Tag Archives: Mifflin-St Jeor

Tracking Net Calories with LoseIt and Apple Watch

I use an Apple Watch to track my total calories burned and use LoseIt to track my total calories consumed. LoseIt can show you net calories – calories consumed minus calories burned – with a big caveat that took me a while to figure out:

A LoseIt Break-Even “Bonus” of Zero

LoseIt gives a “Calorie Bonus” (that is, actually subtracts calories burned) only if you exceed the number of calories they expect you to burn in a day. If you burn less than that many calories, there is no “Calorie Penalty” and you don’t have a reliable way to directly see your net caloric balance for the day.

I’ve determined that a “zero calorie” exercise balance (the maximum level burn that still gives a “bonus” of exactly zero) on LoseIt has nothing to do with your LoseIt weight loss goals, manual adjustments to your LoseIt caloric intake goals, or any goal settings on the Apple Watch. The “zero calorie” exercise balance is the Mifflin-St Jeor Equation for BMR times a seemingly arbitrary activity factor (overall multiplicative factor) of 1.4518. LoseIt’s FAQ says they account for “dressing, showering, housework…” (without mentioning the equation or 1.4518 multiplier) but this multiplier is clearly the number to which they feel the activities they list corresponds. This is the same as what LoseIt uses as the baseline for your caloric intact, before subtracting weight loss goals (e.g. 2lbs/week = 1,000 calories/day) and before subtracting any additional manual adjustments.

That is, the exercise bonus on LoseIt is entirely determined by your weight, height, age, and gender, and it is the same equation for the number of calories LoseIt suggests for maintaining current weight. This equation gets me exactly the LoseIt “zero-calorie bonus” (the break-even point) every day for the last 25 days (as far back as I bothered to check) except for two days when it’s one calorie off and another day where it’s two calories off (i.e., this is correct to within rounding error):

Lose-It Break-Even = 1.4518 * 10*pounds*0.45359237 + 6.25*inches*2.54 - 5*age + gender

gender in the Mifflin-St Jeor Equation is +5 for men and -161 for women.
0.45359237 is the conversion factor for pounds to kilograms
2.54 is the conversion factor for inches to centimeters

While LoseIt imports the Apple Watch data, LoseIt doesn’t care about what you’ve set as your Apple activity goal. LoseIt’s definition of a “normal” amount of activity (multiplication of BMR by 1.4518) is already baked into their formula. LoseIt just compares your total calorie burn to the formula above. I compared my Apple Activities App “Total Calories” daily history to LoseIt’s record of total calories burned and found a match within one calorie for all 25 days studied. I vaguely recall that on either LoseIt or MyFitnessPal, my profile included my to provide (via a drop-down) an estimate of my activity level. If that was LoseIt, then the words I chose (“minimal active” or something like that) might be where the 1.4518 came from. Or maybe the same activity factor is universally applied to all users.

Anyway, plotting my weight (over the last 25 days) against what LoseIt would have considered a break-even “bonus” of zero calories (that is, after a day ends, you can see by how much you came up short or how much extra you burned) gives a very nice line:

LoseIt Break-Even Calories as a function of Don’s Weight

Apple’s Non-Move Calories

While LoseIt treats non-active calories as a function of weight, age, gender, and height, it seems that the Apple Watch approach also includes sensor-based measurements when determining non-Move calories. You can even see the variation in non-Move (“Resting Energy”) calories in Apple’s iOS “Health” app:

iPhone screenshot of Apple Watch data

Indeed, there is no obvious correlation between daily Apple Resting Energy calories and weight, and there is at most a weak correlation between daily Apple Resting Energy calories and daily Move calories:

Conclusions

Because LoseIt is using a RMR formula that is a function only of the day’s weight (presuming height, age, and gender are fixed) and because Apple Watch is using biological measurements for both Move calories and for Resting Energy calories, there is no way to know the exact number of Move calories one will need in order to achieve a calorie bonus from LoseIt. Thus, there is no way to know the exact number of calories one needs to burn in order for LoseIt to be a reliable provider of net calories relative to one’s goal…. Except to make sure that one burns a margin-of-error more than what one will guess to be the break-even point. For me, I need roughly 500 active calories on top of the seemingly random “Resting Energy” I get each day to reach the LoseIt break even point. As a result of this analysis, which I initially performed in February 2019, I set the Move goal for my Apple Watch to be 550 calories per day (500 plus a 10% margin of error) and so far, every day that i’ve hit this Move goal, I’ve received at least a few Bonus Calories from LoseIt, and thereby have an accurate net caloric balance from LoseIt’s visualization.