Testosterone levels by age: the real reference ranges, honestly explained

Search “testosterone levels by age” and you’ll be handed a tidy chart promising a precise normal number for your decade. The honest version is less tidy and more useful: there is no separate official “normal range” for each age band. Clinical labs use a single adult reference range, the best-studied version of that range comes from a deliberate harmonization effort, and the age effect everyone fixates on is real but modest — and partly about weight, not age alone. This page lays out the actual numbers, where they come from, and the one rule that the charts always skip: a number by itself diagnoses nothing.

The actual reference range: where 264–916 comes from

Most lab reference ranges for testosterone were historically built locally, on different assays, in different populations — which is why two labs could call the same blood result “normal” or “low.” To fix that, a 2017 study pooled four large cohort studies across the United States and Europe and harmonized their testosterone measurements to a common standard. In a reference sample of healthy nonobese men aged 19 to 39, the resulting normal range for total testosterone was approximately 264 to 916 ng/dL (the middle 95% of values).^[1]That figure — not a per-decade chart — is the closest thing to a modern, standardized “normal” for adult men, and it’s why a single number is best read against this band rather than a decade-specific one.

Two caveats come attached. First, this is a range for young, healthy, nonobese men by design, which is what makes it a clean reference point — not a promise that every 60-year-old should land in the upper half of it. Second, an individual lab’s printed range may differ, because reference intervals still vary by assay and by the population a lab calibrates against. Always read your result against the range your own lab reports.

“By age”: the decline is real, modest, and not purely about age

Testosterone does fall as men get older. The cleanest evidence is longitudinal — following the same men over time rather than comparing different men at different ages. The Baltimore Longitudinal Study of Aging tracked healthy men and found total and free testosterone declined progressively with age, on the order of roughly 1% per year in total testosterone after the third-to-fourth decade, with free testosterone falling somewhat faster because of changes in binding proteins.^[2] So the “by age” instinct isn’t wrong — it’s just smaller and slower than the dramatic charts imply.

The more important nuance is why levels look lower in older men. A great deal of what reads as age-related decline actually tracks with weight gain, chronic illness, and medicationsthat accumulate over a lifetime — not the passage of time itself. Obesity in particular is a powerful, partly reversible driver: a meta-analysis found that losing weight raises testosterone in proportion to the weight lost, reverting much of the obesity-associated suppression.^[3]That’s the part the “normal level for my age” framing erases — a low-ish number in a 50-year-old may say more about waistline and sleep than about birthdays. We unpack that lever in TRT vs natural testosterone.

Total vs free testosterone, and the SHBG twist

“Testosterone level” usually means total testosterone — everything in the blood, bound and unbound. But most circulating testosterone is stuck to carrier proteins, chiefly sex hormone–binding globulin (SHBG) and albumin. Only the small unbound and loosely bound portion — free and bioavailable testosterone — is active.

This matters with age because SHBG tends to rise as men get older, which can keep total testosterone looking acceptable while the free fraction quietly falls. That’s exactly the pattern the longitudinal data show, with free testosterone declining faster than total.^[2] In practice, when total testosterone is borderline or when SHBG is abnormal, clinicians look at free testosterone. A practical wrinkle: most labs estimate free testosterone by calculation from total T, SHBG and albumin, or by a specialized assay — the cheap “direct” free-T immunoassays are widely considered unreliable, which is why a single free-T number from a basic panel deserves skepticism.

The cutoff number — and why it isn’t a diagnosis

Guidelines do offer a working threshold. The American Urological Association’s testosterone deficiency guideline uses a total testosterone of below 300 ng/dL, confirmed on two early-morning measurements, as the level that defines low testosterone for diagnostic purposes.^[4]It’s a useful flag — but it is a flag, not a finding. The critical point, stated plainly by the Endocrine Society, is that you diagnose hypogonadism only in men with symptoms and signs of testosterone deficiency AND unequivocally, consistently low testosterone on a repeat morning test.^[5] Both halves are required.

That repeat-and-morning detail isn’t pedantry. Testosterone follows a daily rhythm, peaking in the morning, and a single reading bounces around enough that borderline values often normalize on a second draw. The large European Male Aging Study went further, showing that “late-onset hypogonadism” is best defined by the combination of specific sexual symptoms with a low testosterone threshold — symptoms without low T, or low T without symptoms, didn’t reliably mark a treatable condition.^[6] In other words: a number just under a cutoff, in a man with no symptoms, isn’t a diagnosis — and chasing it with treatment is how men end up on therapy they don’t need.

So what should you do with your number?

Read it against the adult reference range your lab prints, not a decade chart. Note whether it was drawn in the morning and fasting — if not, it’s a weak data point. If it’s low or borderline and you have real symptoms (low libido, erectile changes, persistent fatigue that isn’t explained by sleep or stress), the next step is a repeat morning test plus a look at SHBG and free testosterone, not a leap to treatment. And if a reversible driver is in play — excess weight, poor sleep, certain medications — addressing it can raise the number on its own.^[3]The full workup is laid out in low testosterone treatment.

Only when low levels are confirmed alongside symptoms does the treatment conversation begin. What therapy actually does — and doesn’t — is covered in our TRT evidence review, with its side-effect trade-offs and the diagnosis-first path to getting TRT. For men who want to raise their own testosterone while preserving fertility, the alternative is enclomiphene. And to see where testosterone-related interventions sit against the wider longevity field on actual evidence, use our longevity evidence matrix.

The honest bottom line

There is one adult reference range, not a chart per decade. The best-standardized version puts normal total testosterone in healthy young men at roughly 264–916 ng/dL,^[1]levels do drift down with age at about 1% a year,^[2] and a meaningful slice of that drift is weight and illness rather than age itself.^[3] A total testosterone under 300 ng/dL is a common flag,^[4] but the flag isn’t the diagnosis: hypogonadism takes low levels on repeat morning tests plus symptoms.^[5][6]Treat the whole picture, never the number alone.