Physiqonomics Weekly #75: Your Smartwatch Is Lying

24 March 2022 | by Aadam


It’s time for another instalment of Physiqonomics Weekly.


Q-All your discussions about carbs is always in the context of weight gain/loss. I never read any discussion about the long term health effects, e.g., high blood sugar, that could lead to diabetes or other chronic disease. Has your research suggested that high carb intake is also not the enemy when it comes to chronic diseases?

That’s because the two aren’t mutually exclusive. When people lose weight, their health also improves. And there are dozens of examples of this.

Like the “Twinkie” professor who restricted his calorie intake and subsisted on a steady stream of Twinkies and a number of other ‘junk food’ items. ⁣At the end of the two months, the “Twinkie professor” had lost 27 lbs while increasing his ‘good’ cholesterol levels and decreasing his ‘bad’ cholesterol levels. ⁣ In 2015 a science teacher lost 56 lbs in six months eating nothing but McDonald’s; in 2016, Pasquale Cozzolino lost 100 lbs eating pizza; in 2018, Anthony Howard-Crow lost 32 lbs eating ice cream.⁣ And more recently, Andrew Flinders Taylor ate nothing but potatoes for an entire year and dropped 117 lbs, while improving his health markers.

I’ve seen the same thing with clients I’ve worked with. As one example, the client pictured below lost just under 70 lbs when we worked together, and all his health markers drastically improved.

Of course, it’s not just about weight loss. Regular exercise is equally important for overall health and longevity.

But, it’s never just one thing that leads to poor health, it’s a combination of factors. This is why solely blaming carbs is comical. People drink excessive alcohol, don’t get enough sleep, are constantly stressed and increasingly sedentary, but it’s just carbs that are the problem?

Anyway, this is a topic I’ve written about in a ton of detail here. And I’d encourage you to read it to gain a better insight into a topic that’s highly nuanced.


Smartwatches Are Still Terrible at Estimating Energy Expenditure

In 2019, I wrote an article where I argued against “eating back” exercise calories. In that article, I reviewed several studies that showed how inaccurate activity trackers can be.

But that was 2019 and as of writing this, we’re on the seventh generation of the Apple Watch. And with tech companies constantly innovating and improving their products and algorithms, has the accuracy of smartwatches improved?

Well, that’s what a recent study by Hajj-Boutros and colleagues (PMID: 34957939) decided to find out. The researchers investigated the accuracy of three recently released wrist-worn devices––the Apple Watch 6, Polar Vantage V, and Fitbit Sense––for heart rate and energy expenditure during sitting, walking, running, resistance training, and cycling.

A total of 60 participants (30 men and 30 women, mean age: 25 years) were equipped with the reference devices the watches were measured against (Polar H10 chest strap and MetaMax 3B; an indirect calorimetry system), and the three wrist-worn devices (Apple Watch 6, Polar Vantage V, and Fitbit Sense).

The participants then performed five different activities for 10 mins each:

1) Sitting on a chair

2) Indoor walking on the treadmill at the participant’s normal pace

3) Indoor running on the treadmill

4) Machine-based resistance exercises (chest press, leg press, and seated cable row) for three sets of 10 reps with a 1 min rest between sets

5) Cycling

The researchers performed a number of statistical analyses to compare the accuracy of the devices.

All the devices were fairly accurate for measuring heart rate (though the Apple Watch had the highest level of accuracy across all activities measured, while heart rate accuracy varied for the Polar Vantage V and Fitbit sense depending on the activity).

Unfortunately, as the image below illustrates, all three devices did a poor job of estimating energy expenditure for all activities measured.

As the researchers write:

Healthcare care professionals, athletes/coaches and the general population may want to proceed with caution on the clinical utility of energy expenditure of these devices during the implementation of an exercise training or nutritional programme.

I don’t think there’s much more to add here and what I said in my 2019 article still stands: You shouldn’t be putting too much stock in the energy expenditure numbers your smartwatch gives you, and you definitely shouldn’t be eating back those calories.

Instead, use these devices as a way to quantify movement and to encourage you to be more active. For example, if you’re aiming for a certain number of steps, and you find you’re short of your daily target, you can get outside for a walk. We also know energy expenditure can drop as you get leaner, and tracking your step count can help you maintain your activity levels so progress isn’t hindered.

Before wrapping up, I want to address a question I’m frequently asked about this topic: What should you do on days where energy expenditure is unusually high? For example, maybe you do a long hike once per week and according to your smartwatch, you burn, I dunno, 2000 kcals, should you “eat back” those calories?

I wouldn’t recommend it for two reasons. The first, and most obvious, is this number isn’t going to be very accurate. Second, the calories you burn during an activity (or training session) may not translate to an increase in total energy expenditure throughout the day. If an energetically costly training session results in increased fatigue and tiredness, you’ll likely move less the rest of the day, so any increase in acute energy expenditure could be negated or reduced. Instead, eat a bit more on those days until you’re satisfied while being mindful of hunger and satiety. Basically, don’t use it as an excuse to eat until you’re sick.

The same holds true in instances where you’ve consistently increased your energy expenditure, like say, you added additional cardio or sports to your weekly training. Instead of eating back the calories or adjusting your diet based on erroneous values from your smartwatch, you should monitor changes in weekly body weight and adjust your intake based on that.

For example, if you added three cardio sessions to your weekly programming and find you’re consistently losing weight faster than you should be, you can increase calories to account for the change in weight loss. On the other hand, if you’ve increased your activity but find the rate of weight loss is unchanged, then you wouldn’t need to adjust anything. The same holds true if you’re looking to gain muscle (increase calories until weight gain is back on track) or maintain your weight (adjust calories so you’re maintaining weight).

(Share on Facebook, Twitter, WhatsApp, or Email)

Thanks for reading. If you enjoy my weekly emails, please consider telling others.

A lot of time and effort goes into writing these weekly emails to make them as helpful as possible for you.

If you enjoy and find value in them, it would mean a lot to me if you could let others know about Physiqonomics Weekly. It helps Physiqonomics grow and keeps me motivated to continue writing the content you love.

You can share this week’s issue on Facebook, Twitter, WhatsApp, or Email.


Enjoyed this?

You just read an email from my weekly fitness newsletter, the Vitamin. Every Thursday I drop some knowledge bombs on your face to help you reach your goals quicker while avoiding all the bullshit.

Get the next issue in your inbox by dropping your name and email address below–

No spam. Ever. Unsubscribe at any time.