Pace is not showing correct while running

There's been a lot of discussion about instantaneous pace on here, especially regarding GPS pace. There are a number of solutions, I beleive Anders has a great Pace datafield - or you can use a smoothed pace datafield that averages it out over a few seconds. Its largely due to the nature of GPS positioning that instantaneous pace can be weird

The Art of Pacing with a Garmin | DC Rainmaker

An older article but worth a nose

I would love it if we had a sticky at the top of the forum for:

1. Pacing
2. GPS accuracy
3. Battery woes.

4. OHR

Ɛpsilon said:

Its largely due to the nature of GPS positioning that instantaneous pace can be weird

No it is not. It is largely due to Garmin's faulty algorithm. The pace reported by the watch is not a pure GPS instant pace. That would be way way worse than what the watch is reporting. I welcome you to look at any FIT file produced by the watch and see how much it is jumping. What you need to look at is the GPS distance in every sample and how it changes every second - that is your true GPS speed. So Garmin is rightfully already post-processing the pace and averaging it over multiple seconds.

But, unless GPS signal is very strong, Garmin's pace seems to be based primarily on accelerometer data. There seems to be some sort of continuing calibration of accelerometer based on the GPS signal, and then the accelerometer input seems to be used as the main input for the pace. However this algorithm doesn't seem to be working correctly, and that is what causes pace dropouts - always on the slow side.

I should add that the reason the pace is correct after the run is that the run statistics show the average pace which is basically the total time in minutes divided by the the total distance. That has nothing to do with instant pace displayed on the watch during run. Mile or kilometer lap pace is also correct, because it is also based on time and distance. 

Very well explained! But sadly, Garmin seems not to care about it. My wife also experiences exact the same thing while running.

Yes this is the one thing that does annoy me about my F6. In the end I got a second hand Garmin footpod and it gives better pace. Not perfect, but it works.

I agree though we shouldn't have to hack around it.

silentvoyager said:

Its largely due to the nature of GPS positioning that instantaneous pace can be weird

No it is not. It is largely due to Garmin's faulty algorithm.

Instantaneous pace has ALWAYS been a problem for sports watches. Yes, algorithm's can make this worse - but the nature of the device, and GPS, doesnt help

GPS Accuracy of Garmin, Polar, and other Running Watches - Fellrnr.com, Running tips

"I evaluated the real-world accuracy of GPS watches while running over 12,000 miles/19,000Km and recording over 50,000 data points as part of my evaluation of the Best Running Watches. Under good conditions most of the watches are remarkably good, but when things get a little tough the differences become more apparent. However, none of the watches have GPS accuracy that is good enough to be used for displaying your current pace. As a result, I've added the test results for various Footpods as they can be far more accurate than GPS, but more importantly they tend to have far less moment-to-moment variation so they can give a far better display of your current pace. (Note that my accuracy tests focus on the ability to measure distance, not the moment in time position, though the two are obviously related.)"

silentvoyager said:

What you need to look at is the GPS distance in every sample and how it changes every second - that is your true GPS speed.

Then factor in the positional error (in the order of metres) per reading. That is a huge error. And when you go under tree cover, you see that for yourself as this is very much where instant pace goes haywire. I see it myself on tree covered trails. The error goes up further.

Hence the use of accelerometer algorithm's to try and bring that back down. And as you said, it looks better after the activity because then trendlines etc can be plotted and the data, en masse, can be used to drag errors down to a more acceptable level.

Hence why its often best to use either a x-second smoothed pace, or if you can use it, lap pace.