You didn't regain the weight because your metabolism broke. You regained it because a diet has a finish line and maintenance never got a plan. The day you hit your goal weight, every rule you'd been running expired at once — the target, the tracking, the weekly check-in — and nothing took their place. Metabolic adaptation is real, but for a normal dieter it's worth somewhere between about 50 and 180 calories a day, not the 700 the headlines sell. That's a rounding error next to six months of eating without a number.
So this isn't a physiology article. It's the system that runs after the diet ends — the part nobody sells you.
Why do I keep gaining the weight back?
Because the plan that got the weight off was built to end, and you never built the one that replaces it. Every diet ships with a stop condition — a goal weight, a wedding, a beach. Maintenance has no stop condition, so most people don't recognize it as a phase at all. It just feels like being finished.
Then the drift starts, and it's slow enough to be invisible. Eat 200 calories a day above maintenance — one extra beer, or a serving size you stopped weighing — and at roughly 3,500 calories per pound of fat, that's about 1.7 lbs a month. Twenty pounds a year. Nobody notices 1.7 lbs a month. You notice it much later, in a photo — which is exactly how it happened to me the second time.
That's the whole mechanism. Not a metabolic collapse, not a willpower failure — a small, unmeasured gap running unopposed for a long time. (If you're still in the losing phase, how long a cut should be covers that side of it — this article starts the day the cut ends.)
Is my metabolism broken from dieting?
No. It's a bit slower, and the slowdown is not what put the weight back on you.
The study everyone quotes for the opposite view is Fothergill and colleagues, 2016 in Obesity — the "Biggest Loser" follow-up. Fourteen former contestants, six years after a 30-week televised competition: they lost 58.3 kg on the show, regained 41.0 kg of it, and their resting metabolic rate was still 704 kcal/day below baseline. Adjusted for the body composition they were carrying, the unexplained part — the actual metabolic adaptation — was 499 kcal/day. That's where "your metabolism is destroyed" comes from, and it's a real finding.
Read the rest of the paper, though. Weight regain was not correlated with metabolic adaptation at the end of the competition (r = −0.1, P = 0.75). And the contestants who had kept the most weight off at six years were the ones with the greatest metabolic slowing. The slowdown tracked with success, not failure. Whatever that slowdown is, it isn't the reason the weight came back.
But 30 weeks of extreme restriction inside a television competition is not your cut. For the rest of us there's Martins and colleagues, 2020 in the American Journal of Clinical Nutrition, titled, unambiguously, "Metabolic adaptation is not a major barrier to weight-loss maintenance." They followed 171 women with overweight for two years, measuring resting metabolic rate under conditions of weight stability. Adaptation came out to −54 ± 105 kcal/day, it didn't correlate with regain, and in the 46 women with data at every time point it was gone by the one-year mark. In the MATADOR trial, the continuously-dieting men finished 749 kJ/day below prediction after adjusting for body composition — about 180 calories.
Caveat where it's due: those were women with overweight and men with obesity, not lifters. But the direction holds across all three. Your maintenance number is genuinely lower than a guy who was always this weight — by roughly one snack, which is exactly the size of gap a measured system absorbs and guessing never sees.
What does a maintenance system actually look like?
Four settings, and only one of them changes when the cut ends.
- Recalculate maintenance at your new bodyweight. Bodyweight in pounds × 12. At 200 lbs that's about 2,400 calories a day, and you eat at it — not below it, not "intuitively." Most regain starts here, because people finish a cut and go back to what they ate at their old, heavier bodyweight. That number belonged to a different body. (Same arithmetic as how big a calorie deficit you need — you're just not subtracting.)
- Protein stays at 1 gram per pound. 200 g at 200 lbs, same as during the cut — it's the most filling macro per calorie and it holds the muscle you spent months protecting. Drop it and you arrive back at the same scale weight looking softer than before. (How much protein you actually need has the research behind the number.)
- Training doesn't change at all. Four lifting days a week — Push, Pull, Legs, Upper — still adding weight or reps, plus 10,000 steps a day. Nothing about maintenance says train less — the training is the reason the calories can be that high. Not the muscle itself: a pound of muscle burns about 6 calories a day, so it’s the work that buys you the calories, not the tissue.
