December 8, 2013

Calories: True as a concept, futile as a law

This is a guest post from the good folks at, who do a great job providing objective research-based info on pretty much all areas of nutrition and supplementation. It’s an important topic that all-too-often is misunderstood by the general public. Thanks to Sol Orwell and his team for explaining the concept in an easy-to-understand manner and exploring its practical applications. For those interested in the technical aspects, I would recommend you read the paper by Thomas et al. A Mathematical Model of Weight Change with Adaptation.

Calories in versus calories out is one of the fundamental ‘laws’ of nutrition. Though it upholds both the law of thermodynamics and the conservation of mass, the calories in versus calories out theory doesn’t hold up when applied to the human body. If anything, it is a guiding statement based on a law.

When looking at a true closed system (humans don’t count), any possible conversions of energy can be predicted down to a single joule, with nearly 100% accuracy. This is because our equations for energy transfers in closed systems are very good.

When looking at an open system, like the human body, we also have a set of equations we can rely on. Unfortunately, they’re not quite as good as their closed-system counterparts. As an example, lets look at the most commonly used equation to determine caloric expenditure and your basal metabolic rate.

The Harris-Benedict formula for men:

88.362 + (13.397 x bodyweight in kg) + (4.799 x height in cm) – (5.677 x age in years)

The Harris-Benedict formula for women:

447.593 + (9.247 x bodyweight in kg) + (3.098 x height in cm) – (4.330 x age in years)

Don’t forget to multiply by the ‘activity’ factors. Use 1.2 if you are sedentary, 1.375 for light exercise, 1.55 for moderate, 1.725 for heavy, and 1.9 for intense exercise.

The above calculations are pretty good, since the parameters being measured (gender, weight, height, age, and activity level) account for a lot of variation in the metabolic rate between individuals.

Still, it’s not an absolute rule. There are many other factors that influence the metabolic rate that are not included in the Harris-Benedict equation. These include:

• Adipokine status, including leptin, adiponectin, and resistin.
• Thyroid hormone status
• Steroid hormone status, including estrogens and androgens, as well as DHEA.
• Mitochondrial efficiency and ‘uncoupling’, also known as thermogenesis.

The formula doesn’t include these factors because it is impractical to expect people filling out a BMR equation to know their mitochondrial efficiency.

Moreover, the activity levels are vague. What feels ‘intense’ to one person could be ‘moderate’ to another. A misinterpretation of activity levels for someone with a 2000kcal intake could result in a difference of up to 700kcal.

Calories in, as a concept, is fundamentally hard to predict. Differences in food absorption, whether due to genetics, nutrient co-ingestion or even the shape of the chyme in your stomach are hard to account for. The differences in nutrient partitioning after absorption, or whether the nutrients go to muscle, glycogen or fat are even harder to predict.

Don’t forget that calorie counts are not absolute. Though a food label may list “70kcal” as the caloric value for one serving size, a more accurate description is “70kcal, give or take a 5% margin of error.”

Does this mean I shouldn’t worry about calories?

No, no, and no. Despite the above complications, calories do matter in concept, and having even a rough grasp on caloric intake is invaluable. The only change that needs to happen is how you view your caloric intake.

Instead of deciding to eat 2,357 calories a day to maintain weight, eat “a little above 2,000 calories.” Rather than deciding to cut 500 calories from your daily diet to lose a pound of fat per week, aim to eat approximately 500 less calories per day, and track how your body responds.

At any point during the day, you should be able to accurately estimate your caloric intake, as well as how many calories you have left to eat that day. Instead of expecting a perfect caloric count, aim for +/- 200 calories of the true value.

A relaxed but consistent style of nutrient tracking means you’ll be hitting your goals and stressing less about it. Best of all, you’ll never feel constricted because your favorite food would put you over your limit by 50 calories. Just remember: being relaxed doesn’t mean you’re not determined.

Written by Kurtis Frank and Dr. Spencer Nadolsky. Both are directors at, an independent and un-biased organization that focuses on evidence for supplementation and nutrition.


  1. Nothing in the above refutes calories in vs. calories out/thermodynamics as an inalterable law. EVERY example just points out the the numbers can vary. They still have to add up and balance. That things affect the numbers only means that they affect the numbers.

    Not that the law is wrong.

    This article is stupid as hell.

