Normal Testosterone Levels by Age: The 300 Cutoff Came From Older Men
There isn’t one.
There is a range, it is enormous, it depends on your age and the time of day and which machine ran the sample — and the single number most labs draw the line at was worked out on men older than the ones it is now used on.
Let us start with the range, because it is the part almost nobody sees.
What “normal” is, and where it comes from
In 2017, researchers took 9,054 men from four large cohort studies in the US and Europe and cross-calibrated every assay against a reference method at the CDC — so that a reading from one study meant the same thing as a reading from another. [1]
For healthy non-obese men aged 19 to 39, the middle 95% ran from 264 to 916 ng/dL. The median was 531. [1]
Sit with those numbers. The man at the top of the normal range has roughly three and a half times the testosterone of the man at the bottom. Both are normal. Neither has a problem.
That is because a reference range is not a health target. It is a description of where most of a reference population landed — conventionally the middle 95%. Which means, by construction, that one healthy man in twenty falls outside it. Not because anything is wrong with him. Because that is what “middle 95%” means.
So where does 300 come from?
Three hundred nanograms per decilitre is the number in widest clinical use in the US, and it is a reasonable line for the population it was drawn from. The problem is who it gets applied to.
In 2022, a team at Michigan asked a simple question: what if you took the American Urological Association’s own definition of a normal level — the middle third of the population — and calculated it separately for each age group of young men? [2]
They used NHANES, the nationally representative US health survey, kept only morning blood draws, and excluded men on hormonal medication. 1,486 men aged 20 to 44.
| Age | Middle third | Low, for that age |
| 20–24 | 409–558 | below 409 |
| 25–29 | 413–575 | below 413 |
| 30–34 | 359–498 | below 359 |
| 35–39 | 352–478 | below 352 |
| 40–44 | 350–473 | below 350 |
| The cutoff in common use | — | below 300 |
A 27-year-old with a reading of 380 ng/dL is comfortably above the standard cutoff. He is also below the bottom of the middle third for men his own age. Both of those statements are true at once, and only one of them appears on his lab report.
The authors put it plainly: it is “surprising that young men are evaluated for testosterone deficiency with the same cutoff of 300 ng/dL that was developed from samples of older men.” [2]
Now the part that genuinely surprised us
Everyone knows testosterone falls with age. It is the premise of an entire industry.
In 2024, researchers went to the UK Biobank and looked at 197,883 men — comfortably the largest analysis of this question ever run. [3]
In men without metabolic syndrome, average testosterone by decade came out at roughly 358 ng/dL in their forties, 354 in their fifties, and 353 in their sixties. In men with metabolic syndrome it was about 308, 305 and 307 across the same decades — lower throughout, and just as flat.
Two things moved the number. Having metabolic syndrome moved it. And the time of day of the blood draw moved it, from about 367 down to 315 ng/dL. Season moved it slightly too.
Age did not.
Their conclusion, in their own words: “We found significant variation in serum testosterone according to the presence of metabolic syndrome and time of laboratory draw, but not according to age. These data challenge the prior dogma of age-related hypogonadism.” [3]
(Those are our conversions from the paper’s nmol/L into the ng/dL your lab reports. And a caution the paper’s own argument demands: UK Biobank used an immunoassay, so these absolute values are not interchangeable with mass-spectrometry numbers from other studies. Compare the pattern, not the digits.)
Except a very good study says the opposite
We are not going to pretend that settles it, because it doesn’t.
The Massachusetts Male Aging Study followed 1,156 men for seven to ten years and measured them repeatedly. Within the same men, total testosterone declined about 1.6% a year, and bioavailable testosterone 2 to 3% a year. [4]
So which is it?
The interesting thing is what happens when you look at what each study could see. The UK Biobank analysis is a snapshot: it compares different men of different ages at one moment. The Massachusetts study followed the same men through time. And the Massachusetts authors flagged something odd — their longitudinal decline was steeper than their cross-sectional one, which should not happen. Their explanation: “incident poor health may accelerate the age-related decline in androgen levels.” [4]
In the same cohort, men in apparent good health — no chronic illness, no prescription medication, not obese, not heavy drinkers — carried androgen levels 10 to 15% higher. [4]
Which points at the same thing the Biobank found. It is not that the birthdays are harmless. It is that what usually arrives alongside them does most of the work.
The single most useful sentence in this literature
A follow-up study of that same Boston cohort, now with three rounds of measurement, tried to separate ageing from everything that happens while you age. [5]
The finding: a rise in body mass index of 4 to 5 points, or the loss of a spouse, was associated with a fall in testosterone comparable to about ten years of ageing. [5]
Gaining thirty pounds does to your testosterone roughly what a decade does. So does bereavement.
The authors’ conclusion is the reason this page exists: “comorbidities and lifestyle influences may be as strongly associated with declining T levels as is aging itself,” and therefore “age-related hormone decline may be decelerated through the management of health and lifestyle factors.” [5]
Your testosterone is not simply a readout of how old you are. A meaningful part of it is a readout of how you are doing.
What a low number actually predicts
Here is the last piece, and it reframes the question.
The European Male Ageing Study surveyed 3,369 men aged 40 to 79 across eight countries, measured morning testosterone by mass spectrometry, and asked which symptoms actually travelled with a low level. It then split the data into training and validation halves to check its own answer. [6]
Several symptoms were related to testosterone: poor morning erections, low sexual desire, erectile dysfunction, inability to do vigorous activity, depression, and fatigue.
But only three of them hung together as a syndrome. In the paper’s words: “only the three sexual symptoms had a syndromic association with decreased testosterone levels.” [6]
Poor morning erections, low sexual desire, erectile dysfunction. Those three.
Fatigue, low mood and feeling physically flat — the complaints that actually walk into a men’s health clinic — were related to the level but did not form a syndrome with it. They are far too common, and far too easily caused by everything else in a life, to point at a hormone on their own.
The thresholds where those sexual symptoms started appearing sat between 230 and 370 ng/dL. [6] Not a line. A zone.
The verdict
“Below 300 means you are low” is not supported as a universal rule, and the strongest objection to it comes from urology’s own definition of normal applied properly by age.
What is true is narrower and more useful. Normal is a wide band, roughly 264 to 916 ng/dL in healthy young men. Where you sit inside it says very little on its own. The cutoff your lab prints was not built for your age if you are under 40. Time of day moves your result more than a decade of life does. And the symptoms that reliably travel with genuinely low testosterone are three sexual ones — not tiredness.
If you are holding a number and wondering what it means, the honest answer is that it means very little without three other things: what time it was taken, whether it was confirmed on a second morning, and whether you have symptoms that the evidence actually links to it.
And if the number is lower than you would like, the most evidence-backed thing you can do about it is not a supplement. Two studies of the same cohort say that carrying less weight and being in better general health is worth about ten years.
Part 2 of a series on testosterone. Part 1 asked whether you should get tested at all. Next: free versus total testosterone, and why the number your lab prints may not be the one that matters.
Related: should you get your testosterone tested · ashwagandha benefits · does lack of sleep cause weight gain · is creatine safe
