I really enjoyed this essay and feel I have something to add since you mentioned the "trouble" of taking daily creatine: based on everything I have been able to find, there's minimal downside to putting your creatine in the hot beverage you probably drink every morning. I've been putting a scoop into my first hot drink of the day for a while now and haven't noticed a difference in taste or texture.
Creatine is a protein, and the temperature of a hot drink and the amount of time the creatine is in it for before being consumed (<15 minutes) mean there is minimal protein breakdown (since, after all, people don't lose their mind about protein/creatine loss for braising, grilling, pressure cooking, smoking, or otherwise cooking meat for a long time or at a high temperature).
and https://www.nature.com/articles/s41598-024-54249-9, showing that single-dose creatine significantly improves cognitive performance in subjects exposed to 21 hours sleep deprivation compared to placebo (n = 15)
"Say you think there’s a 50% chance creatine increases IQ by 1 point and a 50% chance it’s useless. Is it actually worth the trouble of taking 5 grams of creatine every day for an expected increase of 0.5 IQ points? I’m not sure."
But don't the clear strength benefits justify basically anyone taking creatine, and if there are marginal cognitive benefits on top, then great?
That's a totally reasonable view. I just don't want to make any assumptions about anyone else's values! (I'd imagine that only a minority of the population regularly do any activities where creatine would make much difference?)
I was writing this comment with the median reader of your blog in mind, who, presumably, engages in some form of strength training if they’re interested in health optimization enough to read essays like this one
Just to weigh in: after creatine supplementation my ability to concentrate greatly increases. 5 g provides a small boost; at around 20 g I feel much more energetic. The effect is noticeable around 30 minutes after a single 5 g dose. On days when I don't supplement creatine, I have noticeable brain fog.
This doesn't rule out a placebo. And I don't have any evidence that creatine improves my cognitive performance other than my reported internal states. Still, I think the effect is real. Creatine makes me feel better.
As a side note, I'm likely not neurotypical. My ability to focus has always been very poor. I wonder if creatine improves attentiveness in those with ADHD?
Very interesting. There are several theories of dementia that have to do with energy failing to reach neurons. Here's hoping someone tries to inject creatine past the BBB.
Just sharing a mechanistic hypothesis for you to consider as to why some report cognitive benefits, I don't have any interesting data beyond what you already shared.
Have you evaluated the use of large single doses of Creatine in cases of sleep deprivation? Even if modest continued supplementation doesn’t yield extra cognitive benefits— being able to mega-dose away some of the cognitive loss from lack of sleep seems hugely valuable (especially given that creatine is far easier to get than Modafinil).
Anecdotally, the megadoses have helped me when sleep deprived.
Also— anectdotally, ~10-15g per day continued supplementation I notice no particular increases in my processing abilities, but i do seem to have more sustained cognitive energy. I can work on hard-ish things seemingly far longer before my brain burns out.
Either way, it also makes working out easier and more intensive— so that probably also has an indirect advantage to boosting cognitive benefits if you can do better HIIT.
Or or or, it is simply that I remember to supplement creatine only when I am in a hypomanic and already energized state, which with me atleast, is the very likely cause.
I'm not aware of any experiments that tested anything like that, so I can only speculate. Mechanistically, you could argue that a large one-off dose might work better for the brain because it won't have time to downregulate cerebral production of creatine. On the other hand, transport between the body and brain seems quite limited. Regarding if it gives you the ability to work longer, I suppose it's possible, but the problem is that thinking hard just doesn't change energy usage to any significant degree. (And that the analogy to exercise isn't promising; creatine doesn't help (much) with endurance exercise.)
I suspect there's just not much we can say with confidence, just because we don't understand the mechanisms to say much with confidence and because (AFAIK) there are no long-term RCTs: There actually *are* a bunch of creatine RCTs that ran for 1-2 years, but they all seem to be in older adults from the general population, not weight lifters.
I've been running some self-experiments that use my own reaction time as a proxy for cognitive function, following a protocol developed by the late psychologist Seth Roberts. As a long-time vegan, I take in very little dietary creatine, so I expected creatine monohydrate to give me a noticeable boost in processing speed. It didn't. Supplementing had no measurable effect on my speed.
The graph below captures the result. Lower values are better, since they represent faster reaction times:
The same graph shows an earlier finding that surprised me: metformin made me faster. I don't have a full write-up of the creatine experiment yet, but I wanted to share the data here.
If anyone is interested in collaborating on this kind of work, I'd be glad to hear from you. There's a contact email on the site. These projects tend to be more fun and productive with more than one person involved.
> As far as I can tell, the list of normal nutrients that have been proven to increase strength is: protein, creatine, beta-alanine, the end.