- The 7-day average is the scorecard, and flat is a win. Weigh in every morning, post-bathroom, before food, and judge nothing but the 7-day average. During the cut you wanted that line going down. Now you want it going nowhere. Judging the week instead of the day is also what makes a dinner out survivable — calorie banking is just spending the same week's budget unevenly, on purpose.
That's the same No-Guess Loop — track, review, adjust, build — steering toward a different number. Identical machinery: measure, read the trend instead of the day, change one thing only when the trend says to. The only edit is the target.
Here's the version of me that system produced — 37, 8–10% body fat, and the weight still off.
How do I know I'm drifting before the weight is back?
Set a ceiling 3–5 lbs above your goal weight and treat crossing it as a signal, not a verdict. It works because it fires while the problem is still two weeks of work instead of two months.
Two rules make it work. The tripwire sits on the 7-day average, not on any single morning — a salty dinner puts 3 lbs on the scale overnight and means nothing. And you don't get to move the line after you cross it.
When the average crosses, run a two-week corrective block: maintenance minus 500, protein and training unchanged. At roughly 3,500 calories to a pound that's about 2 lbs in two weeks — precisely the size of the problem. Then back to maintenance. Not a new diet. Not a fresh start in January. Two weeks and it's closed.
None of it works if the log underneath is lying to you. I weigh in every morning and read only the trend, and I still spent a stretch eating roughly 300 calories a day more than I thought, because the apples I was buying were much bigger than the "standard apple" my tracking app assumed. In a cut that shows up as a stall you can at least see. At maintenance it doesn't show up as anything for months — it just quietly becomes 20 lbs. A trend that isn't behaving means you've adapted or your log is wrong, and why the scale won't move in a deficit audits the second one.
What do people who actually keep it off do differently?
They keep measuring. That's the finding, and it's less exciting than anyone wants.
The best long-term dataset on this is the National Weight Control Registry. Thomas and colleagues, 2014 in the American Journal of Preventive Medicine followed 2,886 members over 10 years, all of whom had lost at least 30 lbs and held it for at least a year. Mean weight loss was 31.3 kg at entry and still 23.1 kg a decade later, with more than 87% estimated to still be maintaining at least a 10% loss at 5 and 10 years. Keeping it off is not rare among people who build a system for it.
The regain half is the useful half. Three behaviors predicted greater regain when they declined: physical activity, dietary restraint, and frequency of self-weighing. Note the direction — the measuring tapers first, the weight follows. It's observational, so it can't prove the scale is causal. But the habit stopping and the weight returning travel together, it costs ten seconds a day, and it's the one item on that list you can restart tomorrow morning.
Two more details from the same registry, because the popular version of both is wrong. On activity: members average 2,621 kcal a week (Catenacci and colleagues, 2008) — with a standard deviation of ±2,252, and a quarter reporting under 1,000. The "they all do an hour a day" line is an average balanced on enormous variance; don't build a plan on it. On weekends: Gorin and colleagues, 2004 asked 1,429 members whether they ate the same way all week or dieted harder Monday to Friday. The consistent group had 1.58 times the odds (95% CI 1.2–2.2) of staying within 5 lbs over the following year. Strict all week, amnesty on Saturday, is the most common maintenance pattern there is — and it's the one that loses.
Why did it work on the third attempt and not the first two?
Because the first two attempts had finish lines and the third one didn't. Three cycles, two failures, and the difference wasn't discipline.
The first started in Seattle at 285 lbs, in a dark apartment, at the bottom. Two years of work took about 80 lbs off. Then a startup, a girlfriend who loved restaurants, and a foot injury from wakesurfing put it back on, and I rebuilt. The second ended in a photo from a trip through Rome and Croatia, where I looked at myself and understood I had done the entire thing over again.
Both cycles ran the same shape. Get lean. Stop measuring. Drift. Notice too late. What changed the third time was the question — I stopped asking how to lose weight and started asking how to build a system I'd never have to ask it inside of again. Nine months of tracked, adjusted work later I was in the best shape of my life, and 285 lbs had become about 200 lbs at 8–10% body fat, held year-round. That was over a year ago and the number hasn't moved, because the loop never turned off.
The honest read: the first two failures are what made the third one work. I'm 37, I work full-time as a software engineer, and my calendar is overwhelmingly maintenance rather than cutting — which is why I think maintenance is the actual product and the cut is just the setup. If you've been around this loop twice already, that's the exact conversation I have in coaching. The fix is almost never a harder diet; it's the plan for the six months after it.