Tip · Nutrition

Raising testosterone naturally

Q: How much testosterone does poor sleep cost me?

In the laboratory study, the daytime testosterone level after eight nights of five hours of sleep each was 16.5 instead of 18.4 nanomoles per liter, measured between 8 a.m. and 10 p.m. The difference was greatest in the afternoon and early evening. The sample was ten healthy young men – small, but under controlled laboratory conditions and with a clear direction.

Q: Does strength training raise testosterone levels?

Shortly after training, yes, but this rise achieves nothing. Two studies have separated this cleanly: in one, the same man trained one arm with and one without an accompanying hormone rise for 15 weeks, and both arms grew equally. In the other, there was no relationship whatsoever between the hormone rise and gains in strength or muscle in 49 men over twelve weeks. Strength training works, but via local signals in the muscle.

Q: Does zinc help?

In zinc deficiency, yes; otherwise not detectably. In nine older men with marginal zinc deficiency, testosterone levels rose from 8.3 to 16.0 nanomoles per liter after three to six months of supplementation. Conversely, it fell drastically in four young men during 20 weeks of a low-zinc diet. In the study in men with already normal baseline testosterone, zinc changed nothing.

Q: What about boron?

The only randomized study in men, seven weeks in 19 bodybuilders, found no effect on testosterone, lean mass or strength. The often-cited rise comes from a study in twelve postmenopausal women aged 48 to 82. In addition: the German Federal Institute for Risk Assessment proposes a maximum level of 0.5 milligrams of boron per recommended daily intake for food supplements. The figure common in advice guides is twelve times that, without any proof of efficacy for it.

Q: Do I have to cut out alcohol completely?

The data do not justify that. A meta-analysis of 17 observational studies found no difference in testosterone levels between drinkers and non-drinkers. A review article differentiates: acutely, small to moderate amounts raise the level, large amounts lower it, and chronic heavy consumption has a clearly negative effect. Cutting down sharply if consumption is high is justified; cutting it out completely is not.

Q: Is the 80 percent of the leverage correct?

No, this figure has no source. There are human data for the individual building blocks, but not for their interplay. No study has tested sleep, strength training, zinc, boron, magnesium, body fat percentage and abstaining from alcohol as a package against a control condition.

Strength training, sleep, zinc in case of deficiency: there are data for the building blocks. The often-cited “80 % of the leverage” is an estimate.

Strength training, sleep, body fat, zinc: the bundle sold under the heading “raising testosterone naturally” consists of parts with very different levels of evidence. This page sorts them by strength of evidence.

In short

Three building blocks have human data: sleep, body fat and dietary fat intake. Eight nights of five hours of sleep lowered daytime testosterone levels from 18.4 to 16.5 nmol/l. Weight loss raised total testosterone by 2.87 nmol/l across 24 studies on a calorie-reduced diet, and the greater the weight loss, the greater the rise. A low-fat diet lowered total and free testosterone (SMD −0.38 and −0.37 respectively). Zinc only works in case of deficiency, for boron and magnesium there is no proof in men, and the often-cited 80 percent of the leverage has no source. Strength training remains worthwhile – but not via the hormone surge after the session.

What is behind it

Testosterone is produced from cholesterol in the Leydig cells of the testes, controlled via LH from the pituitary gland. The level follows a daily rhythm with a peak in the morning, and this rise is coupled to sleep. That is why sleep restriction acts directly on the daytime average – in the laboratory, the difference was most pronounced between 2 p.m. and 10 p.m., while cortisol remained comparable between conditions.

In obesity, a separate mechanism is at work: adipose tissue expresses aromatase, which converts testosterone into estradiol, and this inhibits the release of LH via negative feedback. The meta-analysis on weight loss describes exactly this reversal – with weight loss, estradiol fell and the gonadotropins rose.

What does not work

The most popular notion in this field has been refuted: that strength training works via the acute rise in testosterone. Twelve young men trained their elbow flexors for 15 weeks, each arm in a different hormonal environment – one in isolation, the other trained identically but immediately followed by extensive leg training that significantly increased growth hormone, IGF-1 and testosterone (p < 0.001).

Muscle cross-sectional area increased by 12 % and 10 % respectively, with no difference between the conditions (p = 0.25). In 49 strength-trained men over twelve weeks, there was likewise no correlation whatsoever between the acute rise in any of the supposedly anabolic hormones and the change in strength or muscle mass. Muscle growth runs via local signals, not systemic ones.

Zinc, boron, magnesium

Zinc is the only micronutrient with a clear chain – but only via deficiency. In four healthy young men, serum testosterone fell from 39.9 ± 7.1 to 10.6 ± 3.6 nmol/l (p = 0.005) during 20 weeks of dietary zinc restriction. In nine older men with marginal zinc deficiency, it rose from 8.3 ± 6.3 to 16.0 ± 4.4 nmol/l (p = 0.02) after three to six months of supplementation. In 15 men with normal baseline testosterone, by contrast, nothing changed.

