How long should a cut last?

Twelve weeks of restriction is a sensible block, and it takes about fifteen weeks of calendar if you take a week at maintenance in the middle. If your target needs more than that, run one block, spend six to ten weeks at maintenance, and start another rather than extending the first — the failure at week six is adherence, and a longer block does not fix it.

Will I lose strength while cutting?

Some, and less than most people expect if you keep the load. Across the trials Murphy and Koehler pooled, strength gains in a deficit were not significantly different from strength gains in energy balance, although the point estimate still favoured eating enough. Falling top sets across several lifts for two weeks running is the signal that the deficit has gone too far.

Do I need cardio to lose fat?

No. Cardio widens the deficit and does nothing that eating less does not already do. It has a cost — the interference effect scales with how much and what kind you do, and running interferes more than cycling. Set the deficit with food, hold a daily step floor, and add structured cardio only when you have taken intake down as far as you are willing to.

Why did the scale stop moving in week five?

Usually because the deficit closed itself. Expenditure falls as you get lighter, and by more than the change in body mass predicts, so the 500 calories you set in week one is not 500 calories by week six. That is what the weekly weight average is for: when the trend flattens for two weeks, you take another 200 calories out or add steps.

Most Cuts Die in Week Six. Here Is One That Doesn't

Written while building GETSIXPACK — a training log that scores the session against your range and calls the next load.

Almost nobody quits a cut in week one. Week one is easy — the scale drops two kilos, most of it water, and the plan feels like it is working faster than promised.

Week six is where it goes. The scale has been flat for nine days, the top set that felt easy in week two is now a grind, you are thinking about food during meetings, and the whole thing starts to feel like it is not worth it.

That is not a willpower problem. It is what an under-specified plan looks like when it meets a body that has started adapting. So here is the specified version: four numbers, twelve weeks, and an ending — because a cut without a planned ending is the reason the abs you got in April are gone by August.

The deficit: 500 calories is a ceiling, not a target

This is the number everyone sets too aggressively, and there is now a reasonably clean answer for where the limit sits.

Murphy and Koehler pooled every randomised trial they could find that ran resistance training in an energy deficit for three weeks or more. Lean mass gains were clearly impaired compared with training in energy balance. Strength gains were not — the difference there was not statistically significant. When they ran the deficit size against lean mass across all the studies, a deficit of around 500 calories a day was the point where gains in lean mass stopped happening at all. Their recommendation is blunt: if you are lifting to keep muscle while losing fat, do not go past a 500 calorie daily deficit.

Meal prep containers of chicken, rice and vegetables lined up on a counter. — The deficit is the only variable that removes fat. Everything else on this page exists to protect what is underneath it.

The other common guideline comes from Helms and colleagues, whose review of natural bodybuilding preparation recommends losing roughly 0.5 to 1% of bodyweight per week to maximise muscle retention. Those two numbers do not agree as neatly as they look, and it is worth seeing where they collide.

A kilogram of body fat is worth roughly 7,700 calories. Run that both ways:

Bodyweight	0.5%/week	Daily deficit	0.7%/week	Daily deficit
70 kg / 154 lb	0.35 kg	~385 kcal	0.49 kg	~540 kcal
80 kg / 176 lb	0.40 kg	~440 kcal	0.56 kg	~615 kcal
90 kg / 198 lb	0.45 kg	~495 kcal	0.63 kg	~695 kcal
100 kg / 220 lb	0.50 kg	~550 kcal	0.70 kg	~770 kcal

That table is a rough conversion and it flatters the deficit slightly: 7,700 calories is the energy density of fat tissue, while Helms’ percentage is of bodyweight, and some of what comes off is water and some is lean tissue. Read it as an order of magnitude, not a prescription.

Even so, the point survives. The whole right-hand column is above the 500 calorie ceiling, and the left-hand one clears it only up to about 90 kg. The two guidelines agree at the bottom of the Helms range and nowhere else. Take the lower end: half a percent of bodyweight a week is the target, and 1% is what you use when there is a deadline and you have accepted the cost.

Garthe and colleagues ran that comparison directly in elite athletes: one group losing 0.7% of bodyweight per week, the other 1.4%, both lifting four times a week on individualised diets. Over the intervention the slow group gained more lean mass (2.0% versus 0.8%) and considerably more upper body strength (12% versus 6%), and shed a larger share of the fat they started with (31% versus 23% — both groups lost about the same total bodyweight; the fast group simply got there in fewer weeks).

