How Long Sex Lasts by Age: What the Research Actually Shows
I've done two Ironmans and ten marathons, which means I have spent an unreasonable portion of my adult life staring at a watch.
Endurance sport will do that to you. You learn to think in splits. You know your pace per mile for every distance, you know what your heart rate should be at mile eighteen, and you know — with real precision — that your times are going to get slower as you get older. That's not a tragedy in running. It's just physics plus physiology. You adjust your expectations, you train smarter, and you keep going. Nobody quits the sport at forty-seven because they're no longer running a personal best.
And then there's the one arena where I've learned the stopwatch is actively working against me.
Because here's what's strange about sexual duration: it's the only performance metric in a man's life where measuring it makes it worse. Watching the clock during sex doesn't improve your time. It pulls you out of your body, puts you in the role of spectator, and reliably degrades the exact experience you were trying to optimize.
But people want the numbers anyway — and specifically, people in their forties, fifties, and sixties want to know whether what they're noticing is normal. So let's look at the actual research on how duration changes with age. Then let's talk about why the number going down is not the story you think it is.
The Numbers, By Age
The best data we have on this comes from a 2005 study published in The Journal of Sexual Medicine by Marcel Waldinger and colleagues. What makes it valuable is the methodology: rather than asking men to estimate, researchers gave 491 couples across five countries — the Netherlands, Spain, Turkey, the UK, and the US — an actual stopwatch and had the female partner time the interval from penetration to ejaculation over four weeks.
The overall median came in at 5.4 minutes. That's the figure that usually makes headlines, and I've written about why it's so much lower than most people assume.
But the study also broke the results down by age, and that breakdown is the part almost nobody reports:
Ages 18–30: median 6.5 minutes
Ages 31–50: median 5.4 minutes
Ages 51 and over: median 4.3 minutes
So yes — measured duration declines with age. A man over fifty lasts, on average, about two-thirds as long as a man in his twenties.
This surprises people, and I want to be specific about why. The popular assumption runs exactly the opposite direction: young men are supposed to be the ones who finish too fast, and older men are supposed to have developed control. The data says the reverse. Whatever cultural script we're all operating from, it does not match what the stopwatch found.
A few other findings worth knowing from the same study: condom use made no measurable difference to duration. Neither did circumcision. And the individual variation was enormous — 14% of men had a median under three minutes and twenty seconds, while 26% were over ten minutes. The spread between people dwarfs the spread between age groups.
Why the Number Goes Down
The reason has almost nothing to do with stamina, which is the word most men reach for and the wrong one.
Ejaculatory latency isn't an endurance event. It isn't governed by cardiovascular fitness or muscular conditioning or willpower. It's governed by a reflex arc — a neurological threshold involving serotonergic signaling, pelvic floor nerve conduction, and the sensitivity of the ejaculatory reflex itself. Those systems change with age the way every other system does.
As men age, a few things shift at once. Sensory nerve conduction changes. Erections may be somewhat less rigid, which paradoxically can reduce the duration of the plateau phase in some men. Testosterone declines gradually. And medications become more common — blood pressure drugs, SSRIs, and others can push latency in either direction.
Translating that into locker-room terms as "I've lost my stamina" is a category error, and it's an expensive one. It takes a neurological change and converts it into a character verdict. I see this constantly in coaching: a man in his fifties who's noticed a change and has quietly concluded that he's declining as a man, when what he's actually observing is a reflex threshold that moved a few minutes to the left.
What the Declining Number Doesn't Capture
Here's where the numbers get interesting, because the duration of penetration is a wildly incomplete measure of what sex is.
Consider the arithmetic problem. A 2020 study in The Journal of Sexual Medicine by Bhat and Shrivastava used the same stopwatch approach to measure time to orgasm in women in stable heterosexual relationships. The mean was roughly 13.4 minutes, with a range from 6 to 20 minutes. Earlier self-report research with over 2,300 women found a similar average of about 14 minutes during partnered sex.
Hold those two findings next to each other. Median penetration time: 5.4 minutes. Average time to orgasm for women: around 13 to 14 minutes.
That is not a stamina problem. That is a structural problem — a mismatch between how long one event takes and how long another event takes, which no amount of lasting longer will fix, because the clock on the woman's side usually starts well before penetration does. The research on time to orgasm measures from the beginning of genital stimulation, not from penetration. Which means the answer to the gap was never "last longer." It was always "the encounter is bigger than intercourse."
And this is exactly where age becomes an advantage rather than a deficit.
Because as the measured duration of intercourse goes down, the couples who do well are the ones who expand the definition of what they're doing. Everything that happens before penetration counts. Everything that happens after counts. The hands, the mouth, the conversation, the long sensory runway that so many women need in order to actually arrive. A couple in their fifties who have figured this out are having objectively better sex than a couple in their twenties running six and a half minutes of intercourse and calling it a night.
