Two Hours of Lifting a Week — and a Wall Where "More" Stops Paying
June 21, 2026 · Nesse & Williams, Why We Get Sick~5 min read
You'd assume the math is simple: if a little lifting is good for you, a lot must be better, and a gym rat who trains five hours a week is buying himself more years than the rest of us. A June 2026 study of 147,374 adults, followed for three decades, draws a different shape — a line that climbs, levels off near two hours a week, and then just stops paying. More isn't more. After a point, the extra effort buys almost nothing.
The number is smaller than the gym wants you to think
Here is what the researchers, writing in the British Journal of Sports Medicine, actually found. Across 31,540 men and 115,834 women tracked for 30 years — 35,798 of whom died over that span — the people who did resistance training landed in a clear sweet spot at 90 to 119 minutes a week. In that band, all-cause mortality ran about 13% lower; death from cardiovascular disease, 19% lower; death from neurological disease, 27% lower. Not a miracle, but real and worth two short sessions. Then the line goes flat. Beyond 120 minutes a week, the paper reports no additional reduction in risk at all. Train three hours, four hours, five — the curve doesn't reward you for it.
Dose–response for resistance training: all-cause mortality falls to about 13% lower in the 90–119 min/week band — the sweet spot — and shows no additional reduction beyond 120 min/week. Source: British Journal of Sports Medicine (2026), 147,374 adults, 30-year follow-up; this is observational and cannot prove causation. Popular-science interpretation, not medical advice — consult a clinician.
Why a flat line is exactly what a body should give you
That plateau looks like a disappointment. It's closer to a fingerprint. Nesse and Williams, in Why We Get Sick, make a point that reframes the whole chart: natural selection never tuned your body for maximal health, let alone gym aesthetics. It tuned you for reproduction, and it did so through trade-offs — every system set to a level that paid off on average, not pushed to its ceiling. A body built that way runs on dose curves, not on "more is better." Think of watering a plant. Dry soil kills it; the right amount makes it thrive; a flood drowns it. The benefit of water is a hump, not a ramp. Strength training behaves the same way: a stress the body answers by getting stronger, useful in the right dose, with diminishing returns and then real cost past the point where you needed it. The wall at 120 minutes isn't the data failing to find more benefit. It's what an evolved system looks like when you finally graph it honestly.
The dose even shifts depending on what you're dodging
One detail makes the dose-curve idea hard to wave away: the sweet spot isn't the same for every cause of death. For cancer mortality, the benefit showed up at the lowest doses — about 21% lower at just 1 to 29 minutes a week, and 18% lower at 30 to 59 minutes. A tiny weekly habit, and the cancer-death curve already bends. Stack strength work on top of cardio and the numbers climb again: 60 to 119 minutes of lifting plus moderate aerobic exercise tracked with roughly 45% lower mortality, and high aerobic volume (45-plus MET-hours a week) plus any strength training with 53 to 58% lower. Different threats, different optimal doses, every one of them a hump rather than a ramp. No single "max it out" number governs the body, because the body was never built around one.
Before you rearrange your week — the catch
This is the part a careful reader has to sit with, because it's where the headline overreaches. The study is observational. It watched who trained and who died; it did not assign anyone to lift or not. So it cannot tell you that strength training caused the lower mortality. The people who keep up a structured strength habit for years may simply be healthier, more disciplined, better resourced, less sick to begin with — and any of those could be doing the protecting while the dumbbells ride along. Correlation walks beside causation here; it does not prove it. Read the 13% as a signal worth acting on, not a dose you can swallow as a guarantee. The honest version is modest, and modest is the point.
What this means for you
So here's the realistic target, the kind that survives both the data and the caveat. Two strength sessions a week — call it 90 to 120 minutes total — is the band where the curve pays best; if you can pair it with some aerobic movement, better still. And then permission to stop chasing: the third and fourth hour aren't buying the years you'd hope. The body you're training was shaped to run on enough, not on maximum, which is oddly freeing. You don't have to out-train anyone. You have to clear the wall — and once you're over it, the smarter move is to go home. If you've got a heart condition, an injury, or a real medical question, that's a conversation for a clinician, not a chart.
Strength training pays out on a dose curve, not a ramp — clear the two-hour wall, and "more" mostly buys you a more tired week.
The 13% signal is encouraging but observational: a reason to start lifting, not proof that lifting alone adds the years.
This is a popular-science interpretation, not medical advice, and not a prescription — for any concern about your heart, joints, or starting strength training with a health condition, consult a qualified clinician and follow their guidance. Framework paraphrased from Randolph Nesse & George Williams, Why We Get Sick: The New Science of Darwinian Medicine (the body as a bundle of evolutionary trade-offs and dose curves; more of a good thing is not better). The figures refer to an observational study published in the British Journal of Sports Medicine (2026): 147,374 adults (31,540 men, 115,834 women) followed for 30 years, 35,798 deaths; a sweet spot of ~90–119 min/week of resistance training (13% lower all-cause, 19% lower cardiovascular, 27% lower neurological mortality), with no additional benefit beyond 120 min/week, and cancer benefit at low doses (21% at 1–29 min). Being observational, the study cannot prove strength training caused the lower mortality; defer to the original peer-reviewed paper.
研究者が『British Journal of Sports Medicine』で実際に見つけたものを見よう。30年追跡された男性3万1540人と女性11万5834人のうち、その間に3万5798人が亡くなった。筋力トレーニングをした人は、はっきりしたスイートスポットに収まった——週90〜119分だ。この帯で、総死亡率は約13%低く、心血管疾患による死は19%低く、神経系疾患による死は27%低かった。奇跡ではないが、本物で、短い2回のセッションに値する。そして線は平らになる。週120分を超えると、論文が報告する追加のリスク低下は——ゼロだ。3時間、4時間、5時間と練っても、曲線はそのぶん多く報いてはくれない。
筋トレの用量反応:週90〜119分のスイートスポットで総死亡率は約13%低くなり、週120分を超えても追加の低下は見られない。出典:British Journal of Sports Medicine(2026)、147,374人、30年追跡。観察研究であり因果は証明できない。本稿は科学解説であり医療助言ではない——医師に相談を。