see how they work for you...

OK, I admit I prefer current power since our intervals are designed for a specific power level, not the average of wildly fluctuating power. And constant power spikes tire us out; that can jeopardize our power during a tough workout. Check out the following graphs, which charted my power during two similar intervals (the yellow line is power in Watts):



Which one do you think yielded the best average power? The left one did, which may not be surprising if you look at how smooth the yellow power line is on it, compared with the jumpy yellow power line on the right graph. I hope those graphs make my point for me: Use the current power display and try to keep the power exactly where you want the average to be when you're done with the interval.

But it brings up another topic: pacing strategies for time-trials (solo races against the clock). The steady pace that is best for intervals isn't always best for TTs (or other races). But even then I'm convinced current power display is better because it allows us to nail our power level exactly to suit the TT course. But... that is a big topic worthy of a separate blog post. In the meantime, here are some links to pacing articles:
Happy pacing!

Comments