Why Calories Don't Count

Chapter 1: Calories, Calories Everywhere

Overview

Amid endless food options and equally endless advice about them, the simplest question remains unresolved: what makes a calorie count? The difficulty is not finding numbers but deciding which ones are real, and that uncertainty shapes every meal.

San Francisco occupies barely seven miles north to south and seven east to west. Oracle Park, home of the Giants, sits a few blocks from the Bay Bridge. The author grew up in the area, attending St Ignatius High School and UC Berkeley, and remains a devoted 49ers fan.

American football's violence explains its short sixteen-game season, where every contest matters. Baseball runs roughly 180 games, each less crucial and lasting over three hours, so attention wanders to beer and food. What spectators eat in the stands therefore becomes the story.

Key Takeaways

• Football's violence makes its games scarce and therefore decisive, while baseball's marathon season drains individual games of consequence.
• A game that lasts over three hours across 180 contests inevitably turns spectating into an occasion for eating and drinking.
• Ballpark food emerges less as a sideline indulgence and more as the natural centerpiece of the baseball fan's experience.
• The Giants and their stadium matter in this account because the author's own San Francisco history is the lens through which the baseball/food dynamic is seen.

Introduction

Overview

Calorie counts promise to rank foods from good to bad, yet a 1,400-calorie hot dog and a 10,000-calorie burger make a nonsense of that. The numbers are supposed to guide choice, but they leave a deeper question unanswered: what do they actually measure, and does that even matter?

Ballpark food

Ballpark fare moved from hot dogs and burgers to nachos in the 1980s and 90s. At PNC Park in Pittsburgh, the Brunch Burger is a beef burger with bacon, fried egg and cheddar between a doughnut with sprinkles, 900 calories. At the Great American Ballpark in Cincinnati, the Meat-lover's Hotdog is a deep-fried bacon-wrapped sausage with beef chilli, cheese and fried salami, 1,400 calories. Does a lower number mean better, and are all calories equal?

What a calorie is

A calorie is a unit of energy: the small-c version heats 1 ml of water by 1°C. Food labels use the capital-C Calorie, the energy to heat 1 litre, called a kilocalorie or kcal; a Mars bar label says 228 kcal. A joule is the energy to accelerate 1 kg at 1 m/s² over one metre. One calorie equals 4.184 joules. In ordinary speech “calorie” means kcal, and this book follows that usage.

Nutrition labels

Nutrition labelling is compulsory on nearly all pre-packed food in North America, the UK, Europe and Australasia, including packaged meat but not the butcher counter. UK and EU labels must state energy in kJ and kcal, fat, saturated fat, carbohydrates, sugars, protein and salt. Front-of-pack traffic lights are voluntary, with red over 25% of reference intake per 100g or 30% per portion. US labels list sodium rather than salt, since 1g salt contains 0.4g sodium, require trans fats, which raise heart-disease risk, and separate added sugars from naturally occurring ones.

Serving sizes and eating out

Serving sizes distort the numbers: 30g of cereal and 75g of pasta are less than most people eat. US FDA rules have required chains with twenty or more locations to provide calorie information since May 2018, including the note that 2000 calories a day is general advice. UK menu labelling is voluntary. In January 2019, Deliveroo got 500 restaurants to list calories and urged all 17,000 on its platform to do the same; a poll of 2,200 British adults found 54% wanted more. Restaurants argue independents can't easily comply; measuring calories is difficult.

Lavoisier's discoveries

Burning wood in a sealed container leaves total mass unchanged; ash only seems to lose weight because gases escape. Lavoisier vaporised a diamond with sunlight focused through a huge magnifying glass and found it weighed the same as a gas, and when charcoal burned to the same gas, he identified both as forms of one element, carbon. This defined the conservation of mass: matter is neither created nor destroyed in any operation. He named oxygen and hydrogen, proved water is not an element, and showed that combustion requires oxygen: phosphorus and sulphur gain weight when burnt because they combine with air. After Joseph Priestley described the gas from heated mercury oxide, Lavoisier concluded air is 21% breathable oxygen and 79% unbreathable, mostly nitrogen.