The younger couple has the better split time. The older couple has the better race.
What Happens to Frequency
Since we're being honest about numbers, frequency declines too — and more steeply than duration.
Analysis of General Social Survey data covering more than 26,000 American adults found that people in their twenties average roughly 80 sexual encounters per year, while people in their sixties average around 20. (Notably, that same research found the broader decline in American sexual frequency over recent decades is driven more by birth cohort than by age — people born in the 1930s had the most sex of any cohort studied, and those born in the 1990s the least. Make of that what you will.)
But the most useful data on later life comes from the National Social Life, Health, and Aging Project, published in The New England Journal of Medicine in 2007. Researchers surveyed 3,005 adults between 57 and 85. What they found:
Ages 57–64: 73% were sexually active in the past year
Ages 65–74: 53%
Ages 75–85: 26%
And among those who were sexually active, most reported having sex two to three times a month — including in the oldest group, where 23% reported sex once a week or more. That's people between 75 and 85 having sex weekly. Nearly a quarter of them.
The study also catalogued what gets harder. Among sexually active men: erectile difficulties (37%), lack of interest (28%), climaxing too quickly (28%), performance anxiety (27%). Among women: low interest (43%), difficulty with lubrication (39%), inability to climax (34%), pain at entry (17%). About half of sexually active older adults reported at least one bothersome problem.
I include that list on purpose. The honest picture isn't "sex is great forever, don't worry about it." The honest picture is that real physical changes arrive, they're common, most of them are addressable, and the people who stay sexually alive are the ones who treat them as logistics rather than verdicts.
The Trap of the Declining Number
Here's what I actually want you to take from all of this.
Every one of these figures describes an average of a population, and not one of them describes you. Remember that in the stopwatch study, the variation between individuals was far larger than the variation between age groups. A 55-year-old at the upper end of his age bracket is lasting longer than a 25-year-old at the lower end of his. The distributions overlap almost completely.
What the declining number does, if you let it, is give you a story. And the story is the problem.
In sport psychology, there's a basic distinction between state and trait. A state is a temporary condition — you're tight today, you slept badly, your legs are heavy. A trait is a stable characteristic of who you are. Athletes get wrecked when they take a state and file it as a trait: one bad race becomes "I'm not fast anymore," and then they train and compete as a man who isn't fast anymore, and the prophecy closes.
Men do exactly this with duration. One shorter-than-expected encounter at fifty becomes a verdict about aging, and the verdict produces anxiety, and the anxiety produces spectatoring — that cognitive step outside your own body to monitor and evaluate your performance — and spectatoring reliably makes everything worse, including the duration. The number was never the problem. The story about the number did all the damage.
What to Actually Do With This
Three things, if you're in the back half of this curve.
Stop measuring the wrong event. If you're tracking anything, track how present you were, not how long penetration lasted. Those two variables are inversely related more often than not, which should tell you something about which one to optimize.
Expand the encounter. The research gap between 5.4 minutes and 13.4 minutes closes from one direction only — by building a longer runway and a longer landing, not by extending the middle. This is the single highest-leverage change available to most couples, and it gets easier with age, not harder, because you have more skill and less urgency than you did at twenty-five.
Treat physical changes as logistics. Erections change. Lubrication changes. Recovery time changes. Nearly all of it is manageable with some combination of medical input, more deliberate arousal, lubricant, and different positioning. None of it is a referendum on your worth. I've written specifically about what sex can look like in your fifties, and the short version is: different, not diminished.
I'm 47. By the Waldinger numbers I've got a few years before I move into the bottom bracket. And I can tell you with complete confidence that my intimate life right now is dramatically better than it was at 27, when the number was higher and I had no idea what I was doing.
The watch was never measuring the thing that mattered. In running, at least it measured something. Here, it measures the least interesting variable in the room.
Brittney and I built Coelle because presence is the actual variable — and it's much easier to access when something else is holding the structure. Guided audio puts both partners in the same attentional state at the same time, which is precisely what the stopwatch can never give you.
If duration anxiety, performance pressure, or changes you've noticed in your forties or fifties are affecting your intimate life, that's something I work on directly with clients. Book a free discovery call here.
Sources: Waldinger MD et al., "A Multinational Population Survey of Intravaginal Ejaculation Latency Time," The Journal of Sexual Medicine, 2005. Bhat GS & Shrivastava A, "Time to Orgasm in Women in a Monogamous Stable Heterosexual Relationship," The Journal of Sexual Medicine, 2020. Twenge JM et al., "Declines in Sexual Frequency among American Adults, 1989–2014," Archives of Sexual Behavior, 2017. Lindau ST et al., "A Study of Sexuality and Health among Older Adults in the United States," The New England Journal of Medicine, 2007.