Two things worth reporting against that, because the same paper says them. When the athletes were retested twelve months later the differences between the groups had washed out — and body mass and fat mass were back to baseline in both. Everybody regained. That is a 23-athlete follow-up, so it is an underpowered null rather than a demonstration that rate does not matter. What it does say is that how you cut decides what you keep on the way down, and what you do afterwards decides whether you keep any of it — which is what the last section of this article is about.

Set intake from your own data, not a calculator

Every online calculator estimates your maintenance calories from height, weight, age and an activity multiplier you pick from a dropdown. The estimate is a population average with a wide spread, and the activity multiplier is a guess about a guess.

There is a better two-week move.

Eat the way you normally eat, without trying to change anything. Weigh yourself every morning under the same conditions and average the seven readings each week. If the two weekly averages are the same, your average intake over those two weeks is your maintenance — not an estimate of it, a measurement of it. Then subtract.

The reason this beats the formula is not just precision. It is that your maintenance number will move during the cut, and the only instrument that can see it move is your own weight trend. Kevin Hall and colleagues modelled this in The Lancet: the body’s weight response to a change in intake is slow, with a half-time on the order of a year, because expenditure keeps falling as you get lighter. A static calculation done in week one describes a person who no longer exists by week eight.

Protein: the number you do not negotiate

Three numbers get quoted here, they use different denominators, and that is where most of the confusion lives.

1.6 g per kg of bodyweight. From Morton and colleagues’ meta-analysis of 49 studies and 1,863 participants: protein intake beyond about 1.62 g/kg/day produced no further gains from resistance training. That is the ceiling for building muscle in energy balance — not a cutting number. It is also softer than it gets quoted: the break point carries a confidence interval of 1.03 to 2.20 and the paper itself reports it as not statistically significant.
2.3 to 3.1 g per kg of lean body mass. The Helms review’s recommendation for contest preparation. Note the denominator: lean mass, not bodyweight. For an 80 kg lifter at 15% body fat, that is 68 kg of lean mass, so 156 to 211 g a day.
2.4 g per kg of bodyweight. What Longland and colleagues fed one arm of a four-week trial in which young men ate at a 40% deficit while training six days a week. The high-protein group gained 1.2 kg of lean mass and lost 4.8 kg of fat. The 1.2 g/kg group gained 0.1 kg and lost 3.5 kg.

Four weeks, a 40% deficit, six training days a week — and the only difference between the two groups was protein

Fat lost — 2.4 g/kg protein4.8 kg
Fat lost — 1.2 g/kg protein3.5 kg
Lean mass gained — 2.4 g/kg protein+1.2 kg
Lean mass gained — 1.2 g/kg protein+0.1 kg

All four bars are kilograms on one axis. Longland and colleagues, 2016 — 40 young men, four weeks. A proof of principle at an extreme deficit, not a template.

Two chicken breasts searing in a heavy pan, shot close. — Longland's high-protein group gained lean mass on a 40% deficit. The control group, eating half the protein on the same calories, did not.

That last study is worth flagging as the outlier it is. A 40% deficit with six training days a week for four weeks is a proof of principle, not a template, and the participants are described only as “young men” — how trained they were is not something the paper tells you. But the direction is consistent everywhere: the leaner you are and the deeper the deficit, the higher protein needs to go.

Practical version: 2 g per kg of bodyweight, every day, no exceptions for weekends. It sits inside all three recommendations and it is a number you can hit without a spreadsheet.

Training: cut the volume, keep the load

This is the single most important decision of the twelve weeks and most people get it backwards. They keep the set count and drop the weight on the bar, because lighter weights feel more appropriate to how tired they are.

Spiering and colleagues reviewed what it actually takes to maintain strength and muscle size over time rather than build it. In younger populations, both were held for up to 32 weeks on as little as one session a week and one set per exercise — provided the relative load was maintained. Their conclusion names the variable directly: intensity is the thing that preserves performance, even when frequency and volume fall a long way.

Being straight about the join: none of that data was collected in an energy deficit, and the review says outright that there is not enough evidence to make specific recommendations for athletes. Applying it to a cut is an inference. It is a well-supported one — the mechanism does not obviously change when calories drop — but it is an inference, not a finding.

Racked barbells and stacked weight plates in a gym, in near-monochrome. — Volume is the fatigue you cannot recover from on low calories. Load is the signal that keeps the muscle.

So: keep the weights, cut the sets. Take your weekly working sets per muscle down toward the lower end of whatever you have been running, protect the top set of each exercise, and stop taking every set to failure.

Murphy and Koehler found no significant difference in strength gains between lifting in a deficit and lifting in balance, which is the closest thing to permission the literature offers. Read it carefully, though: the point estimate still favoured energy balance, and a non-significant result in a small meta-analysis is missing evidence rather than proof of no effect. The useful version is not “your strength is safe” but “your strength is the instrument” — a top set that starts falling across several lifts is the earliest signal you get that the deficit is too large.