The first calorimeter

Lavoisier's further leap: respiration is combustion. He built the proof on latent heat, Joseph Black's discovery that melting ice absorbs heat without rising above 0°C. With Pierre-Simon Laplace he made a calorimeter holding a guinea pig in an ice-filled, snow-insulated chamber with air piped through. Melt-water was the measure, 80 kcal per kilogram; in ten hours the guinea pig melted 0.37 kg, about 29.6 kcal, matching the heat from burning carbon to the same carbon dioxide it exhaled. So respiration is slow combustion, and the instrument's name joins Latin for heat with Greek for measure.

The bomb calorimeter

Lavoisier never used the word calorie; Nicolas Clément first defined the big-C Calorie, the kilocalorie, in 1824. The bomb calorimeter, developed by Paul Vieille in 1878 for measuring explosive heat, was soon applied to food. Platinum wires ignite a sample inside a vessel pressurised with pure oxygen at about thirty times atmospheric pressure; the heat warms a water jacket, and the temperature rise gives the count. Soups are dried first, as water carries no calories. According to the book, the bomb answers only the first of three distinct numbers:

• A: the calories actually in the food
• B: the number printed on the pack
• C: the calories the body can actually extract

Getting from A to B requires Wilbur O. Atwater.

Key Takeaways

• A food label's "Calorie" is a kilocalorie, the energy needed to warm a litre of water by one degree, so everyday calorie counts are a thousand times larger than the physicist's calorie.
• Serving sizes written on packaging and menus are often far smaller than typical portions, and many restaurant chains still list calories voluntarily or not at all, so the printed number is a weak guide to what you actually eat.
• Lavoisier's calorimetry proved that animal respiration is a slow form of combustion, generating the same heat as burning the carbon that the animal exhales as carbon dioxide.
• The calorie count from a bomb calorimeter is only the first of three different numbers: the energy actually in the food, the figure printed on the packet, and the energy your body can extract, and only the last one matters for weight.
• Asking whether a 1,400-calorie hotdog is worse than a 900-calorie burger presumes all calories are interchangeable, a presumption the rest of the book puts in question.

Chapter 2: The Atwater Factor

Overview

The calorie count printed on food describes what burns in a machine, not what happens in a body. If a calorie is not a calorie, then every label, diet plan and nutrition guideline rests on a shaky foundation.

Friday nights in Bottisham end with a pint at The Bell pub, where a pack of pork scratchings declared 440 calories, almost all from 32.6 grams of fat and 34.2 grams of protein. Such labels do not reflect what the body actually extracts.

Rubner, the bomb and caloric availability

The bomb calorimeter, invented in France in 1878, was developed for nutritional science by Germans studying farm animals, not humans. Max Rubner, trained under Carl Voit, adapted it for food energy and built the first self-recording whole-body calorimeter. A dog trial balanced intake against heat and waste, proving the first law of thermodynamics applies to life. Rubner's isodynamic law made protein, fat and carbohydrate calories interchangeable: "a calorie is a calorie."

The bomb burns food in pure oxygen at thirty times sea-level pressure, extracting every calorie. Digestion is gentler, so food structure and processing change availability: sugar yields over 95 per cent of its energy, sweetcorn far less, and tortillas or cornbread far more than raw kernels.

Atwater's method

Wilbur Olin Atwater (1844-1907), a Connecticut chemist, imported the German Agricultural Experiment Station model and ran the Storrs station. His method had three stages:

1. Collate typical foods and their components.
2. Burn each in a bomb calorimeter for gross energy.
3. Feed each to humans, analyse the waste, and subtract.

Atwater treated food as five components: water, ash (minerals), protein, fat and carbohydrates. He named the extracted value metabolisable energy.

The numbers

Animal fat burned at 9.4 calories per gram, plant fat at 9.3. Faecal losses ran 5 per cent for animal fat and 10 per cent for plant fat. Weighting for the animal-dominated diet of the era, Atwater settled on 8.9, rounded to 9 calories per gram.

Total carbohydrate was calculated by difference. Animal carbs, chiefly lactose and glycogen, burned at 3.9 calories per gram and plant carbs at 4.15. With 95 per cent of intake plant-based, Atwater settled on 4 calories per gram.

Protein, the most complex, contains nitrogen. Unused protein converts to fat. The nitrogen leaves as urea, uric acid and creatinine, feeding the nitrogen cycle.

