← Back to blog

Strategy

Creatine: What It Actually Does and What It Does Not

Effect sizes from the meta-analyses, where creatine works hardest, and why loading is not required.

· 4 min

Key takeaways

  • The effect is small but consistent: 0.17 for body composition, 0.42 for upper-body exercise.
  • A meta-analysis of 53 studies reported a global effect size of 0.317 for upper-limb strength.
  • Loading shortens time to effect but does not raise the ceiling — 3–5 g daily is enough.
  • You will see it in maintained reps across sets, not in a new personal record.

The Effect Is Small, Consistent, and Biggest Up Top

Creatine is the best-studied supplement in strength training and also the one whose effect is most often oversold. A meta-analysis found small but significant effect sizes, including 0.17 for body composition, with a larger effect for upper-body exercise (0.42) than lower-body (0.21) or total-body work (0.13) [1].

A separate meta-analysis of 53 studies on upper-limb strength reported a global effect size of 0.317, concentrated in efforts under three minutes and largely independent of dose or protocol [2]. Two independent data sets, the same picture: a real, moderate benefit.

The mechanism explains this distribution well. Creatine raises muscle phosphocreatine stores, meaning a greater capacity to rebuild ATP quickly between reps and sets. The largest benefit therefore lands on short, repeated efforts — which dominate upper-body work.

The same mechanism explains why the effect is hard to see in a single heavy attempt. One maximal rep does not deplete the phosphagen system enough for its larger capacity to matter.

Where creatine actually shows up

effect size vs placebo

Upper-body exercise0.42Loading protocol (body comp.)0.26Lower-body exercise0.21Body composition (overall)0.17Total-body exercise0.13Maintenance dose (body comp.)0.04

Branch (2003) meta-analysis. All the effect sizes are small, and they are not all the same outcome — exercise-performance effects and body-composition effects are listed side by side for scale. The pattern is consistent: real but modest, and biggest on short, upper-body efforts.

Source: Branch JD. Int J Sport Nutr Exerc Metab. 2003;13(2):198-226.

Loading Is Not Required

In the same analysis, the body-composition effect was larger after a loading protocol (0.26) than on a maintenance dose (0.04) [1] — but that comparison mostly measures how fast the muscle saturates, not the long-run outcome. Loading shortens the time to full effect; it does not raise the ceiling.

Practically that means 3–5 g a day with no loading phase reaches the same place in a few weeks. Loading is worth it only when you need the effect by a specific date.

It is also worth remembering what these numbers do not say. An effect size of 0.3 is a few percent on the bar, not the difference between a stall and progress — your training dose still comes from volume and recovery.

It is worth settling the practical questions that generate the most confusion. Timing does not matter, because the effect depends on muscle saturation rather than blood concentration at any moment. Supplementing on rest days makes sense for the same reason.

Hydration deserves attention, though not for the usual reason. Creatine increases intracellular water content, so fluid needs rise slightly — an argument for drinking normally rather than for any special protocol.

Where You Will See It in Your Data

Creatine usually does not change your e1RM on a single heavy set — it changes how many reps you hold on the second and third set. If you are looking for the effect, watch rep drop-off across sets at the same load rather than your best single.

The second visible effect is a faster gain in body weight over the first two weeks, driven mainly by water retention inside the muscle. Worth knowing so you do not read it as new tissue, or as a failing fat-loss phase.

The last thing worth settling is the order of priorities. An effect size around 0.3 is real, but smaller than the difference between a plan with 10 and 16 weekly sets per muscle, or between six and eight hours of sleep.

Creatine is a good addition to a working system rather than a way around its gaps. Supplementing while missing three sessions a month changes less than recovering those three sessions would.

FAQ

Do I need to cycle off creatine?

There is no convincing basis for it in the data. A break simply means re-saturating the muscle when you return.

Does creatine work in a deficit?

The mechanism does not change in a deficit, so the help with set performance remains. Only the scale reading shifts because of water retention.

Which form of creatine is best?

Monohydrate is the best studied and the cheapest. More expensive forms lack data that would justify the price difference.

References

  1. [1]Branch JD. Effect of creatine supplementation on body composition and performance: a meta-analysis. Int J Sport Nutr Exerc Metab. 2003;13(2):198-226. PubMed 12945830
  2. [2]Lanhers C, et al. Creatine supplementation and upper limb strength performance: a systematic review and meta-analysis. Sports Med. 2017;47(1):163-173. doi:10.1007/s40279-016-0571-4

Bibliographic sources via PubMed.

Want to turn this into your own plan?

Compare Load7 plans, review the free workflow, and see when Premium adds the most value for planning and analysis.

We use cookies

We use essential cookies for login and app functionality, and optional analytics cookies to improve the product.

Learn more in Cookie Policy