The guide

What creatine does, at its real size

By GyroscopeUpdated November 14, 202615 sources

With lifting, creatine adds a modest, real edge. Strength rose 20% on creatine against 12% without it across 22 studies, and lean mass rose 2.4 lb (1.1 kg) more than lifting alone across 35 trials. Without training it does almost nothing. Its effects on memory are small, and its effect on mood is below what people notice.

Creatine gets sold two ways. On one side it’s the miracle powder, good for muscles, brain, mood, aging, bones and recovery. On the other it’s just water weight that does nothing. Both are easy to believe, because the muscle effect is real but small, and small effects are easy to inflate or dismiss.

The way through is to look at each claim with its number beside it. Some hold up well, some barely, and a few don’t hold up at all.

It works like a small raise

An extra rep on each hard set is like a small raise at work. On any one payday you’d hardly notice it. Over a year it adds up to real money, as long as you keep showing up. Creatine’s raise only gets paid on days you train.

How much stronger does it make you?

A little, and reliably. In 2003, Rawson and Volek pooled 22 studies of creatine with weight training. Strength rose 20% on creatine against 12% on placebo, an 8 point difference. Reps at the same share of your maximum rose 26% against 12%. On a set of 10, that’s about one or two more reps. The spread between studies was wide, with bench press gains ranging from 3% to 45%.

Two larger reviews from 2015 and 2017 pooled 60 studies of leg strength and 53 of upper body strength, counting only efforts under three minutes. Creatine helped both, with what statisticians call a small effect. In plain terms, it’s real, and it shows up over months of training more than in any one session.

How much muscle does it add?

About 2.4 lb (1.1 kg), with lifting. A 2022 review pooled 35 trials and 1,192 people. With resistance training, creatine added 1.10 kg (2.4 lb) of lean mass over training alone, with a range of 0.56 to 1.65 kg. With no exercise, it added 0.03 kg, which is nothing.

The same review found men gained 1.46 kg (3.2 lb) and women 0.29 kg (0.6 lb), and the women’s number was too small to be sure of. A 2026 pooling of seven trials in women after menopause found 0.37 kg (0.8 lb) of extra lean mass and 7.5 kg (16.5 lb) more on the leg press, mostly in trials that used at least 5 g a day with lifting. Creatine for women has more.

People in their late 50s to 70 did well too. Across 22 trials of older adults lifting two or three days a week, creatine added 1.37 kg (3 lb) of lean tissue and more strength on the chest and leg press. Creatine after 50 covers it.

The best case against it

Some of that lean mass is water. Body scans count water as lean tissue, and creatine brings water into muscle. A 2025 trial showed how big that effect can be. After one week on 5 g a day, people had 0.51 kg (1.1 lb) more lean mass on a scan before they’d lifted anything. Then both groups lifted for 12 weeks and gained about 2 kg each, with no extra gain on creatine.

The answer to that comes from studies that looked at the muscle itself. In a 2003 trial, Burke’s group took muscle samples before and after 8 weeks of lifting, and the fast-twitch fibers grew more in people on creatine than on placebo. Fiber size measured in a muscle sample is much harder to explain with water than a scan is, though cell water can swell fibers a little too. The fair summary is that some of the early lean mass is water, and some of the long-run gain is real muscle.

Does creatine work without exercise?

Barely, by the pooled evidence. Across 35 trials, lean mass didn’t move without exercise.

A 2026 trial got a lot of attention for finding otherwise. 64 adults aged 45 to 65 finished 12 weeks on either 10 g of creatine a day or a placebo, and each chose beforehand whether to join an exercise and diet program. In the group that didn’t exercise, creatine added about 1 kg (2.2 lb) of lean tissue. That group had 13 people at the end, it took twice the usual dose, and the scan it used counts water as lean tissue. It’s an interesting result, but it doesn’t outweigh 35 trials.

Does it help endurance?

It’s not the reason to take it. The strength trials only counted efforts under three minutes, which is where phosphocreatine does its job. The International Society of Sports Nutrition lists possible benefits for longer efforts, like helping muscles store glycogen and carrying energy from the mitochondria, but those are a smaller and less tested part of the story.

Does creatine help your brain?

A little. About 5% of your creatine is outside muscle, much of it in the brain. A 2024 review pooled 16 trials and 492 people. Memory improved slightly, as did attention and processing speed. Overall thinking and planning didn’t change.

The most dramatic result came from a lab in Germany. In 2024, 15 young adults took one large dose, 0.35 g per kg, during a night without sleep. Their processing speed held up better than on placebo, with the biggest effect about four hours after the dose. That dose is about 28 g at 176 lb (80 kg), a trial dose and not advice, and sleep is still the fix for no sleep.

One popular idea didn’t replicate. A 2023 trial of 123 people, half of them vegetarian, found that vegetarians didn’t get any more brain benefit from 5 g a day than meat eaters did.

Does creatine help depression?

Maybe a little, but not enough to notice. A 2025 review pooled 11 trials and 1,093 people. Creatine moved depression scores by 2.2 points on the scale doctors use, below the 3 points that counts as a difference people notice. The researchers rated the evidence very low and wrote that the true effect “may be trivial or null”. If you have depression, creatine isn’t a treatment, and never change a medicine because of it.

Who gets the most from it?

People who start with low stores and people who train. Vegetarians start lower, and in a 2003 trial they gained more stored creatine, lean tissue and total work than meat eaters on the same dose and program. People who don’t train get almost nothing, whatever they eat.

Every claim, side by side