Apple Watch Calorie Equation Ignores Crucial Factor

A few weeks ago, I updated my Apple Watch (which I’m going to have to do again tomorrow), and I noticed something curious while setting up my Activity profile: Apple gives you three options to describe your sex. Now, dont misunderstand the issue here. “Male,” “Female,” and “Other” is a perfectly acceptable rubric for describing — in a very general way — most folks’ interpretations of who they are. Indeed, there are dozens of subcategories within that “Other” category, and those are all fine too. Nobody’s personal identities and lawful life choices are to be questioned. Ever.

But when it comes to Apple Watch’s goal of accurate calorie tracking during workouts, this is a big problem. While there is a tremendous amount of empirical data on the dramatic differences between male and female metabolisms and calorie burning at rest and in action, there is no reliable equation for the very diverse “Other” category. Worse, Apple allows users to skip this section altogether.

As a former certified personal trainer (big whoop, I know), that set off a few alarms inside my head. The goal here, obviously, is nothing more or less than maximum accuracy, and biological gender is a genuinely important variable in all medically accepted caloric intake and expenditure equations. So, to make a long story short, I wanted to see what would happen if I did the same exercise — for the same amount of time at the same height, weight, and approximate heart rate — using all three gender categories offered. (To do this, I had to completely reset my Apple Watch before each workout, as simply changing the parameters in the iPhone’s Apple Watch Health app often doesn’t take on the wearable for some reason, and I wanted to eliminate any potential contamination of the data.)

My routine was simple: I walked around a flat, half-mile neighborhood circle for 40 minutes while keeping my heart rate at roughly 100 BPM. Then, having saved the workouts, I compared the details of the data per the following screens (in order of Female, Male, Other):

(The timestamp for the “Female” workout’s setup screens dont match because I forgot to take them prior to the workout. I reset the Apple Watch afterwards and got the screens then.)

Notice something strange? There’s no appreciable difference between any of these workouts re: calories burned.

Notice something stranger? The resting calorie rate is identical across all genders.

I even did a fourth workout where I selected the “Skip” option, and it gave me the same results. This seems to indicate that Apple’s Activity tracker doesn’t take sex into account, which makes it impossible to ballpark even a reasonably accurate calorie count for users. More curiously, however, Apple does seem to imply that gender matters to some degree, as you can see that, when “Male” is selected, the daily calorie target for the lowest activity setting (which I used for all tests) is 460 calories compared to 380 for “Female” and “Other” (and “Skip”).

Whether in the interest of keeping things politically correct or simply because Apple’s research team has decided gender-based hormonal metabolism doesn’t impact calories burned, it seems that gender is not a part of whatever calculations are going on here.

Using industry-accepted equations for calories burned during exercise (there are several to choose from, but all account for gender), here are what I expected my results to be:

• Female:

  [(32 x 0.074) – (245 x 0.05741) + (101 x 0.4472) – 20.4022] x (40.09 / 4.184)

  = [2.368 – 14.06545 + 45.1672 – 20.4022] x 9.58173996

  = 13.06755 x 9.58173996

  = 125.209866

  = 125 Cal

• Male:

  [(32 x 0.2017) – (245 x 0.09036) + (100 x 0.6309) – 55.0969] x (40.06 / 4.184)

  = [6.4544 – 22.1382 + 63.09 – 55.0969] x 9.57456979

  = -7.6907 x 9.57456979

  = -73.6351439

  = -74 Cal

  Error. For these equations to work, there is a minimum threshold of heart rate that constitutes a functional workout.

• Other:

  N/A

As you can see, there are issues — or, rather, boundaries — with the traditionally accepted ways to calculate calories burned during exercise, and there’s even debate as to how to properly establish one’s Basal Metabolic Rate (BMR, which Apple seems to refer to as “Resting” calories). Whatever equation or assumptions Apple is using, though, they’re demonstrably wrong for one simple reason: Apple Watch’s cited BMR is identical for all three selected gender categories.

I was certainly not expecting that, and I dont see how a calorie calculator for a workout can be remotely accurate without accounting for the fundamental differences between male and female physiology. And even had Apple included those figures properly, the “Other” category couldn’t conceivably provide accurate data without a series of sub-questions to ascertain the exact hormonal and compositional body type of the individual in question.

Going into this, I thought that the “Other” selection would be misleading and unhelpful for many Apple Watch users. It turns out that the entire formula for calorie expenditure is borked. Sure, it’ll work for tracking trends in your exercising, but all the numbers will be significantly off.

Here’s hoping watchOS 2 fixes this problem.

Note: I could be completely wrong about all this. Maybe there’s something I missed or transposed incorrectly in my calculations. Maybe I used the wrong units, or maybe Apple’s hit on a revolutionary new and accurate way to count calories without needing to know an individual’s gender. I doubt that’s the case, but who knows? If you do, drop me a line or comment on Twitter so we can get to the bottom of this weirdness.