Why does the overall max heart rate auto detection feature result in a value lower than my measured activity max heart rate?

You will find some answers here: How Are High Heart Rate and Max Heart Rate Values Calculated on My Garmin Watch? | Garmin Customer Support

Anyway, I'd recommend getting your Max HR from a reliable source (lab test, or a field test), and then disabling the Max HR auto-detection.

Thanks for the response. Before posting this question, I did read on that page

max heart rate can also adjust upward or downward based on a proprietary algorithm designed by Firstbeat Analytics

But it seemed wrong to me that the estimate resulting from this "proprietary algorithm" would be preferred over actual (repeated) measurements during actual activities. Apparently that is the way it is...

Also, I was disappointed that this estimate of overall max heart rate was more than 10 bpm below my actual overall max heart rate (as found by measurements). That is not a good estimate, so what is the quality of the estimates of any of the stats that (heavily) depend on overall max heart rate, like VO2max?

7414949 said:

I was disappointed that this estimate of overall max heart rate was more than 10 bpm below my actual overall max heart rate (as found by measurements)

JFYI, the Max HR is no way the highest HR you reach, and never should be. The Max HR is the HR you can sustain at an all-out effort for at least  a few tens of seconds (at trained athletes it can be up to a minute, but typically is it around 20-30s). That's why your Max HR will practically never be equal to the highest detected HR during an activity. Still, it is much safer to get the Max HR from a lab, and disabling the auto-detection, because the auto-detection is dependent on the accuracy of the HR sensor, and that may be a problem if you do not use a HRM strap, or when the HRM strap malfunctions (dry electrodes, low battery, connectivity problems, etc).

Right, I forgot about that. Keeping that in mind, I still see activities that have several tens of seconds of sustained heartbeat a few BPMs higher than the max overall estimated value by Firstbeat and I still think that is too bad.

This discussion has made me wonder how the estimated max overall heart rate will evolve as I increase my running activities though and I may now switch back to automatic max heart rate detection just out of curiosity :-) 

7414949 said:

how the estimated max overall heart rate will evolve as I increase my running

Depends on your fitness level when you started. It’s not so much your HRmax increasing, it’s more your fitness increasing and giving you an ability to reach a higher heart rate. If you’re a well-trained athlete it’s very unlikely there’ll be any increase. The heart is a muscle that responds well to training but can only do so much just like any muscle. 

Dr Phil said:

It’s not so much your HRmax increasing, it’s more your fitness increasing and giving you an ability to reach a higher heart rate.

Yes, indeed, Max HR does not increase with fitness. Researches tell the Max HR only linearly drops with age, and is independent on the fitness level. You will actually reach the Max HR much quicker if you are unfit, than when you are highly trained, but the limit remains the same regardless of the fitness level, in contrary to the popular belief.

Maximal heart rate declines linearly with age independent of cardiorespiratory fitness levels - PubMed (nih.gov) 

In fact, some studies even found that the Max HR may decrease with increasing physical fitness, due to the higher cardiac output of the heart when you are trained. The higher effort and longer time required at a trained person to reach the Max HR contribute to it too. Some links to whitepapers explaining it, are listed for example here:

Maximum Heart Rate: The COMPLETE Guide (upsidestrength.com)

Dr Phil said:

It’s not so much your HRmax increasing

I am aware that the actual HRmax will not change (much). What I was talking about is the estimated HRmax by the Firstbeat model, just because I am curious how the model works. I know for sure that the current estimate is too low, so I wonder what it takes for the model to get to an improved estimate.

trux said:

Maximum Heart Rate: The COMPLETE Guide (upsidestrength.com)

Thank you for sharing, that is an interesting read.

7414949 said:

just because I am curious how the model works

The patent filed in Europe by researchers from Firstbeat is very likely the way it works.

data.epo.org/.../document.pdf

That is a great find, thank you.