The Maintenance Statistics
The deficit phase gets the headlines; the twelve months after it get the data nobody wants on a poster. Most lost weight comes back, but the statistics also carry the less comfortable news — that some people keep it off for years, and what separates them is measurable and mostly unglamorous. This page puts the regain numbers on the table, names their limits, and explains why the post-diet phase is the real test of whether a weight-loss effort actually worked.
What the evidence supports
- On average, dieters keep only a small fraction of their loss years out — roughly a few percent of starting body weight in pooled follow-up.
- A meaningful minority does maintain substantial loss long-term, and that minority shares a reproducible behavioral profile.
- Once the deficit stops, the biology that pushed weight down keeps pushing back for at least a year — regain is not simply a willpower failure.
What remains uncertain
- Pooled averages hide a wide distribution — some people keep everything, most regain partially, and the labels flatten that into one scary number.
- Registry data relies on volunteers who self-select for success, so it overstates how common durable maintenance really is.
- Whether the newer weight-loss drugs change the long-term curve is, as of this writing, still being measured.
Evidence last reviewed: August 15, 2026. Conclusions may change as new research is published.
the real test
The Numbers Nobody Puts on a Poster
The landmark reference point is a meta-analysis of longer-term weight-loss studies: averaging across trials that followed people for five years or more, participants maintained only about three percent of their initial body weight — roughly a quarter of what they had lost on the program (Anderson et al., American Journal of Clinical Nutrition, 2001). Let the two framings land together: three percent of starting weight is a small, almost demoralizing number, and it also means the average person did hold on to something five years later rather than returning precisely to baseline. The direction is unmistakable, but the headline hides the shape of the distribution — which is where this page lives.
A second, often-quoted figure comes from a national survey rather than a clinic: roughly one in six US adults with prior overweight or obesity had ever maintained a loss of ten percent or more of their body weight for a full year (Kraschnewski et al., International Journal of Obesity, 2010). The honest reading is neither "nobody succeeds" nor "anyone can" — durable maintenance is a minority outcome with a reproducible recipe, and the recipe is the subject of the rest of this series.
The Regain Curve, Quantified
The cleanest way to see the problem is to chart where the lost weight goes over time. The pattern below — the large drop, the partial return in year one, the slow bleed thereafter — is the typical shape across pooled follow-up data. The percentages are rounded averages, not anyone's personal forecast.
Read the third bar as the design brief for the rest of this series: the most valuable work happens in the window the chart makes visible — the first year, where about half the eventual regain typically lands. That is exactly why the maintenance protocol treats the transition period as a phase to engineer rather than a finish line to celebrate.
Who Keeps It Off: What the Registry Says
Set against the averages is a group that beats them: the National Weight Control Registry, which tracks adults who have lost at least 13.6 kg (30 lb) and kept it off for at least a year. The registry's earliest large description found that members reported keeping weight off for an average of more than five years, and most of them follow a strikingly ordinary playbook — roughly an hour of activity a day, a modest and consistent calorie intake, regular breakfasts, and, in the more recent waves, regular weighing (Klem et al., 1997; Thomas et al., American Journal of Preventive Medicine, 2014). The registry is a biased sample — it attracts the winners — but that is precisely its use: it is the best existing catalog of what durable maintenance looks like when it happens, and the profile is replicable.
The Two Ways the Average Lies
- 📉 The average hides the distribution. "Three percent kept" is a mean across people who kept everything, people who kept half, and people who regained more than they lost. The spread matters more than the mean, and no single figure can describe your run.
- 🧾 The registries self-select. Registry members volunteered after succeeding, so registry-based success rates overstate the population baseline. Both directions are real: unmanaged regain is common, and managed maintenance is possible — they are not in tension.
- 🔬 The follow-up window thins out. Beyond five years the datasets get sparse, so the decade-plus numbers are the least certain in this whole story — worth remembering whenever someone quotes "the 95 percent" of dieters who fail.
That last point deserves a beat. The famous "95% of diets fail" figure traces to a handful of clinic follow-ups and a self-referential survey, not to a single authoritative dataset, and serious reviews do not support it as a literal number. The data that does exist is bad enough without being inflated: the honest summary is that most lost weight tends to return in the absence of a system, and a minority keeps it off with one.
