On Sunday, August 11, 2019 at 2:09:28 PM UTC-4, Tom Kunich wrote:
I have some information on power development he

http://www.cyclingpowerlab.com/cyclingpoweroutput.aspx

Using this information it appears that I'm making over 200 watts on a good day.

I did 22 miles on Thursday with one long 6% climb and four over 5% humps in an hour and 25 minutes. After a short break I continued into a 25 knot wind right on the nose for another 6 miles and then a hard climb up to 1,000 feet with up to 11% recorded max grade (though these three hard grades measure 12% as often as hot.)

I got home with 42 miles and a riding time of 3 hrs and 22 minutes and a total of 2,404 feet of climbing for an average speed of 12.2 mph.

I was wondering if my average speed was including my breaks so I installed MapMyRide and checked my average speed that way. I got pretty near the same average speed. This leaves me wondering how the hell that people tell me that they do a flat ride that crosses about 20 stop lights and get an average speed of 17 mph.

The other responders are correct, it's the auto-pause function, and they aren't the same from device to device. It's likely both your computer and the Map My Ride app do not have the auto-pause turned on, hence the similar averages. I have an old Polar 720 strapped to my commuter, but always use strava on my phone as well. The polar autopause is quite quick - it stops accumulating within about two seconds of the stop, and one second upon restart. The strava app takes quite a bit longer to stop. As a result: The actual moving time between the two devices on my 20 mile commute can vary by as much as three minutes (I go through 16 stop lights), but the total time is well over 5 minutes, sometimes as high as ten depending on how gracious the traffic light gods are feeling that day.

NOTE: Another annoying "feature" of GPS tracking devices is that they can shorten the actual travel distance depending on the sampling rate of the device - in essence, the function sometimes doesn't take enough data points which results in the traveled distance being quite a bit shorter. I had an early Garmin device that had a fixed five-second sampling rate, which really sucked on technical single track.