Cardio is a lever, not a fat burner

Cardio does not burn fat in any sense that matters here. It enlarges the deficit, which is a useful thing for it to do, and it has a cost.

Wilson and colleagues’ meta-analysis of concurrent training found the interference effect is dose-dependent — it scales with the modality, duration and frequency of the aerobic work, with running interfering more than cycling. Small amounts of low-intensity work are close to free. An hour of hard running on your leg day is not.

A man walking on a treadmill in a gym, seen from the front. — Steps first, machines second. Cardio's job is to widen the deficit without taxing the sessions that keep your muscle.

The order of operations that wastes the least: food first, steps second, cardio third. Set the deficit with intake. Hold a step count you can actually keep — pick a floor and hit it daily rather than chasing a big number twice a week. Add structured cardio only when intake has come down as far as you are willing to take it, and put it as far from your hardest lifting sessions as the week allows.

Sleep, and the part of the deficit that vanishes quietly

Nedeltcheva and colleagues took ten overweight adults, put them on the same moderate calorie restriction twice, and changed one thing: 8.5 hours of sleep opportunity in one arm, 5.5 in the other.

Same diet, same deficit, two weeks each. On short sleep, the proportion of the weight lost as fat fell by 55% — 1.4 kg of fat versus 0.6 kg — and the loss of fat-free mass rose by 60%, from 1.5 kg to 2.4 kg. Hunger was higher. The total weight lost was similar; what changed was what came off.

That is the clearest demonstration available that the composition of your weight loss is not decided by calories alone. It is also the cheapest variable on this entire page.

The other quiet leak is the one your body does on purpose. Rosenbaum and Leibel describe it as adaptive thermogenesis: as you lose weight, energy expenditure falls by more than the change in body mass predicts, and Trexler and colleagues lay out what that means specifically for athletes — reduced non-exercise movement, lower resting expenditure, and a deficit that shrinks while you are still eating the same amount. You cannot switch it off. You can notice it in the weight trend and adjust, which is the entire reason for weighing daily.

Diet breaks: what they do, and what they don’t

Peos and colleagues ran the cleanest test of this — 61 resistance-trained adults over twelve weeks of energy restriction, randomised to continuous restriction or to three-week blocks separated by one-week returns to energy balance.

Fat loss was the same. Fat-free mass retention was the same. Strength and endurance were the same. What differed: the intermittent group reported significantly lower hunger, less desire to eat, and greater satisfaction.

So a diet break is not a fat loss accelerator, and anyone selling it as one is overselling. It is an adherence tool, and adherence is what week six is actually about. Worth knowing that it costs calendar time — the intermittent arm needed fifteen weeks to complete twelve weeks of restriction.

The twelve weeks

Two weeks of baseline measurement happen before the clock starts. Do not skip them; they are what makes every number below yours rather than a guess.

Weeks	What you are doing	What you are watching
1–3	Intake at maintenance minus ~450 kcal. Protein at 2 g/kg. Step floor set. No cardio yet.	Weekly average weight, not daily. Weeks 1–2 will over-report — that drop is glycogen and water.
4	First real checkpoint. Three weeks of trend data.	Losing 0.4–0.6%/week? Hold everything. Losing nothing? Cut another 200 kcal or add steps.
5–7	The hard part. Volume comes down, load stays.	Top sets. If they are falling across multiple lifts, the deficit is too large — not a sign to push harder.
8	One week at maintenance. Same training, same protein, more carbohydrate.	Hunger and sleep resetting. The scale will rise 0.5–1.5 kg. That is water.
9–11	Back to the deficit. This is where cardio earns its place if it has not already.	Adherence. Anything you cannot sustain for three more weeks is not part of the plan.
12	Hold, then exit.	The five signals below.

Bare feet on a digital bathroom scale on a wooden floor. — Daily readings, weekly averages. One morning's number is noise; four of them in a row is information.

When to stop, and why “leaner” is not automatically better

Five signals, any two of which mean the cut is finished regardless of what the mirror says: top sets falling across several lifts for two weeks running, sleep breaking down, libido gone, hunger you are thinking about during work, and adherence that has stopped being a decision and started being a fight.

For the version of this with numbers attached, there is a well-documented case study. Rossow and colleagues tracked a drug-free male bodybuilder for six months before a competition and six months after. Body fat went from 14.8% to 4.5%. Testosterone fell from 9.22 to 2.27 ng/mL. Total mood disturbance went from 6 units to 43.

Most of that came back. Testosterone returned to 9.91 ng/mL and mood to 4 units within the six months after the show. One thing did not: strength decreased during preparation and had not fully recovered six months later.

