Optical heart rate is all the rage these days, its comfort just too nice to pass up on.
Every Garmin fitness and sports wearable above the most basic, almost all new Suunto watches – oHR is everywhere.
As we’ve already seen in the look at optical heart rate from a Garmin fenix 5X, it’s a bit of a risky proposition in actual use.
oHR delivers results that are better for a little guidance than nothing, but no good for exact analysis.
After the look at Garmin’s fenix 5X with their Elevate sensor versus a (Suunto) chest strap, it’s time for an in-depth look at the Suunto performance with their Valencell “wrist heart rate,” as they call oHR.
The watch I have, to use for that (and received to beta test new firmware for Suunto), is a Spartan Sport WHR Baro, so here goes with that.
Suunto Spartan Sport WHR Baro – Wrist Heart Rate Performance
The “Big” Comparison of October 1
We finished the look at the Garmin fenix 5X oHR with these October 1 recordings that included a HR belt connected to a Suunto Ambit3 Peak and the Spartan Sport WHR Baro’s optical HR measurement.
As already discussed, there are many spots where it looks like the Spartan Sport WHR Baro provided results closer to the heart rate chest strap than the fenix 5X – but the opposite can also be found.
Things get interesting with other comparisons (where I did not have the fenix 5X on loan from Garmin anymore, but “only” a Suunto Spartan Ultra and Sport WHR Baro, if you’re wondering)…
November 26 Cycle-Run-Cycle
End of November, it’s gotten cold, I went out into the dark. oHR went slightly crazy:
This is not untypical – for oHR to have serious issues during a warm-up period, that is – but very telling a problem for it, nonetheless.
oHR is funny that way: It is good for 24/7 observation of one’s heart rate (although that’s hardly “warmed up”), but not so good during the beginnings of an active phase.
During running, things got to be better, but hardly as good as one would like them to be…
… and the HR recording went on similarly on my bicycle ride back: Much better than during the warm-up, but still with odd peaks/differences.
It takes a bit more than such impressions to get an idea of what is going on and how oHR is performing, though…
November 28 Trail Running
Two days later, and I was getting rather better warmed-up… and going on for quite a bit:
How does that look?
Enough similarity in the curves, from a qualitative look, to not make oHR useless (same as we saw with the Garmin fenix’s oHR). And still some pretty decisive peaks from the wrist heart rate that should rather not be there.
It’s worth having a look at the averages and zones data recorded on that same run:
The Suunto Spartan Ultra, of course using a heart rate chest strap, recorded between 90 and 182 bpm and an average HR of 153.
- Easy: 0:38’21
- Moderate: 1:03’20
- Hard: 1:13’12
- Very hard: 2:03’07
- Maximal: 0:09’24
EPOC Peak: 84 ml/kg
The Suunto Spartan Sport WHR Baro recorded between 88 and 201 bpm (okay on the low end, quite a peak compared to the Ultra on the upper) – and the average HR came out at 151 bpm, only a 2 bpm difference from the Baro. Hardly bad.
Similar results with other of that data:
- Easy: 0:51’21
- Moderate: 1:00’19
- Hard: 1:09’37
- Very hard: 1:49’21
- Maximal: 0:16’46
EPOC Peak: 85 ml/kg
The qualitative fit – the one saying that things don’t look *that* bad – is not only supported by the calculations and averages above; the correlation coefficient for the two heart rate data series also points to a strong correlation (coming in at a value of 0.90).
[Altitude data, however, comes out to have a correlation coefficient of 0.9999 😉 ]
Back through November
Similarly, things went with other runs…
The above comparisons are interesting, again.
Sometimes, WHR skews towards higher heart rates, sometimes towards lower.
Some peaks are definitely off, more often than happens with heart rate chest straps (where false peak readings do also occur).
Averages are, as in the November 28 example above, quite okay again; I must also say that things didn’t look too bad judging by the HR I saw shown whenever I looked at a watch during a run.
One thing that is really important to consider: the likely effect of the weather.
oHR under the Weather
We like to think that modern technology, based on current science, ought to give accurate results no matter what. GPS is space-science, you know?
Things are not like this, unfortunately, and oHR particularly so.
It is interesting technology, certainly. Shining light on your skin to recognize blood flow underneath with which to measure heart rate? Pretty nifty!
Just think about it, though:
How much blood is being pumped to your hands? How well is that visible to the LEDs? Are your veins constricted or widened? Is that because of exertion and/or cold? How do the two interact? What about muscular contractions? Light getting in from the sides of the watch?
Nevermind what happens if your skin is more pigmented (lots of work went into that, from what I know), let alone if you have tattoos…
I should also mention that I am always wearing the Spartan WHR Baro where I would wear any watch, (low) on my wrist.
This is not the ideal position for better readings from a WHR device, but given the shape of my arms, the watch would not stay in the recommended position higher up (and if it moved down from there, it would be so loose on the wrist, the oHR LEDs would be “measuring” only the bounce of the watch and the air around my wrist).
One likely problem and the major culprit in the above data (or rather, its recording conditions): It was all recorded as the weather was getting colder and colder. Which one final comparison serves to show quite well…
October 15, Running in Warmer Weather
This was a nice little trail run, recorded with an Ambit3 Peak and HR chest strap and the Suunto Spartan Sport WHR Baro – mid-October, in warmer weather.
Of course, there are still some differences between the chest strap and the oHR recordings, but they are rather less bad than those in all of the November recordings above – and nicely similar, overall, to the shape of the recordings in the “big” comparison shown at the beginning. That was also in warmer overall conditions (if in bad weather), at the beginning of October.
[Interestingly, the correlation coefficient for the two HR data sets here is calculated as being 0.91, so the two data sets only correlate slightly more strongly than those of November 28! They still do correlate strongly, but a mere look at the curves may have led one to think that the data here was considerably better.]
Overall, the conclusions to draw are the same with Suunto wrist heart rate from the Valencell sensor as with Garmin and their Elevate oHR:
The data is usable enough for a bit of guidance with the comfort of getting it from your wrist, without need for a chest strap.
For any sort of physiological analysis that needs exact heart rate data, though, oHR is not recommendable.
Averages look to be alright, generally; snapshot views of heart rate once warmed up can be quite alright, but HR-based training focused on staying in certain HR zones should better be done using a chest heart rate belt.
To oHR or not to oHR?
Just thinking of the decision between a Suunto Spartan Ultra or a Sport WHR Baro, which only differ in the bigger battery or the oHR, respectively… I don’t know.
The indication of overall health status that an average heart rate during one’s sleep and 24/7 overview of heart rate can give do rather appeal to me, and a Sport WHR Baro that gives that data could still be combined with a chest strap if more-exact data were desired. And if not, then wrist heart rate is probably sufficient.
On the other hand, to be serious about HR-based training and about running overall, I’d rather not have oHR when I want exact data or might consider not using heart rate at all. Instead, I might want to run freely, by feeling, and take the longer runtime that a Spartan Ultra’s bigger battery gives.
Suunto or Garmin?
Something of both could be had from a Garmin fenix 5(X). That is heavier (which tends to make oHR performance worse) and just overall a very different device.
If I considered not looking at HR on a Suunto Spartan (of whichever kind), I would definitely not want to suffer the intrusively helpful approach of Garmin – but for guidance and smartwatch functions, a fenix it would be.
Again, supposed performance in measurements, judged from a single device under somewhat haphazard conditions, is probably not the best guide to base a purchasing decision on. Much better to look at the overall package and, actually, to decide by the feel of a device in use.
[Ad / Affiliate Link – if you buy from there, I will get a little commission that helps run this site]