Wrist based vs chest strap HR monitor

Currently using my Fenix 6 for HR during all rides and TR sessions. Anyone got any thoughts on how accurate it is compared to chest strap? Do they typically read high/low or spot on?

I have a garmin watch. It’s fine for walking around. Utter and complete trash for on the bike.
It typically reads about 20-70bpm on the low side.


It depends :slight_smile:

I have a strap (HRM-Run) and a Forerunner 245 which can broadcast HR to my laptop or bike computer. I almost always use the strap, the only time I don’t is when I realize I’ve forgotten it after I’ve started a ride then I’ll turn on the broadcast.

When the wrist based monitor works it works, but when it doesn’t you are just SOL. The less pigment you have the better, apparently. I find that as long as I keep it pushed down from my wrist and not necessarily tight but it no play it works really well. Either way it works much better on the bike than it did when I was running when it would sometimes show my cadence or double or halve my HR. It was maddening.

1 Like

Same experience.

I used a Wahoo arm/wrist HRM. It was picky about placement. Never had much luck on wrist (same problem with a Garmin FR) - needed to be tight and kept in place. Upper arm worked better. But, a chest HRM seems to be more consistent and less likely to give wonky results.

1 Like

Have a 945 and the HRM-PRO. I use the HRM-PRO with TR but sometimes have the watch on. I was perplexed for months about why the readings were off on the trainer, often never breaking 80-90 on the 945 when at 160+ via the HRM-PRO. It worked fine running and swimming, or even on hikes. But on the trainer didn’t work.

Then I learned somewhere on here that if it is recording an activity it is WAY more accurate. It will also be more accurate if it is put in the transmit/Broadcast mode (can’t remember the correct name). I tested that on 2 TR rides, never actually connecting it just broadcasting it and now finally whenever I looked at the screen vs the watch they were pretty close to each other.

I don’t understand why it will show me a correctish HR if I run up some stairs or mow the lawn, but not while riding on the trainer. Maybe it thinks I’m not moving and it can’t possibly be right? Anyway, if you are recording or broadcasting its probably decently close, but I personally prefer the strap.

1 Like

Does it actually matter? If it is off always same amount, isn’t this already good enough – shouldn’t we look at trends, not specific values anyway? :thinking:

There are a few other threads about wrist based hr that you might want to check out. Even some people who LOVE the Whoop and the Fenix will tell you they aren’t very accurate for cycling HR data, for example (of course there are also people who will tell you they are very accurate for themselves too).

I think the bottom line is that “it depends on the user and the monitor”.

My personal experience is that a Fenix and a Whoop (on my wrist) aren’t very accurate most of the time, especially for things like weight lifting, but I did get very accurate data from a Scosche.

1 Like

My Whoop when on my wrist is heavily affected by vibration and excessive movement and measures high for large parts of almost any outdoor ride. I went back to wearing my Garmin Tickr chest strap and recording that separately on my 530 to get some data. While the low readings basically match, the max reading can be 25+ beats higher on the Whoop and the spikes are frequent enough that the average is 15 to 20 beats high too. Its definitely the wrist placement that is the issue. If I wear the Whoop on my upper arm using the bicep strap, ride data is within 1-2 beats for min, max and average of my my chest strap readings.

1 Like

If it were only off by a few bpm, that would be fine.

But, it’s often off by 20+ bpm. And not consistently. The wrist strap is more prone to interruption by movement and reading differently based on placement.

FWIW, my Apple Watch is more consistent than my Garmin FR ever was, but it doesn’t broadcast, so I’ve only tested it running, not on the bike/trainer.

Check out https://www.youtube.com/@TheQuantifiedScientist. Not sure if Fenix 6 was reviewed but 7 was.


Personally hate the feeling of a chest HR monitor. Have a forerunner 55 and it is accurate verified through me taking my pulse. It has misread on occasion but is rare. I would say data is correct 99% of the time.

1 Like

The watches are just pulse monitors, but the straps detect electrical signals. I would not trust the data from wrist based metering.

From a recent cyclingweekly article:
In summary, for most types of cycling a wrist-based sensor will be acceptable, but if you’re using heart rate in any serious capacity or for racing, it’s still best to use a chest strap.

A chest strap measures heart rate, a wristband measures pulse. Not every heartbeat will lead to a pulse beat, especially when the heart beats too fast for the left ventricle to fill up, so…

As has been mentioned, for accurate results from your wrist based Garmin the key is to turn on broadcast HR and/or start an activity. The readings otherwise are less reliable.

For me, I exclusively use my Garmin Instinct 2 now and it is consistent, accurate, and easy to use.

I had a Wahoo Tickr arm one that was pretty accurate but sometimes had times where it just didn’t want to pick up the HR. I have just as much of a consistency problem with getting the chest mounted HR to get going properly.

I think there’s not much wrong with using wrist based these days.

Like I explained above pulse and heart rate are fundamentally 2 different things, often the same but not necessarily, not every heartbeat leads to a pulse beat but every pulse beat is always the result of a heartbeat

You’re posting in a thread full of people saying their wrist based HR does or does not work for them. It’s clearly an individual thing and a case of “how much bad data are you willing to accept” for many.

1 Like

That’s not how it works. When reading well, wrist-based HR monitors can be reasonably accurate. When not, they’re basically random number generators.

Relying on a wrist-based HR monitor for anything other than entertainment value is a bit like relying on GPS for speed when doing aero testing. Yeah, you can, but you’ll never get great results.

Yeah, you’re right – when it works (Garmin Fenix 7), it produces smooth HR graph, but ~20% of times it is really jumpy, definitely not real values. Anyway, I’m mostly using it as 100cal interval timer to get some fuel.

1 Like

I’d suggest you run some tests, that’s what I’ve always done with the garmin watches I’ve owned - 935 and now 955.

Download Auxiliary HR - Connect IQ Store | Free Watch Faces and Apps | Garmin and test your watch against something you know to be accurate.

I used a Polar H10 which usually gives accurate results to initially compare. The 935 was absolute garbage, and unfortunately the 955 for me wasn’t great on the bike or running either - both would miss peak HR during intervals/hill repeats, so I’d get a square profile instead of a sharkfin/hill in the HR graph.

To make matters worse I found the Polar H10 would often die halfway through runs/rides (battery was fine). When it worked it was accurate, but it lost connection on me a lot. I moved to the Polar Verity Sense and I’ve found it to be perfect, easy to live with and accurate enough for me. Intervals are always captured correctly and despite the Quantified Scientist finding the H10 more accurate, I find the Verity Sense more reliable and more than accurate enough. Never had a single issue now with a couple years of use. Unfortunately despite the glowing reviews for OHR on recent watches I don’t find they work as well as reported on my wrist.

Do a test yourself to check :wink:


I’ve spoken to Garmin about this issue and they are really doing themselves a disservice by either not understanding the issue or not relaying the importance of starting an activity in order to get (more) accurate HR readings.

My original 235 used to only give good data during exercise if using the “broadcast HR” or recording an activity. My current Instinct Solar only gives good data when recording an activity. (I believe the sampling method hasn’t been setup correctly). Anyway, you can even see it in the data when using TR. The data is trash until I begin recording an activity on the watch, then it is really quite good (I have compared it to other sources). So now I just use activity recording on the watch while I record a ride on TR or an outside ride on the Garmin Edge and discard the watch recorded afterwards.