You Can't Cheat Thermodynamics — And It's Exactly Why Counting Calories Fails

Article

Calories are real. “Calories in, calories out” is still a broken measuring tool — because your body was never a bomb calorimeter. “Calories in, calories out.” It sounds like physics, and half of it is. You genuinely cannot cheat the conservation of energy — anyone who tells you calories don't matter is wrong. But “count your calories” quietly smuggles in two assumptions that are both false: that you can measure the calories going in, and that the calories going out are a fixed number you control. Neither holds. The law is real. The ruler is broken. First, the part that's true Let me be the last person to tell you energy balance is a myth. The first law of thermodynamics isn't negotiable: absorb more energy than you expend and the surplus is stored — full stop. So this isn't a “calories don't count” argument; those always end badly. It's a more uncomfortable claim: the two numbers you'd actually need to count are, in a living body, unmeasurable and constantly moving. “Calories in” isn't what you eat — it's what you absorb The calorie printed on a label comes from a bomb calorimeter: literally setting food on fire in a sealed chamber and measuring the heat. That number assumes your body extracts every last joule. It doesn't. Your gut is a leaky, selective, microbe-filled system, not a furnace. When USDA researchers actually measured it, whole almonds delivered about 20% fewer calories than the label — 129 instead of 160 per serving — because the intact cell walls and fibre keep some of the fat from ever being absorbed. Cooking, processing, fibre, and your own microbiome all shift the real number. The “calories in” you're counting is a laboratory abstraction, not what your body took. The calorie on the label was measured by setting food on fire. Your body is not a furnace. “Calories out” isn't a fixed number — it's a thermostat, not a dial But the first law isn't the only law of thermodynamics — there's a second one the calorie math quietly ignores. It's every bit as unbreakable: physics, not opinion. And it's exactly where "calories out" falls apart. The second law says no engine converts fuel into useful work with perfect efficiency — some energy is always lost as heat, and how much is lost is never fixed. The human body is exactly such an engine. It never captures 100% of a food's energy as something it can use or store; a portion is always dissipated as heat — and, critically, that fraction changes with what a person eats and the metabolic state they're in. Digesting protein, for example, costs the body 20–30% of protein's own energy just to process it — a large efficiency loss — while fat costs almost nothing. So 100 calories of protein and 100 calories of fat leave very different amounts of net energy behind, even though a calorie counter records them as identical. Just as no two engines burn fuel at the same efficiency, no two foods — and no two people — extract and store energy the same way. And it gets worse for the ledger: total “calories out” adapts. When “The Biggest Loser” contestants lost dramatic weight, their resting metabolism didn't just fall with their size — it stayed roughly 500 calories a day below expected, and was still suppressed six years later, even as the weight came back. Eat less, and the engine quietly runs leaner. “Calories out” is not a dial you set; it's a thermostat that adjusts. None of this breaks physics — energy is still perfectly conserved, and the first law holds. The point is subtler, and more damning for calorie-counting: because the body's efficiency varies and cannot be measured from the outside, “calories out” simply is not a fixed quantity anyone can count. The second law guarantees the losses; it just won't let you write them down in advance. And the same calorie sends different signals There's a final layer the arithmetic ignores entirely: calories carry hormonal information. Two hundred calories of soda and two hundred calories of eggs are identical on a spreadsheet but opposite in your body. The soda spikes insulin, delivers no satiety, and leaves you hungry an hour later; the eggs are protein-dense, blunt hunger, and don't make the same demand of your fat-storage machinery. Same “calories in” — completely different effect on how much you eat next and how readily you store what you just ate. That's the part that gets me as an engineer: a single number is treated as the whole story when it's barely the beginning of one. (It's also why, at Aleré, we pay attention to a meal's hormonal footprint — its effect on insulin and satiety — not just its calorie total.) It fails in practice, too If the science were the only problem, calorie counting might still be a useful discipline. But it fails on its own terms. In a randomized controlled trial, researchers gave overweight primary-care patients the most popular calorie-tracking app for six months and compared them with patients who got usual care. The result: no meaningful difference in weight. Simply handing people the app changed essentially nothing. And most people don't last six months anyway. The majority abandon food-logging apps within the first few weeks, and the number-one reason isn't cost — it's effort. One 2021 study found that people who could log a meal in seconds were more than three times as likely to still be tracking half a year later as those for whom it took a couple of minutes. Friction, not willpower, decides who keeps counting. So calorie counting fails twice over: the number you're counting is wrong, and the counting itself is so tedious that almost no one sustains it long enough for it to matter — and the few who do don't lose more weight than the people who never bothered. So what do you actually do? Not “ignore calories.” Stop counting the uncountable, and work the levers that actually move the system: prioritize protein and fibre (more satiety, higher thermic cost, gentler hormonal response); lean on whole-food matrices over processed calories your body absorbs too efficiently; move after meals; protect your sleep — and pay attention to how your own body responds, which is information a label can't give you. The goal isn't a perfect ledger. It's a metabolism that isn't fighting you. The honest version It isn't “calories don't matter.” It's that “count your calories” is precise-sounding advice built on an imprecise number and a moving target. You can't cheat thermodynamics — but if you've been dutifully counting, here's the twist: you were never really counting it in the first place. The tools that finally move the needle won't ask you to count at all. They'll be effortless enough that people actually keep using them, and smart enough to read how your body responds instead of tallying a label off a package. That's a very different bet from a calorie counter — and, I'd argue, the only one with a real chance of lasting. If you've tried calorie counting, what actually worked for you — and what quietly stopped working? I'd genuinely like to hear. I'm an engineer and applied physicist, not a physician or dietitian. This is education, not medical or nutritional advice — decisions about your diet belong with you and your clinician. References Novotny, Gebauer & Baer, Am J Clin Nutr 2012 (almond calorie absorption). Fothergill et al., Obesity 2016 (“Biggest Loser” metabolic adaptation). Laing et al., Ann Intern Med 2014 (MyFitnessPal randomized trial — no significant weight loss vs. usual care). Food-logging retention: 2021 study in Appetite; industry app-retention data. Thermic effect of food: established human-nutrition physiology.