Sleep.
Same 14-day deficit, run twice in a randomised crossover.
Once with 8.5 hours of sleep opportunity, once with 5.5.
Short sleep cut fat weight loss by 55%, from 1.4 kg to 0.6 kg.
It increased fat-free-mass loss by 60%.
Hunger rose, ghrelin rose, and substrate use shifted away from fat.
(Nedeltcheva et al., 2010, Annals of Internal Medicine; R§44)
Identical food. Different tissue leaving your body.
Ten people.
So here's the second trial: 36 people, eight weeks, about an hour less sleep on five nights.
Same total loss, smaller share of it as fat.
Same direction, smaller effect. (Wang et al., 2018; R§44)
Short sleep tilts a fat-loss protocol toward muscle loss at the same intake. It is the cheapest variable in this whole subject to fix, and it is the one most often traded away.
Training.
Almost everything sold to you lowers the input.
Training changes what the deficit takes.
I want to be precise about how big this is, because it has been measured.
Here is what the trials say.
Across 34 randomised trials and 1,455 people restricting calories, exercise prevented about 46% of the lean loss, roughly 0.9 kg.
Strength training alone prevented about 0.8 kg of lean loss. (R§42)
In 140 women on 3,350 kJ (800 kcal) a day, the ones who did not train also cut their everyday movement, the fidgeting and walking that nobody counts.
The ones who lifted or did cardio held it. (R§37b)
And training raises the rate at which you can oxidise fat: a meta-analysis of 13 studies and 519 overweight and obese adults found both interval and steady-state training raised maximal fat oxidation against no training. (R§34)
Fair caveat: one trained-versus-untrained study in women found a higher rate across the working range but no higher peak.
None of that is a ceiling going up.
It's the deficit being paid from the right account.
And it's a lever people think exists, but it doesn't.
You'll hear that certain drugs work as a "metabolic furnace", that glucagon raises your energy expenditure and takes the fat off. (which I initially believed)
Glucagon does raise energy expenditure by about 15%, which sounds enormous.
Three problems.
It doesn't work through brown fat: no temperature rise, no brown-fat activation on scanning.
In one study, about 29% of the rise came from oxidising protein, which is muscle, not fat.
And over 72 hours in people with overweight or obesity, the actual population this matters for, glucagon infusion changed sleeping, basal and 24-hour energy expenditure by nothing at all.
(Salem et al., 2016; Nair, 1987; Whytock et al., 2021; R§52)
Over thirteen hours in lean men, the effect is 210 to 290 kJ (50 to 70 kcal) a day, which the researchers themselves called of marginal clinical significance.
And one more thing about protein.
High protein protects lean mass in a deficit.
That's well established, and I stand by it.
But there's a limit nobody mentions.
In a ten-day military exercise producing a deficit of about 18,000 kJ (4,300 kcal) a day, 19 men ate 2 g/kg of protein, and 19 ate 1 g/kg.
Body mass, fat mass, bench press, leg press, and jump height: no difference between the groups. Cortisol actually rose more in the high-protein group.
(Øfsteng et al., 2020; R§11)
Protein is insurance that works at moderate deficits.
Push the deficit far enough, and the insurance stops paying out. It is not a licence to cut harder because your protein is high.
And one thing for the people who cannot face the food. (Me)
Four meat feeds a day is a lot of chewing, and some people have an aversion to it that no argument fixes.
Your protein floor is grams a day; it does not say where the grams come from.
Whey isolate, a beef protein isolate with a full essential-amino-acid profile, or free essential amino acids all count.
A 35 g meat feed is about 970 kJ (232 kcal);
the whey feed that does the same job is about 560 kJ (134 kcal);
a leucine-enriched amino-acid dose is about 200 kJ (48 kcal), and it lands in ten minutes to two hours instead of one to seven.
That changes the energy you eat, not the protein you need, and not your floor.
Three things meat gave you without asking, which you now add on purpose: the fat, ten grams a block; creatine, B12, iron, zinc and iodine; and the leucine, 2 to 3 g a feed, which some beef isolates do not have because they are mostly collagen. (R§70b)
What's your current split between "eat less" and "train so the body fat loss comes from the right place"?