Video Reviews on YouTube
Visit me on Instagram
Follow me on Strava
Visit me on Facebook
RSS Feed
Visit me on Twitter
Share by Email
Visit me on Google+

Daily Steps Accuracy Tests

I have noticed while testing various products alongside each other that the STEPS tracked between different watches vary quite a bit. With this in mind I have performed some simple tests on the following products

Test 1 – Walk around Block

The first test was a quick qalk around my local area, down to the seafront and back again. While walking I mixed my pace up from dead slow to a gentle jog. I have previously performed several manual calibration tests on my Huawei P9 phone counting each step and checking the phones accuracy throughout. I have found the phone to be highly accurate and never more than 1% out from numerous tests up to 2,000 steps. As a result of these previous tests I am using the Huawei P9 as the CONTROL.

Result from Test 1

As can be seen from the image above, the first walk test recorded 1,080 steps on the Huawei P9 CONTROL. The watches on test recorded, in order of similarity to the P9

  1. Garmin Forerunner 35 > 1,091 steps > Difference = 11 steps / 1% variation
  2. Suunto Spartan Ultra > 1,131 steps > Difference = 51 steps / 4.7% variation
  3. Polar M300 > 992 steps > Difference = 88 steps / 8.1% variation

Test 2 – Throughout the Working Day

This is where I have noticed the largest variation. I took step readings from the three watches at 2pm and 5pm. There was no specific activity during this time, just working at desk, up and down stairs a few times etc. I did NOT have the phone recording steps at this point so I do not have those readings for comparison. The results of this working day test are shown below …

2PM Recording

  • Garmin = 3,356
  • Suunto = 4,992
  • Polar = 5,360

5PM Recording

  • Garmin = 3835 an increase of 479 steps
  • Suunto = 6329 an increase of 1,337 steps
  • Polar = 5586 an increase of 226

Test 3 – Accuracy During Strenuous Exercise

This final test was performed during strenuous exercise. I took the 3 x watches and the P9 phone for a 4 mile run. I ran on various surfaces from pavement and grass to mainly beach. BEFORE starting each run a recording of each watch’s total step count was taken.

Results from 4 Mile Run

The Huawei P9 tracked 8,116 steps during the exercise.

The watches on test recorded, in order of similarity to the P9

  1. Garmin Forerunner 35 = 8,126 steps > difference 10 steps / 0.1% variation 
  2. Suunto Spartan = 8,274 steps > difference 158 steps /  2% variation
  3. Polar M200 = 7,529 steps > difference 587 steps / 7.2% variation 

Conclusions and Summaries

During Activity

During actual walking and exercise the step accuracy of all products was acceptable. The Garmin came out on TOP for both the walk and run test followed by the Suunto then the Polar (comparing them to the phone). The largest variation I witnessed was 8%. I find this acceptable and put it down to individual sensitivity of the accelerometers.

During Working Day

The confusing results occurred during the “normal” working day. Remember, I wore ALL watches on the same non-dominant wrist, yet they differed GREATLY.

From 9am to 5pm the watches recorded > Garmin @ 3835 steps, Polar @ 5586 steps and Suunto @ 6329 steps. A range from lowest to highest of 2,494 steps, a variation of approximately 40% from lowest to highest … This is quite a significant variation for anyone looking to achieve a daily step goal of 10,000 steps. 10,000 steps on the Garmin would record approximately 14,000 on the Polar and conversely 10,000 on the Polar would be only 6,000 on the Garmin.

More tests to come …

Unfortunately I cannot say which watch is better or worse throughout the day as I did not have the phone on me as a control. I will conduct another test with the Huawei phone on me throughout the day in the near futre

If you found this test informative and interested, then you may be interested to know there are more tests in my BLOG

Please show your support by clicking the adverts below

The CardioCritic website exists solely through the revenue earned from our shopping and advertising partners. If these reviews have helped you in any way please use one or more of the links below to show your appreciation. It costs you nothing to support us, just a click of the mouse. Thank you

Tristan Haskins – aka the CardioCritic

Posted in Testing | Tagged , , , , is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon properties including, but not limited to,,,,, or Likewise, we are members of other private affiliate programs and may earn comissions on purchases done by referrals at no additional cost for the buyer.