    Comment by Lyle Mcdonald — December 8, 2013 @ 11:47 am

  2. Ohhh man that link to “A Mathematical Model of Weight Change with Adaptation” is brilliant! I’m just in the midst of trying to write a blog post on appetite/overfeeding and it’s jam packed full of gems.

    Thank you Brad 🙂

    Comment by Shane — December 8, 2013 @ 11:55 am

  3. Thanks for the input Lyle. Perhaps it’s semantics, but I don’t interpret the article as stating the law is wrong; rather, as you note it implies that the numbers can vary based on a variety of factors. The paper by Thomas et al. that I posted does a nice job in summarizing the complexities IMO.


    Comment by Brad — December 8, 2013 @ 12:25 pm

  4. Lyle, he isn’t setting out to disprove the law of thermodynamics… he’s merely stating that the law doesn’t tell us enough to make calorie counting accurate when applied to human metabolism. We can’t even accurately calculate calories in… so how can we hope to calculate calories out with any kind of accuracy?

    The equation also doesn’t tell us where that caloric excess goes. Most people automatically assume excess goes straight to fat cells, but that is simply not the case.

    Nice strawman nonetheless…

    Comment by DJ — December 8, 2013 @ 2:41 pm

  5. Pretty bad article. I’m sorry but first of all the calories in vs. calories out is the best model we have currently. Second of all, if my trainer tells me “eat a little above 2,000 calories”, instead of giving me an exact number, I’ll tell him to go find another job.
    Oh, and once you’ve adjusted your BMR and TDEE with the help of tracking weight and calorie intake for a certain period of time, you pretty much have enough accuracy to claim you know exactly how much calories you need.

    P.S. Yes, numbers wary but that doesn’t mean the law is futile.

    Comment by Radi — December 9, 2013 @ 5:25 am

  6. This articles argues that, due to hard to measure individual factors, it is difficult to predict the weight loss or gain of an average person in a population based on how many calories the person consumed. But that doesn’t mean that an individual couldn’t predict his or her own weight loss or gain based on how many calories he or she consumed with a fair amount of accuracy. That is assuming, of course, that a given individual’s hard to measure factors, such as mitochondrial efficiency or thyroid hormone status, which vary widely from person to person, don’t vary that much over time within the same person.

    Comment by Dale — December 9, 2013 @ 1:36 pm

  7. Hah, as soon as I read the title of this article I recalled John Kiefer’s article on calories-in/calories-out and the war that followed between him and Lyle McDonald. So when I was reading this article here I wondered what Lyle would have to say about that. And BAM, first comment, not without his no-BS approach.
    I think that what this article fails to show is the approximate importance of this additional energy expenditure variables that are hard to measure. I understand that they make counting calories with single-digit precision void without denying the general rule. But some readers may begin thinking that human body is some kind of magical box that works in unforeseeable ways. Many people actually think this way and reading this article inattentively may enforce such opinions within them. And that’s what leads people for blaming their genes for being fat or embracing some strangest diets.

    Comment by Mike Mela — December 13, 2013 @ 4:18 pm

  8. >>>When looking at a true closed system (humans don’t count), any possible conversions of energy can be predicted down to a single joule, with nearly 100% accuracy. This is because our equations for energy transfers in closed systems are very good.

    When looking at an open system, like the human body, we also have a set of equations we can rely on. Unfortunately, they’re not quite as good as their closed-system counterparts.<<<

    There's no such thing as a closed system if you really want to "go there". But humans are actually much better approximations and biochemical reactions (enzyme complexes etc.) far better regulated than some so-called closed systems. Take a car engine — there will be some uncombusted fuel that goes out in the exhaust. A long chain fatty acid that gets "in" to the human body must eventually be oxidized to get "out".

    The equations for our body are actually better viewed in this context.

    Comment by Evelyn aka CarbSane — December 22, 2013 @ 10:50 am

  9. […] tutkimuksissa usein energiankulutuskomponentti on unohdettu ja se on hankala arvioida, koska niin moni asia vaikuttaa. Tämän perusteella voisi sanoa, että ongelma on siis enemmänkin laskijassa kuin kaloriteoriassa […]

    Pingback by Pahimmat ravintomyytit – osa I – kalori on tai ei ole kalori | Lihastohtori — February 14, 2014 @ 2:53 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.