Average Heartrate and Calories Burned Inaccuracy

[SunsetRunner]

The simple fact is, the Fitibit is not calculating either the average heartrate, or the total calories burned correctly - when compared to many, many other sources of information.

Today I performed an hour of spin with a chest strap connected to Endomondo, and the Fitbit. Heartrate tracks nearly perfectl between the two - which some variation I would expect. However, the Fitibit calcuated my average HR as 126bpm, while Endomondo was 135bpm.

Also, the Fitbit calculated my caloric burn as 549 calories, while Endomondo shows them at 827.

Since I can get a .tcx file from Endomondo (unfortunately not from Fitbit for non-GPS activities) which provides heartrate, and using standard calculations to determine HR to caloric burn, I know that Endomondo is tracking very closely. These readings also track with a number of on-line calcualtors.

So, how exactly is the Fitbit working again? What formulas are being used to calculate these values? And can we get an answer from someone who isn't a moderator and will just suggest I reset my tracker?

[VivisFitbit]

hey @SunsetRunner! Your Surge records your heart rate based on PurePulse technolgy. The first step you may want to take to resolve this is restarting your Surge. I think it would be a good idea to check the wearing tips of our Website. As you mentioned, different devices with different technologies will provide you different readings. 

Keep me posted! 

[SunsetRunner]

Hilarious response. It is almost as if you didn't even read my questions. Here, I'll post them again for clarification:

What formulas are being used to calculate these values?

***And can we get an answer from someone who isn't a moderator and will just suggest I reset my tracker?***

Moderator edit: Format. 

[ChaosqueenB]

They won't tell you. It's a secret. That's what another Mod told me.

And: Did you try to restart your tracker? 😛

[bbarrera]

@SunsetRunner

Algorithms are proprietary, Fitbit hasn't disclosed much AFAIK. There is (some) info in the Fitbit API docs:

https://dev.fitbit.com/docs/activity/#get-activity-intraday-time-series

There is an old blog entry describing general info on how Fitbit comes up with algorithms. For some reason Fitbit has removed older blog entries, so I had to use the Wayback Machine to find it. This snapshot is from September 2015:

https://web.archive.org/web/20150911173055/https://blog.fitbit.com/a-brief-look-into-how-the-fitbit-algorithms-work/

For reference, last week I had a spin session at 135bpm average. Recorded using Garmin and also Apple Watch, here are the stats:

1:11:59 (basically an 1 and 12 minutes = 1.2 hours)

Garmin - 784 calories

Apple Watch - 788 calories

I didn't take the time to calculate the difference between the two heart rate graphs, but visually they appear to match closely. In the past I've output TCX and extracted heart rate (every 3 seconds for Apple Watch), and in more than 9 out of 10 times found the Apple Watch to be within 3bpm of the Garmin chest strap during the workout.

[ChaosqueenB]

Thank you for the Fitbit API doc! 😄

[Rustyw007]

Hopelessly inaccurate for me too. Very disappointed as I like having only a single tracking device.

What I know is that the Surge simply isn't tracking HR most of the time during strength training (it may work perfectly well for other exercises). What it does track averages out well below the actual exercise, and all peaks (from HIIT) are completely missing.

I've worn it 1, 2 and 3 fingers up from my wrist and even worn it with the sensor facing "inside" my forearm (where the vessels are closer to surface of skin).

I'll look like a total dork, but I'm going to wear my ancient Garmin Forerunner 210 + Chest strap _AND_ the Surge at the same time on my next workout to capture the differences.

I've already confirmed (via manual timing of my heartrate during an actual routine) that the Garmin is +/- 5% of my true HR.

I really wanted the Surge to work, but I suspect after the review of my next workout, I'll be returning it.

[Teacherbytes]

I have experienced heartrate and calorie inaccuracies too. In fact the inaccuracies could have killed me because when I was riding in hill country (I live near the coast) my Garmin with a chest strap said I was in Zone 6 and over 100% of my max heartrate. When I checked the Surge I was also wearing, it showed the highest my heartrate got was 150 bpm compared with 172 bpm my Garmin reported. To me this is a dangerous inaccuracy and could have hurt me if I had been only using the Garmin for tracking my heartrate on that ride. 

On the calorie side, I fininished a spin workout and my instructor asked how many calories did I burn. When I told him about 600 he seemed puzzled about this. In another class I wore a chest strap connected to a Vivoactive and the calorie burn was about 800. 

The bottom line I don't know if I can trust the fitbit heartrate monitor but I read an article about how one may have saved the life of another user.