NP vs Average outdoors

From a ride today, I am struggling to understand the relationship between NP and average power in performance analytics. Today I rode in the Lake District, my dig up Kirkstone Pass is in the image. My Normalized Power was 347w for 15 mins however my average was 186. Looking at the effort I only dropped below 200w for a moment. Am I reading things wrong?np

Are you sure those figures aren’t for the whole ride rather than the 15 mins you’ve highlighted?

@AndyGajda Performance Analytics displays numbers for the segment highlighted. The image shows a 15:57 duration at the top, so the numbers are for about 16 minutes. Try opening a ride in PA and selecting a segment and watching all the numbers change as you drag the endpoint.

@Rob1, I think you’re below 200w (gray line above HR curve) for about 1/8 km at the beginning and are close to or slightly below 200w at several points during the climb. You’re showing distance instead of time, so it’s hard to guess how much that affects AP. I’m estimating those valleys near 200w are close to the bottom of your Z3.

Over the effort, power is up and down from Z2 to well over Z6. NP weights power based on zone, so it makes sense the NP and AP are very different. Those Z5/Z6 spikes generate a lot of stress (indicated by NP, 1.26 IF, 26 TSS in 16 minutes ~= 100 TSS/hr) but are short enough they have only a small effect on AP.

Based on info in the TR articles explaining NP: and

Dragging this up after a week another ride today make me think this might be a cadence issue with a single side 4iiii

did ride today with a 45 minute climb, screen shot below. I spend 46.29 with NP of 308. I had two brief 20 second points where my power droped to 0 (letting a car through and opening a gate). The average power is coming out at 175 I’d have thought to be an average of that I’d need a significantly longer period of low power. Is average just a mathematical average


Taking a small section 1.30 my NP is 371, my power doesn’t dip below 275 and yet my average power shows as 159.


Hey Rob!

This is pretty peculiar, would you mind reaching out to our Support Team at They have the ability to look at the data at a deeper level to better understand what is going on and should be able to provide definitive answers for you.