Protein and nitrogen

Protein content was inferred from nitrogen, even though the body holds an estimated 80,000 to 400,000 distinct proteins. Atwater assumed meat averaged 16 per cent nitrogen, giving 6.25 grams of protein per gram of nitrogen. Urinary losses ran about 1.25 calories per gram of protein eaten, and faecal losses 3 per cent for animal protein, 15 per cent for plant. Since protein burned at 5.65 calories per gram, and 61 per cent of the average family's protein came from animals, Atwater settled on 4 calories per gram.

The Atwater general factors

His system set fat at 9, carbohydrate and protein at 4, and alcohol at 7 calories per gram. Rubner's rival figures of 9.3, 4.1 and 4.1 lost out because he ignored faecal losses and assumed nitrogen at 15.5 per cent, a 6.45 factor that overestimated protein. Atwater's general factor system stuck.

The human body as a machine

Atwater extended Rubner's dog work to humans, building the Atwater-Rosa calorimeter, then testing four men on varied diets and exercise from 1898 to 1900. Intake matched output as heat and waste, so the first law of thermodynamics and the isodynamic law both held in people, and mental effort burned no measurable calories. Alcohol's 7 calories per gram fuelled work identically, to the liquor trade's delight and his Methodist-backed employer's displeasure.

Lulu Hunt Peters

Lulu Hunt Peters, a California doctor, weighed 220 pounds after medical school and lost 70 by strict calorie counting. Her 1918 Diet and Health with Key to the Calories, spun from a syndicated column, was the first popular diet book. It reached middle-class American women as wartime metal shortages ended corset use. Her rules:

• Ideal weight: multiply inches over five feet by 5½, add 110.
• Maintenance: 15 to 20 calories per pound of ideal weight.
• Weight loss: cut 500 to 1000 calories a day, using 100-calorie portions from Atwater's tables.

She sold slimming as patriotic rationing. The book sold two million copies, became the first diet bestseller, and turned the calorie into a weapon.

What the factors missed

Over a century on, most food-label calories still rest on Atwater's factors, despite his protein arithmetic's approximations and his dismissal of Rubner's data.

Jones factors

The first flaw in Atwater's numbers sits in his protein estimate. He derived total protein from nitrogen content using a 6.25 conversion factor, based on meat's average nitrogen share of 16 per cent. But protein nitrogen actually ranges from 13 to 19 per cent, which implies conversion factors anywhere from 5.26 to 7.69, and materially different calorie counts. In 1931, D.B. Jones proposed abandoning the universal 6.25 and replacing it with food-specific factors. The resulting Jones factors have been widely adopted, though for the most common protein sources the spread is narrow: meat keeps 6.25, cereals and legumes sit between 5.7 and 6.25, and dairy sits at 6.38.

Carbohydrate by difference

Atwater's second weakness was computing carbohydrate by subtracting everything else from total weight and calling the remainder carbohydrate. That lumps simple sugars, starch and indigestible fibre together. To correct it, a factor of 3.75 calories per gram for simple sugar arrived in 1970, and 2 calories per gram for dietary fibre in 1998. Combined with Atwater's original factors, these additions form the extensive general factor system.

The specific factor system

Because the proportions of macronutrients within a food shift their individual heats of combustion, a third system, really a series of tables, was created. Under it, fuel values vary noticeably by food:

• Eggs: 9.02 fat, 3.68 carbohydrate, 4.36 protein
• Meat and fish: 9.02 fat, 4.27 protein
• Dairy: 8.79 fat, 3.87 carbohydrate, 4.27 protein
• White rice: 8.37 fat, 4.16 carbohydrate, 3.82 protein
• Potatoes: 8.37 fat, 4.03 carbohydrate, 2.78 protein
• Soybeans: 8.37 fat, 4.07 carbohydrate, 3.47 protein

Rice protein, for instance, burns about 20 per cent hotter than potato protein, and digestibility differences such as wholewheat versus finely milled flour change available energy further. A selected list appears in the book's appendix.

What the label still says

The specific system is more accurate but unwieldy, so Atwater's original 9-4-4 general factors still dominate the calorie counts on packaged food. Yet none of these systems, Atwater's or Rubner's, accounts for energy lost as heat while macronutrients are chemically dismantled to drive metabolism. That missing step, the gap between labelled calories and usable calories, requires understanding how food becomes energy in the first place.

