​It's OK that optical HR tech isn't perfect, but we'd like a bit more honesty

Why brands should learn from Under Armour
Wrist-based HR's dirty secret

Over the past few weeks it's become clear that the latest crop of optical heart rate sensors just aren't up to task.

That might not come as a surprise to the more cynical of Wareable readers, but let me justify my thinking.

When Fitbit and Apple started putting heart rate tracking on the wrist, I naturally wasn't convinced – and nor were the likes of Garmin and Polar. Those big guns of the fitness world were staying clear of Mickey Mouse gimmicks, and kept their focus firmly on the chest strap.

Then came Mio. When we spoke to Liz Dickinson, who's credited with inventing optical heart rate tech, she rubbished the work of Fitbit et al, claiming Mio's was the only capable technology.

And the tech industry responded.

Last year TomTom and Garmin both used Mio's tech, and both have dropped it for 2016. There are now a slew of devices, including the latest Vivosmart HR and Garmin Forerunner 235.

That led me to believe, mistakenly, that the technology was ready for prime time. I felt that if Garmin was finally backing the technology, it must be good enough – but that's not the case. And while I'm not an athlete that relies on my workouts being fixed within narrow and specific zones, once I know that something's not accurate, the data becomes junk to me.

And the most telling evidence comes in the form of a product that's not even been released yet.

Under Armour's new band comes bundled with a separate heart rate monitor designed for exercise – a clear admission that the technology isn't capable of cutting through the noise and movement experiences during high intensity workouts.

"We're being pragmatic about what you can do on the wrist and what you can do on the chest. We want to appeal to athletes," said Claude Zellweger, VP of Design at HTC, which created Under Armour's technology.

And while that's refreshing to hear, it confirms what we already know: if you're serious about your data, then the latest optical-toting wearables aren't up to scratch.

So should you disregard these devices altogether? Absolutely not. The Garmin Forerunner 235 is a fantastic device, which will pair with ANT+ chest straps for those HIIT sessions. It's not the technology's failings that sits uncomfortably with us, it's that Garmin and Fitbit aren't being as open as Under Armour about what their tech can and can't do.

The Garmin Forerunner 235, for example, utilises its optical tech to great effect when you're not working out. And if you pair with a chest strap for a HIIT sessions you'd have an incredible set of fitness data – it's just a shame that Garmin never really mentioned that.


  • RichFrankel says:

    This is a really well thought-out commentary.  I've used just about every watch with a HR tracker available from Gears to fitbits, MS band, AW and currently the Apple Watch.  They are not accurate from device to device, but the fitbit and the Apple Watch have both been consistent with themselves.  That is, if I have a resting rate in the 60s with the Apple Watch and I have a cardio rate that shows 120 if I'm doing a specific activity, I can count on the watch to show me in that range whenever I work at that level.  It may not have anything to do with my actual heart rate, and people need to be informed and not rely on their watch while their hearts explode, but as a relative tool, it works.  

    I think the issue with the UA products coming out isn't as clear as their press makes it sound.  Sure, chest straps are far more accurate and will be for some time, but it's a bit disingenuous for them not to point to the fact that by having another piece it adds commercial value can't be ignored. The HIIT issue that you raised is 100% true, and I think even once the optical tech gets dialed in to be accurate, it's going to be some time before it's also *fast* and so even if it dials in a perfect medical-grade HR, if it doesn't get you there in 10 seconds or less, the data is already too old for some HIIT situations.  

    Anyhow great piece.  Optical is the future, but it's got a way's to go.  

    • j.stables says:

      Interesting take especially on the lag time stuff. Thanks for commenting.

  • yafizicist says:

    My trouble is, MIO and Scosche (Valencell) sensors have been shown to be accurate. So why are all the rest so bad? If they can't make it right why not license it? I know there is a big problem with battery life for 24/7 versus intensity training, so let us choose the frequency and accept the battery hit. I had the Vivosmart HR and it was a great device but the HR sucked at cycling. Someone's going to get it right and blow everyone else out of the water.

  • TheWerewolf says:

    "And while I'm not an athlete that relies on my workouts being fixed within narrow and specific zones, once I know that something's not accurate, the data becomes junk to me."

    And there's the problem. You're wrong. If you're measuring the weight of say an atom, sure - but measuring heart rates isn't a precise action, so demanding too much precision isn't just pointless, it's misleading. It's 'theater of precision'.

    if your heart was a perfect machine that operated exactly the same way on each pulse, maybe - but your heart isn't - the exact sequence of contractions and relaxations vary so knowing when a pulse starts is tricky enough.. but you're reading it not at the heart, using a myoelectrocardiogram (which measures the electric signal generate by the muscles of the heart itself), you're reading the effect of that pulse on the thickness of blood vessels all the way down by your wrist - which means the force and timing will be affected by all sorts of things including something as simple as how your wrist is bent.

    At best, it's going to be an estimate - so the real question isn't "is it accurate" (because it's not), but rather is it useful data (which is may well be).

    Except you've just eliminated that because you apparently want precision accuracy where that simply doesn't exist.

    • j.stables says:

      I get what you're saying.

      This is my point: If you're supposed to be training within a specific HR zone (which is the whole point of HR training) if your wrist-based monitor is as much as +/- 10 bpm out, then you could be training in totally the wrong zone, undermining your entire plan. For me average HR is just another piece of data, but to others, it underpins their entire plan. Which goes back to what you're saying about the data being useful.

      Also, the title of the piece is "it's OK that's optical HR isn't perfect", because our points align. It's not going to be totally accurate. But there's a lack of transparency over the usefulness of these readings.

  • SalusLife says:

    The Basis Peak device pictured is one of the most accurate HR devices on the market with validation white papers available showing its accuracy to be within 2.3% of a medical 4 lead ECG. Not bad. 

    "The Basis Peak fitness and sleep tracker demonstrated excellent correlation with clinical- grade heart monitoring equipment (ECG) during a validation study executed at UCSF Medical Center. Basis Peak displayed high accuracy and reliability, consistently reporting heart rate readings across a range of activities." 


What do you think?

Connect with Facebook, Twitter, or just enter your email to sign in and comment.