That is what the extreme end costs, in a compliant, well-coached athlete doing everything correctly — and it is one person, so treat it as an illustration rather than a rate. Nothing about it says a cut is dangerous. It does say that “lean” and “leaner” are not the same project, that the price curve gets very steep at the end, and that the thing slowest to come back is the thing you were cutting to show. Most people reading this want the body fat where abs become visible, which is nowhere near 4.5%.

The exit

The twelve weeks are not over when you hit the target. They are over when you are eating at maintenance again without having regained the fat.

Add calories back gradually — roughly 100 to 150 a week, mostly carbohydrate — until the weight trend flattens. That is your new maintenance, and it will be lower than your old one, partly because you are lighter and partly because of the adaptation described above. Then stay there for six to ten weeks before you do anything else.

Nobody does this part. It is why the same people run the same cut every spring.

Two weeks of baseline before week one. Eat normally, weigh daily, average each week. If the two averages match, that number is your maintenance.
Set intake at maintenance minus about 450. Then leave it alone for three weeks — a plan you adjust weekly has no trend to read.
Protein at 2 g per kilogram of bodyweight, weekends included. It is the one number in the plan with no acceptable bad day.
Keep the load, cut the sets. Protect the top set of every exercise and stop taking everything to failure.
Weigh every morning, decide off the weekly average. One reading is noise. Four in a row is information.
Watch the top sets, not the mirror. Falling across several lifts for two weeks means the deficit is too large.
Write the exit down before week one. 100 to 150 calories back a week until the trend flattens, then six to ten weeks at that number.

If you want to know how many of these twelve-week blocks stand between you and the number you have in mind, that is a separate piece of arithmetic, and it takes about two minutes.

Where this comes from

The 500-calorie ceiling and the finding that strength survives a deficit while lean mass gains do not: Murphy and Koehler, Scandinavian Journal of Medicine & Science in Sports, 2022 (doi:10.1111/sms.14075).

The 0.5–1% per week guideline and the protein recommendation in grams per kilogram of lean mass: Helms, Aragon and Fitschen, Journal of the International Society of Sports Nutrition, 2014 (doi:10.1186/1550-2783-11-20).

The slow versus fast weight loss comparison in elite athletes. The paper linked here is the twelve-month follow-up, which reports both the intervention results and what was left of them a year on: Garthe, Raastad and Sundgot-Borgen, International Journal of Sport Nutrition and Exercise Metabolism, 2011 (doi:10.1123/ijsnem.21.5.426).

The 1.62 g/kg/day plateau: Morton and colleagues, British Journal of Sports Medicine, 2018 (doi:10.1136/bjsports-2017-097608). The 2.4 versus 1.2 g/kg trial during a 40% deficit: Longland and colleagues, American Journal of Clinical Nutrition, 2016 (doi:10.3945/ajcn.115.119339).

Why a static calorie calculation drifts: Hall and colleagues, The Lancet, 2011 (doi:10.1016/S0140-6736(11)60812-X). Adaptive thermogenesis: Rosenbaum and Leibel, International Journal of Obesity, 2010 (doi:10.1038/ijo.2010.184), and Trexler, Smith-Ryan and Norton, Journal of the International Society of Sports Nutrition, 2014 (doi:10.1186/1550-2783-11-7).

Maintaining strength and size on a reduced dose, and why load is the variable that matters: Spiering, Mujika, Sharp and Foulis, Journal of Strength and Conditioning Research, 2021 (doi:10.1519/JSC.0000000000003964).

The dose-dependence of the interference effect: Wilson and colleagues, Journal of Strength and Conditioning Research, 2012 (doi:10.1519/JSC.0b013e31823a3e2d).

Sleep and the composition of weight lost: Nedeltcheva and colleagues, Annals of Internal Medicine, 2010 (doi:10.7326/0003-4819-153-7-201010050-00006).

Diet breaks: Peos and colleagues, the ICECAP trial, Medicine & Science in Sports & Exercise, 2021 (doi:10.1249/MSS.0000000000002636).

The twelve-month case study of a natural bodybuilder: Rossow and colleagues, International Journal of Sports Physiology and Performance, 2013 (doi:10.1123/ijspp.8.5.582).

And what is not from a study. The specific figures in the twelve-week table and the exit — 450 calories to start, another 200 if nothing moves, the 0.5 to 1.5 kg the scale rises during a maintenance week, 100 to 150 calories a week on the way back up, six to ten weeks at the top — are programming decisions, not findings. They sit inside the ranges above, and they are the numbers that have worked for us. Nothing was measured to produce them.