Key Takeaways

• Label calories come from bomb calorimetry, which burns food completely, so the stated figure overstates what digestion actually extracts; food structure and processing can shift that availability significantly.
• Atwater’s 9-4-4 factors were a compromise tuned to late-19th-century American diets, and his protein estimate rested on a nitrogen-to-protein ratio that does not hold across food types.
• Later corrections, from food-specific nitrogen factors to separate values for sugar and fibre, make the system more accurate, but none captures the energy lost as heat when the body dismantles macronutrients.
• Lulu Hunt Peters’s 1918 bestseller turned Atwater’s tables into 100-calorie portions, making the calorie a portable tool for weight control and launching the modern diet-book market.

Chapter 3: How Do We Turn Food into Energy?

Overview

A day’s food contains enough energy to bring many litres of water to a boil, yet the body that stores it never catches fire. The puzzle is not where the energy goes, but how it is tamed into something a living cell can actually use. What allows such violent potential to be released safely at body temperature?

Boiling a litre of tap water takes just 85 calories, while a day's 2,000 to 2,500 calories could boil 25 litres. We don't boil because, as Rubner and Atwater showed, living creatures obey the first law of thermodynamics: food energy cannot be destroyed, only converted. The body converts it into transportable units in two steps:

• digestion, from mouth to bloodstream
• intermediary metabolism, deciding whether they are burned or stored

From mouth to stomach

Digestion is mechanical and chemical: chewing grinds while salivary amylase attacks starch, forming a bolus. In the stomach, at pH 1.5, hydrochloric acid kills bacteria and activates pepsin, fragmenting protein; peristalsis churns the meal into chyme, and the pyloric sphincter releases it.

The small intestine

The small intestine, six to seven metres long, runs duodenum to jejunum to ileum; its villi give a thirty-square-metre absorbing surface, one cell thick. Bicarbonate neutralises the acidic chyme; bile from the liver and gallbladder emulsifies fat into micelles, and pancreatic enzymes finish the chemical breakdown. Glucose and amino acids absorb here; fat, which cannot dissolve in blood, is reassembled into triglycerides and carried in chylomicrons.

Lactose

According to the book, most humans lose lactase in adulthood when a protein switches off the gene, making lactose intolerance the mammalian default. A mutation blocking that switch spread with European dairy herds 7,500 years ago; 85 per cent of Northern Europeans drink milk.

The large intestine

The large intestine produces no digestive enzymes; most nutrients and 90 per cent of water are already absorbed, and the leftover fibre and bacteria solidify into faeces.

Lactose intolerance

According to the book, undigested lactose in the colon limits water absorption and ferments into gas, causing bloating, pain, diarrhoea and nausea thirty minutes to two hours, with severity dose-dependent.

ATP, the energy currency

Glucose, fatty acids and amino acids are transportable intermediates, not energy. The universal currency is adenosine triphosphate (ATP), a molecular rechargeable battery: energy sits in the bond between the second and third phosphates, and splitting off the third releases a burst, returning it to ADP. People cycle 50 to 75 kg of ATP daily, each molecule recycled 500 to 750 times.

The mitochondrial pathway

Metabolism, as Lavoisier's calorimetry showed, is carbon burning in oxygen made gentle, with energy parceled into ATP rather than heat. All nutrients funnel into acetyl-CoA, their common midpoint. Mitochondria, in all cells except red blood cells, run beta-oxidation, the Krebs cycle and oxidative phosphorylation.

Krebs discovered his cycle in 1937 (Nobel, 1953): acetyl-CoA joins oxaloacetate to form citrate, shedding two carbons as CO2 each turn while loading three NADH and one FADH2 with electrons.

Electrons travel through complexes I to IV on the inner membrane, pumping protons to build an electrical gradient like pumped hydro storage. Protons then spin ATP synthase, a turbine producing three ATP per turn; oxygen receives the electrons, forming water.

