GPS accuracy on time trials

I officially have one the other way where my $800 watch said the half course was short because it jumped around. Therefore strava didn't count it as a PR because I didn't "run" 13.1mi.

I wrote a letter to garmin asking for my money back but for some reason I haven't heard back. I'm sure the check is in the mail.

The question was about TT's...

I run all my TT's in one of a few open areas with minimal turns and 0 overhead obstructions. GPS with optimal signal is Very accurate. Def PR worthy data. Especially if you do a straight shot or out&back down a coastline/canal/railtrail. In contrast, if you are running though/near city, buildings, bridges, through a forest, ect... resulting in sub-par sig then Forgetaboutit.

The other weekend I raced a measured 10k on the wide open boardwalk (out and back), and my Garmin instinct measured like 5.9mi, routinely 15s/mi slower at each split.

At any given point I trust it as far as I can throw it, which is probably about 50 feet. In open areas it can be useful for key workouts, and in lieu of races it was nice to do tts to get a ballpark idea of fitness. But these days I mostly run by effort and am pleasantly surprised during races when pace isn't being sandbagged by gps drifting through corners, and much happier than I was when I constantly worried about the BS number it gave me

No, GPS alone doesn't count. I find on tracks, the GPS error goes anywhere from 3.17-3.22 miles. That is a on average...a 3.1% error (heh). Which means you probably ran closer to 3 miles.

Why didn’t you just run around a track

I mean, you can claim whatever you want for a PR, but GPS is not PR worthy. It's at best 1% off. If someone told me their PR was based on GPS, they'd get a pretty big eye roll.

Why would a simple machine with no engine have a gauge that measures the RPMs of an engine?

People often times think GPS or Google Earth, aerial photos, etc are accurate. THEY ARE NOT. I work in mapping and surveying information and I can tell you that those types of information are great for many purposes but not for the kind of accuracy you are looking for. They are wonderful for planning routes and getting rough distances. To know how far you ran you will have to use a calibrated bike or an accurate measuring wheel.

On average how much would you say GPS is out by? Just a ballpark figure. I'm just really curious.

There have been a few days this year where my Garmin has been super wonky, but for the most part it's extremely accurate, generally within a few feet or less (for a mile). I've tested it on the track, on a certified measured course with mile markers, and against my Styrd data. It's perfectly accurate for workout purposes, but not for legitimate PR purposes.

I have used a high quality gps attached to my bicycle's handlebars when measuring course. I set it to metric so I could get readings to the 1/100 of a km (10 m). It's never less then the measured distance (using a calibrated Jones Counter). Sometimes right on. But nearly all the time, the gps device will give me a reading slightly long 10m+ per mile.

I also always carefully map out my routes on Google Earth before measuring. And depending on the available view, I could often get an accuracy within 10 m per mile. FYI - There is an option on Google Earth to use "historical imagery". This may help to see through tree cover for those areas where there is a change in seasons.