Fitness monitors and accuracy

It is so great to have a fitness tracker and get objective feedback on the workout, making sure I train on target. 

I have a Garmin Forerunner 735X which I am very pleased with. It has enough battery power to run GPS for a whole day which saves me from carrying a smartphone. My Garmin can also measure swim distance.


I have noticed that the recorded running distance often appears to be slightly short. I previously used a Fitbit, which also was great. The Garmin records shorter distances in comparison to the Fitbit on the same tracks. This has to do with how distance is measured by the tracker. 

The Fitbit used my estimated stride length, which I had calibrated on a measured course. The backside of that was when I got faster and my strides longer, the Fitbit recorded a marginally shorter distance!

Garmin use a GPS grid and interpolates distance between points. I think this is where the discrepancy starts, if the track has many curves, the grid has a tendency to straighten the corners. Last summer I ran an officially measured 5km race and the Garmin gave me 4.99 km. 

I often run a trail nearby that has markers every 0.5 miles and I run 3 miles (1.5 miles out and 1.5 miles back), but my Garmin usually gives me 2.91 miles. This difference also affects the calculated speed/distance.

When swimming lenghts in the pool, my Garmin seems to occasionally credit me with an extra lenght or two making my time/100yards much better than it really is (most of the time, the tracker measures the lenghts correctly). I am not sure of the source of that problem, maybe the tracker gets confused when I turn on my back.

On the bike, my Garmin seems to be a little stingy. When I bike with my friends and we cover exactly the same course on the same time, I usually have recorded a slightly shorter distance than my friends. While this difference usually is very small, when it comes to elevation gain, it is significant. Here is just one example from a recent bike ride:

person 1: distance 27.18 miles, elevation gain 857 feet

person 2: distance 26.66 miles, elevation gain 863 feet

person 3: distance 27.33 miles, elevation gain 1043 feet

person 4: distance 27.17 miles, elevation gain 858 feet

person 5: distance 27.46 miles, elevation gain 1045 feet

person 6: distance 27.51 miles, elevation gain 1047 feet

While the discrepancy in distance is small and can be explained away by minor deviations on the route, the recorded difference in elevation is alarming. I am not sure what trackers the other people had, but I see two clusters: elevation gain around 860 feet and a second cluster around 1040 feet. 

Elevation changes can be measured from barometer data and from the map data, maybe this is the difference we are looking at? In any case, one must be aware that fitness trackers are not exact and I compare primarily with myself. I look only at the big trends, ignoring minor details as noise.

/Gunilla

Comments

Popular posts from this blog

Do not wait until it is too late!

Carbon plated running shoes

Multisport National Championship Festival Irving TX April 19-23