Why the Post-Diet Year Is the Real Test
The statistics make sense only with the biology underneath them. After weight loss, the body operates as if the lower weight is the emergency: hunger hormones stay elevated and circulating energy expenditure is lower than expected for the new body size, and these adaptations persist for at least a year after the loss (Sumithran et al., New England Journal of Medicine, 2011; MacLean et al., 2011). None of that makes regain inevitable; it makes it engineered — the default when the deficit simply stops and nothing takes its place. The deficit phase stored up a physiological headwind that the maintenance phase has to sail against, and most people are never told the wind is there.
📊 Averages describe groups, not your plot line
Every statistic on this page is a group summary, and groups are not individuals. The point of knowing the regain curve is not to prophesy your failure — it is to design against a real, measured default. The people who keep weight off are not resisting the average through rare willpower; they are running structures the average never included. The next four pages in this series are those structures: the reverse-diet transition, the calorie band, the trigger rule, and the identity shift that makes the others survivable.
The Regain Data at a Glance
| Milestone | Typical finding | Source type | Read |
|---|---|---|---|
| 📉 Loss at program end | Roughly 5–10% of starting body weight | Randomized trials (DPP lifestyle arm ≈7%) | The achievable win |
| ⏳ One year out | A third to a half of the loss typically regained | Pooled follow-up | The critical window |
| 🔁 Five years out | About 3% of starting weight kept, on average | Meta-analysis (Anderson et al., 2001) | The bleed |
| 🏆 The maintainers | 1 in 6 nationally with ≥10% kept for a year; registry shows the recipe | Survey + registry (Kraschnewski, 2010) | The blueprint |
Questions, Answered Briefly
- 😕 Isn't it true that 95% of diets fail? Not as a literal figure — that claim has weak provenance. The real data is sobering enough: most lost weight tends to return without a system, and a minority keeps it off with one. The useful question is not whether diets fail but which phase was never run.
- 📉 Does this mean losing weight is pointless? No — short-term loss reliably improves cardiometabolic markers, and maintenance changes the story entirely. The statistics condemn the off-switch approach, not the deficit itself.
- 💉 Are the GLP-1 drugs changing these numbers? Too early to say from long-term data. Continuous-treatment trials keep weight off while people stay on the drug; the maintenance statistics here describe the lifestyle-default world. The GLP-1 era page owns that question in depth.
- ⚖️ Does self-weighing actually predict success? The association is consistent: people who keep measuring tend to keep weight off — roughly 44% of registry maintainers weigh themselves at least weekly (Wing & Phelan, 2005). The trigger rule is that finding turned into a decision rule.
The Bottom Line
- Most lost weight tends to return without a system — roughly one-quarter of the initial loss is typically kept at five years, about 3% of starting body weight.
- The first year is the critical window — much of the eventual regain lands there, while hormonal push-back from the loss is strongest.
- A minority does keep it off long-term — the registry recipe is ordinary: regular activity, consistent intake, and ongoing measurement.
- Statistics are for design, not prophecy — the regain curve defines the problem the next four pages solve, not your personal forecast.
Related Topics
- Anderson et al., "Long-term weight-loss maintenance: a meta-analysis of US studies," American Journal of Clinical Nutrition (2001)
- Kraschnewski et al., "Long-term weight loss maintenance in the United States," International Journal of Obesity (2010)
- Klem et al., "A descriptive study of individuals successful at long-term maintenance of substantial weight loss," American Journal of Clinical Nutrition (1997)
- Thomas et al., "Weight-loss maintenance for 10 years in the National Weight Control Registry," American Journal of Preventive Medicine (2014)
- Wing & Phelan, "Long-term weight loss maintenance," American Journal of Clinical Nutrition (2005)
- Sumithran et al., "Long-term persistence of hormonal adaptations to weight loss," New England Journal of Medicine (2011)
- MacLean et al., "Biology's response to dieting: the impetus for weight regain," American Journal of Physiology — Regulatory, Integrative and Comparative Physiology (2011)