In this video I dive into the wild world of power meter data recording, showing how even with an accurate power meter your bike computer can still misrepresent your efforts. Using the Favero Assioma PRO RS pedals and the SRAM AXS Quraq as two baselines, I tested five popular head units (COROS DURA, Wahoo ROAM V3, Hammerhead Karoo V3, Garmin EDGE MTB, and iGPSport BiNavi) and found some big problems with how intervals and sprints are logged. I break down why this matters for training accuracy, long-term progress tracking, coaching feedback, esports validation, and overall trust in cycling tech, plus share analyzer data so you can see the numbers for yourself.

In follow up to comments from iGPSport and others regarding the “Including Zeros in Averages” setting for power meter sensors – The issue presented within this video **IS NOT** related to that setting.

IGPSport support has commented: “The possible reason for the discrepancy observed in the blogger’s test may be that the “Average Power Zero-Value Filter” was enabled.”

It was NOT enabled for any of my testing presented in this video.

This reply from iGPSport is indication they didn’t understand the issue, as such their reply is misleading and has resulted in another commenter attempting to call my testing out as being invalid (in not so generous terms).

I’ve spent more time performing additional tests to get a better understanding of exactly what that “Include Zeros in Averages” setting does in regards to the power meter data on both iGPSport and Garmin bike computers.

The summary for those short on time: The average power calculations in my analysis were performed by the DCR Analyzer tool. The DCR Analyzer does NOT use the “ave_power” data stored in the “session” table within the FIT file. The issue of sticky watts / duplicated recording of power values when coasting occurs prior to any ‘averaging’ done on the bike computer. As a bonus – this additional testing has identified the issue is also occurring with the iGPSport IGS800 bike computer.

Further Testing Full Details (August 30th 2025):

Devices:
● iGPSport BiNavi firmware 1.19
● iGPSport iGS800 firmware 1.61
● Garmin EDGE MTB firmware 28.20

Power Meter:
● SRAM AXS Quarq.

Test 1 – Average Power INCLUDING Zeros.

● “Exclude 0 values when calculating the average power” DISABLED on both units. The wording logic is a little confusing. In disabling “Excluding 0 values” the device will INCLUDE 0 values. Sigh….

● This is the default/factory setting for iGPSport cycling computers . This is also the default behaviour of other cycling computers (including the EDGE MTB). iGPSport stated “It’s recommended to disable this setting,”. It was. Nevertheless, I did another controlled test.

● ~6mins outdoor riding with intermittent pedalling (coasting tests).

● Observations:
– Both the iGS800 and BiNavi exhibited the identical issues with power data as in this video review.
– The Garmin EDGE MTB recorded the Quarq data as expected.
– BiNavi Average power: DCR Analyzer = 224.28. FIT FIle Avg Power: 220W
– iGS800 Average power: DCR Analyzer = 223.40W. FIT FIle Avg Power: 219W
– EDGE MTB Average power: DCR Analyzer = 132.66W. FIT FIle Avg Power: 137W

● Conclusions:
– Results similar to testing presented in the YouTube video. It was of no surprise the iGS800 has the same power meter sensor data recording issue as the BiNavi.

Test 2: Average Power EXCLUDING Zeros.

● “Exclude 0 values when calculating the average power” ENABLED on both iGPSport units. This setting will now EXCLUDE 0 values.
● EDGE MTB was set to the non-default setting of “Do Not Include Zeros” for Power Data Averaging.
● This setting IS NOT a recommended setting as it overestimates average watts.
● ~6mins outdoor riding with intermittent pedalling (coasting tests).

● Observations:
– As in Test 1, both the iGS800 and BiNavi exhibited the identical issues with power data as in this video review.
– The Garmin EDGE MTB recorded the Quarq data as expected.
– BiNavi Average power: DCR Analyzer = 250.04W. FIT FIle Avg Power: 257W
– iGS800 Average power: DCR Analyzer = 250.04W. FIT FIle Avg Power: 257W
– EDGE MTB Average power: DCR Analyzer = 149.55W. FIT FIle Avg Power: 163W

● Conclusions:
– All units overestimated the power data averages within the FIT file summary due to not including zeros. Yes, this is a “water is wet” moment. – The numbers on the iGPSport aren’t as far apart as I’d expect… I should have coasted more during this 6 minutes. See an additional test below for a better example.
– Note: I suspect there was a cloud/sync issues with the iGS800/BiNavi resulting in the same FIT stored for this test. This has no impact on the result(s), there was no need to use both. More testing is needed on this specific setting, butthis was irrefutable proof the raw power data recording issue occurs regardless of this setting.

Additional Test

Take Outs:

● The ‘Exclude 0’ option for power data on iGPSport cycling computers has no impact on the sticky watts / repeated power data issue as presented in this video.

● The analysis presented in the video is on the raw data from the FIT file using the DCR Analyzer tool. This tool has its own averaging algorithms to report averages for the highlighted sections (as shown in my video analysis of the data).

So in response to the comment “Self own. You are excluding zero values from power calculation, hence the average power is higher. Loool.” – That’s an incorrect conclusion. This setting has no effect on the raw power data being recorded.