Saturday, April 30, 2011

Normalized Power vs. Average Power Discrepancy

When I got back from my ride today, I downloaded my ride data from my Garmin Edge 500 into Garmin Connect (online) software.  I was "pleasantly" surprised to see my Average Power for my 2 hr. + ride was 250w.  (See Garmin Connect Data on left).  That's a LOT higher than what I normally see on a 2-3 hr. ride...especially since today's ride was a regular route of mine. 

Then, after uploading my Garmin Connect data into Training Peaks WKO+, I discovered that my Normalized Power was 220w and my Average Power was only 170w.  (See WKO+ graph on left).  At first, I didn't know what to make of it.  Why should there be a 30w difference between my Normalized Power in WKO+ and Average Power on my Garmin Edge computer?  I think the following explanation will answer why.

When you ride outside, there are many factors that affect your ride: wind, hills, accelerations, drafting, long steady grinding, coasting, etc.  Because of this variability, Average Power is not a true indicator of your true metabolic demands of your ride.  What complicates things even more is that Average Power can include zero power (not pedaling/coasting) or no zero power.  Naturally, if Average Power includes zeros you're going to see low numbers like my 170w power reading.  If Average Power does NOT include zeros you're going to see high numbers like my 250w power reading.  BTW, I programmed my Garmin Edge NOT to include zeros when computing Average Power...which is why the high number. 

To account for the variableness of a ride, Training Peaks WKO+ software includes an algorithm which provides a better measure of the true metablolic demands of the ride called "Normalized Power".  Normalized Power provides a better measure of the true physiological demands of the ride.

I'll tell you why Average Power (w/ no zeros) computed by my Garmn Edge 500 is NOT a good metric to use.  When I'm riding and not pedaling, I'm essentially recovering. Therefore, when I do pedal again..I'm relatively refreshed and able to put the pedal to the metal and keep the power up.  Besides, with an FTP of 260w (my current hour power) there is no way that I'd be able to average 250w for 2+ hrs. on my ride today (like my Garmin Edge 500 said I did).  Using Average Power (w/ zeros) computed by my Garmin Edge 500 may be a better metric except for the fact that it's demoralizing to ride hard for 2+ hrs. and see that you only averaged a paltry 170w..a power number I usually warmup with on my trainer.  BTW, Normalized Power will equal your Average Power on your trainer because you are constantly pedaling.  If you Garmin owners don't believe me, check it out yourself.  Upload your Edge 500 data into Garmin Connect then Upload it into Training Peaks WKO+ and you'll see the same number for Average Power and Normalized Power.

So, stick with Normalized Power as a metric for analyzing power's a more realistic power number for the work/effort you performed on your ride.

Power ON! Coach Rob


Scott said...

"BTW, Normalized Power will equal your Average Power on your trainer because you are constantly pedaling."

This statement is false, your normalized power tends to be the same on a trainer because you don't have large surging efforts.. pull out the trainer for 10 minutes and alternate 100 for 1:30 400 for :15-:30, your AP and NP will not be the same

Wow Thinking said...

Good post, Just in case if you need to contact Garmin's Customer Product Support, Garmin phone number Call at Garmin help USA 1-916-209-6500 and get solve yourquery.