Garmin Accuracy Test
In most Running Forums there is considerable debate about how accurate the Garmin can be.
Often Runners end up challenging Event Organisers over the course distance by pointing out the numbers recorded on their Garmin devices. It is commonly accepted that the Garmin cannot be used by Pacers for pacing in Racing events since the distance is always known to be long…i.e. the Garmin will register more distance than what it actually is.
In order to put the level of accuracy to the test, I decided to run my 21 kms Half Marathon run of Feb 2013 with two Garmins. The 310XT went on my left hand and the 405 went on my right hand.
For those who do not know me, I have been running a Half Marathon ( either in an event or the distance by itself ) each calender month for the last 3 years. For the month of Feb 2013 I chose to run on Thursday, Feb 14th which was Valentine’s day in order to maintain another tradition. Every year, myself and Usha visit Pondicherry and run Auroville as a Valentine celebration. But this year Usha decided to take a break for our son’s 10th standard exams and I, the cheapskate, decided it was not worth going alone.
Coming back..besides the respective Garmins on each hand I also put the respective Heart Rate straps on my chest. Being a weekday I chose to start from my house and run in the route of BKC/ BWSL / Mt Mary / Bandstand / Pali / Carter road so that I can finish and return quickly.
My Heart Rate strap for the 310XT is a real disaster and does not work consistently. As anticipated, it stopped somewhere halfway..so this study will ignore the Heart Rate.
I started at 5:30 am from the University Gate in Kalina. going down CST road went into BKC. Somewhere near Asian Heart Sud Vishwakarma crossed me, recognised me and decided to run the rest of the way along with me. Sud is a young athletic runner who does a 1:40 HM without any training to speak of. I then met Pratik Desai who also turned back and ran along with me till the end of BKC. It was nice that we could catch up on his Comrades training plan.
Crossing over the Western Express Highway we entered Bandra Worli Sea Link and at the start met Dr Shirish Kataria and the gang. They ran with me till the toll booth of the Sea Link where we all turned back. I went over to the Lilavathi Hospital / Mt Carmel and having finished 10kms decided to buy a bottle of water. It was an unsupported run. Going up Mt Mary slopes and past the Church to Bandra Fort we met with Praful of Striders and I could catch up on things.
Hitting St Andrews, we crossed over to Otters Club / Joggers Park and did Pali Hill taking in Zig Zag road for good measure. Coming down Pali Hill we ran the sea face of Carter road and then at about km 18 stopped for a Nariyal Pani. We finished the last 3 kms on Carter road by running back and forth so as to end it in front of Cafe Coffee Day.
We both stretched and then treated ourselves to a coffee and shared an indulgent brownie
The two Garmin links for the run are uploaded here at
The Garmin 310XT activity
The Garmin 405 activity
Here is how they compare:
Distance: Both Garmins were stopped as soon as they beeped 21 kms. But the 310XT beeped first. I stopped it and kept running till the 405 beeped 21kms. This gap was about 300 metres or so. A 1% variation, which I would say it is immaterial for any practical reason.
Pace: The average pace as per the 310XT was 8:22 mins per km while the 405 showed it as 8:27
Time: This is really strange if you consider that the time taken by 310XT is shown as 2:57:42 while the 405 shows it as 2:57:39. Bear in mind that this is quite an major error since both the Garmins were started at the same time and location in the morning but the 310XT was stopped first about 300 meters before the 405. So something is not quite right with the time function.
Calories: The 310XT registered 1029 calories while the 405 showed I had burnt more calories at 1142. An almost 10% variation.
Elevation: The 310XT showed a 142 meters elevation gain while the 405 showed only 127 meters.
Runners have always known that the Garmin cannot be very accurate, but until this study, one never knew what was the level of inaccuracy in them. This study probably is the most conclusive report of the variation that can be expected when using these kind of tools for running. Bear in mind that each Garmin was in the hands of the same person.
Event Organisers should be able to rest easy and point out this study when runners start challenging them about course distance being inaccurate.
On the way back I stopped the rickshaw at Vakola flyover as my romantic side took over and I got a flowering rose plant for Usha.
The photos of the run are uploaded here