Carbohydrate metabolism

Carbohydrates are digested to monosaccharides, chiefly glucose, which yields 30 ATP per molecule. Glycolysis splits glucose anaerobically into two pyruvate molecules, spending 2 ATP and earning 4 (net two), plus 2 NADH. Without oxygen, pyruvate becomes lactate, recycling NADH for short bursts like sprinting.

Oxygen and the full yield

Glycolysis powers short, intense bursts; its lactate becomes lactic acid, the burning muscle sensation. Lactic acid clears slowly; past a minute oxygen takes over. Oxygen converts pyruvate to acetyl-CoA, entering the Krebs cycle and oxidative phosphorylation, for 28 ATP. Each glucose yields thirty ATP, fifteen times glycolysis's net two.

Gluconeogenesis and the Cori cycle

The liver and kidneys can rebuild pyruvate into glucose, gluconeogenesis, costing four ATP and two NADH. The liver's central job is holding blood glucose between 4 and 7 millimolar: insulin clears excess, gluconeogenesis restores deficit. Muscle lactate travels to the liver and returns as glucose: the Cori cycle. Once pyruvate becomes acetyl-CoA, the path back to glucose closes.

Fat, protein and storage

Beta-oxidation removes two carbons at a time from fatty acids; complete oxidation of palmitic acid yields 106 ATP. Protein sheds its amino group as urea; its carbon skeleton then feeds the Krebs cycle or forms ketones. Glycogen stores cover a day; fat stores of 10-20 kg hold 90,000-180,000 calories, enough for 72-90 days. Amino acids cannot be stored, so all surplus fuel becomes fat.

Starvation, ketones and type 1 diabetes

In starvation, muscle burns fat to spare glucose for the brain, which cannot use fatty acids. After forty hours, nearly all liver glucose comes from gluconeogenesis. With oxaloacetate diverted there, acetyl-CoA accumulates and liver converts it to ketones for muscles, heart and brain. Type 1 diabetes, according to the book, destroys insulin production, so glucose cannot enter muscle or fat and protein breaks down unchecked: 'starvation in the midst of plenty.'

Uncontrolled diabetes also brings diabetic ketoacidosis. When no glucose reaches the cells, fatty acids and amino acids break down rapidly, and the bloodstream is flooded with ketone bodies. The danger is not the ketones themselves but their acidity. According to the book, blood pH normally holds between 7.35 and 7.45 on a scale where 7 is neutral, making blood slightly alkaline; a sudden load of ketones pushes it acidic. Drowsiness begins at pH 7.0 to 7.25, coma sets in below 7.0, and without rapid correction death follows.

Before 1922, type 1 diabetes was invariably fatal. Hospitals ran wards where diabetics, many of them children, wasted away as though starving; food passed straight through their bodies, and most died comatose. That year, Banting and Best, working in J.J.R. Macleod's Toronto lab, discovered insulin. The historian Michael Bliss called watching the first starved, comatose patients revived by it one of the genuine miracles of modern medicine. Banting and Macleod received the 1923 Nobel Prize in Physiology or Medicine.

What the energy is for

Digestion breaks food into its constituents, and the body painstakingly uses those intermediates to build ATP. The final term in the calorie equation, usable calories, still needs defining, because extracting energy from food is not the same as being able to use it. That depends on what the body actually spends energy on, which is the next topic.

Key Takeaways

• Food energy obeys the first law of thermodynamics, so a day's calories are never lost; digestion and intermediary metabolism simply convert them, ending in ATP, the phosphate bond all cells can spend.
• Digested glucose, fatty acids and amino acids are not yet usable energy; they converge on acetyl-CoA and mitochondrial respiration, where a proton gradient drives ATP synthase and oxygen receives the electrons.
• Anaerobic glycolysis buys quick speed at a cost: it nets only 2 ATP per glucose and leaves lactate, but the liver recycles lactate into glucose, while full aerobic oxidation of glucose returns 30 ATP, roughly fifteen times as much.
• Fat is the largest fuel reservoir, with palmitic acid yielding 106 ATP, while amino acids cannot be stored; once glycogen runs out, the body spares glucose for the brain by burning fat and making ketones.
• Herman Pontzer notes that before 1922 type 1 diabetes was invariably fatal; in uncontrolled diabetes the body starves even when well fed, and ketoacidosis kills through acidity, with coma setting in as blood pH drops below 7.0.