Fitness apps and accuracy (again)

I know not to take fitness apps too seriously, but here we go again. Recently, I ran one of my regular routes on a new PR. Great, I might still be improving, despite age being against me. It was a chilly day (for Texas), about 52F (11C). 

My first observation is that the Coros watch I am using right now gave me a distance of 2.61 miles and an elevation gain of 627 ft. Strava gave the same distance, but only a 371 ft elevation gain. Previous runs using Garmin have measured the distance to 2.60 mile and 313 ft in elevation. My Fitbit have given me 2.80 miles on the same route. I have got used to the discrepancy in distance, and I use Strava as the arbiter.

Strava: 2.61 miles. 371 elev. gain

Coros: 2.61 miles, 627 ft elev. gain

Garmin: 2.60 miles, 313 ft elev. gain

Fitbit: 2.80 miles, elevation gain not recorded

This last time (Coros), I ran a PR with 35 seconds! This is a huge improvement, and I was deliberately trying for a new record, meaning that I pushed hard all the way to the end. 

What gets me is that Strava gave me an effort score of 81. A week earlier, I ran the same route 64 seconds slower, and the effort score was 99. Coros suggested that my record run training effect was 2.3 = maintaining. 

I don't agree, the record run was harder and of course improving, it was a huge PR. In addition, and I have been discussing this before in another post, short and hard is not given enough credit in Strava and the fitness apps. 

I also learned something important: 

Cool temperature lets your body work so much more efficiently. Of course, I knew heat is hard for the body, but I did not realize how much of a difference it made. I am not even in particularly good shape, and I made a huge improvement to my PR on a colder day.