Boron fails. In the only randomized study in men, 19 bodybuilders received boron or placebo for seven weeks; plasma boron levels rose measurably, but the analysis of variance showed no effect on total testosterone, free testosterone, lean mass or strength. The often-cited rise comes from a study in twelve postmenopausal women aged 48 to 82. For magnesium, there is only one small intervention study in taekwondo athletes and sedentary people, with no replication to be found.

What is well supported

Sleep is the most directly measured: eight nights of five hours compared with a well-rested reference condition lowered daytime testosterone levels from 18.4 (3.8) to 16.5 (2.8) nmol/l (p = 0.049), in ten healthy men in the laboratory. Body fat is the most broadly supported: across 24 studies, total testosterone rose by 2.87 nmol/l (95 % CI 1.68 to 4.07) on a calorie-reduced diet and by 8.73 nmol/l (95 % CI 6.51 to 10.95) after bariatric surgery. And a low-fat compared with a high-fat diet lowered total testosterone (SMD −0.38; p = 0.04) and free testosterone (SMD −0.37; p = 0.005) across six intervention studies with 206 participants.

What the studies show

One week of sleep restriction in the laboratory

Ten healthy men, mean age 24.3 years, first slept eight hours at home for a week and then spent eleven days in the laboratory: three nights with ten hours of sleep opportunity as a reference, followed by eight nights with five hours. During the waking period common to both conditions, from 8 a.m. to 10 p.m., the values after restriction were lower, most clearly in the afternoon and evening.

Weight loss and testosterone

Meta-analysis with 24 studies included out of 266 screened, in which diet, physical activity or bariatric surgery led to weight loss. The extent of weight loss was the best predictor of the rise in the regression. Limitation: it is a before-and-after comparison across heterogeneous studies, not a pooled analysis of randomized comparisons with a control group.

The post-training hormone rise put to the test

Twelve healthy young men trained their elbow flexors for 15 weeks on separate days under two hormonal conditions. The primary endpoint was muscle cross-sectional area: plus 12 % without and plus 10 % with the hormone rise, no difference (interaction p = 0.25).

What the studies do not show

The figure “80 percent of the leverage” has no source. A joint effect of the bundle of strength training, sleep, zinc, boron, magnesium, body fat percentage and abstaining from alcohol has never been measured; there is no study that has tested these measures together against a control condition. The mark “body fat below 20 percent” is also set, not measured: what is supported is a continuous relationship between the extent of weight loss and the rise in testosterone, not a threshold.

On alcohol, the observational data are neutral: a meta-analysis of 17 observational studies found a standardized mean difference of 0.03 (95 % CI −0.11 to 0.16) between drinkers and non-drinkers, that is, no difference, nor for FSH and LH. And the add-on “especially REM” for sleep is not covered – the study manipulated sleep duration, not individual sleep stages.

How to do it

Ordered by strength of evidence: sleep first, because it is the most directly measured intervention on this list. Then body fat via a calorie-reduced diet, with the effect growing with the extent of weight loss. Then fat intake, which should not be too low; the data base for this is six intervention studies with 206 participants. Zinc is only worthwhile in case of deficiency. Strength training belongs on the list, but via the detour of body composition and not via the hormone surge after the session. For boron there is no proof of efficacy in men, and the maximum level for food supplements proposed by the BfR is 0.5 mg per recommended daily intake – the 6 mg circulating in advice guides is twelve times that.

Safety

The most important point is not a dosage point: anyone who suspects a low testosterone level needs a medical evaluation, not self-medication. Hypogonadism is a diagnosis with clear criteria and partly treatable causes; the measures here improve building blocks and do not replace diagnostics. For zinc, the SCF and EFSA name a tolerable upper intake level of 25 mg per day for adults, and the BfR proposes 6.5 mg per recommended daily intake for food supplements – relevant because zinc from food comes on top and a permanently high intake interferes with copper absorption. For boron, the EFSA value is 10 mg per day; even with extreme occupational exposure, a field study in 212 workers found no boron-associated effects on sperm parameters, FSH, LH or total testosterone. At this order of magnitude, boron is apparently not acutely dangerous – but in men it does not work either.

BK-Score Thin human evidence

Human evidence		5
Mechanism		7
Safety data		8
Hype gap		4
Track record of use		9

For the individual building blocks – sleep, strength training, zinc in case of deficiency – there are human data in each case. The figure “80 % of the leverage”, by contrast, is a free estimate; a joint effect of the bundle has never been measured.

The score rates the state of knowledge, not the effect. “Safety data 9” means well studied – not harmless.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)

Frequently asked questions about raising testosterone naturally

How much testosterone does poor sleep cost me?