A note on supplementation in general:
It's often assumed that supplements that provide short term increases in strength will eventually lead to longer terms increases in strength—because short term increases in workout and training volume should in theory translate to more volume stacked over time, which in theory translates to more adaptation. But this is a mere assumption, and I think it’s generally not well-founded: it’s not that hard to push into injury, burnout, or general exhaustion if you just do too much training volume, and you don’t need a supplement to do so. For example, it's well documented that caffeine can increase performance in long distance running, but runners don't supplement caffeine before all of their runs because they aren't limited by the amount of running they do, but how well they can recover and adapt to the running they actually do.
To me, it seems like the supplements that provide long-term benefits are those that assist in *recovery*, which is where all the gains are made anyway. For example, more protein means more muscle protein synthesis. Many anabolic steroids also work on the same principle, granting greater adaptations for the same amount of training stimulus (instead of allowing the user to dump more stimulus onto their body). And of course, the most anabolic natural supplement of all is sleep.
These points in no way refute anything you've said in this article, I just think it's important to note.
wait, would the cleanest test be sleep-deprived/vegetarian/older people, with brain creatine measured before and after, instead of just asking whether random healthy people feel sharper?
Ah nice, that’s the missing endpoint. I’d really want baseline brain creatine in the model too, because a null average could just be mixing actual brain responders with people whose brains barely changed.
Yeah, absolutely. I need to see scatterplots of Δ(brain creatine levels) vs. Δ(cognitive tests). However, I guess this would be pretty expensive due to the difficulty of measuring (really, estimating) brain creatine levels.
Yeah, even a noisy MRS substudy would be worth a lot. Baseline + Δbrain creatine would at least tell you whether the cognitive null is biology, psychometrics, or averaging over non-uptakers.
I understand the thought. A problem is that there's really two different questions, and both are interesting: (1) does a random health person see a benefit (the big payout, if proven), (2) does ANYONE show a benefit (the easier proof, but far less important demonstration). Depending on what your mandate/interest in, I could justify wanting to bias towards either one of those questions.
Yeah, fair split. I was thinking of the enriched group as a cheap mechanism test: if sleep-deprived/vegetarian/older people with low-ish brain creatine don't budge, I get much less excited about paying for the random-healthy-adult RCT.
I really enjoyed this essay and feel I have something to add since you mentioned the "trouble" of taking daily creatine: based on everything I have been able to find, there's minimal downside to putting your creatine in the hot beverage you probably drink every morning. I've been putting a scoop into my first hot drink of the day for a while now and haven't noticed a difference in taste or texture.
Creatine is a protein, and the temperature of a hot drink and the amount of time the creatine is in it for before being consumed (<15 minutes) mean there is minimal protein breakdown (since, after all, people don't lose their mind about protein/creatine loss for braising, grilling, pressure cooking, smoking, or otherwise cooking meat for a long time or at a high temperature).
You’ve got me here at 11pm bolting upright so I can drink some creatine and then reading further and lying back down then rinse and repeat…
Perhaps the purported effect is secondary to improved psychiatric health?
https://www.psychiatrypodcast.com/psychiatry-psychotherapy-podcast/episode-238-creatine-mental-health-benefits
Two studies to highlight:
https://doi.org/10.1523/JNEUROSCI.3113-14.2015, showing that ~20g/day creatine preserves cognitive performance after artificial hypoxia compared to placebo (n = 15)
and https://www.nature.com/articles/s41598-024-54249-9, showing that single-dose creatine significantly improves cognitive performance in subjects exposed to 21 hours sleep deprivation compared to placebo (n = 15)
"Say you think there’s a 50% chance creatine increases IQ by 1 point and a 50% chance it’s useless. Is it actually worth the trouble of taking 5 grams of creatine every day for an expected increase of 0.5 IQ points? I’m not sure."
But don't the clear strength benefits justify basically anyone taking creatine, and if there are marginal cognitive benefits on top, then great?
That's a totally reasonable view. I just don't want to make any assumptions about anyone else's values! (I'd imagine that only a minority of the population regularly do any activities where creatine would make much difference?)
I was writing this comment with the median reader of your blog in mind, who, presumably, engages in some form of strength training if they’re interested in health optimization enough to read essays like this one
Just to weigh in: after creatine supplementation my ability to concentrate greatly increases. 5 g provides a small boost; at around 20 g I feel much more energetic. The effect is noticeable around 30 minutes after a single 5 g dose. On days when I don't supplement creatine, I have noticeable brain fog.
This doesn't rule out a placebo. And I don't have any evidence that creatine improves my cognitive performance other than my reported internal states. Still, I think the effect is real. Creatine makes me feel better.
As a side note, I'm likely not neurotypical. My ability to focus has always been very poor. I wonder if creatine improves attentiveness in those with ADHD?
Very interesting. There are several theories of dementia that have to do with energy failing to reach neurons. Here's hoping someone tries to inject creatine past the BBB.
https://splittinginfinity.substack.com/p/links-33
Supplementing creatine spares under-methylated people 50% of their daily SAMe that would normally be spent on endogenous creatine synthesis.
Which is why it works so well to improve energy/cognitive performance on the elderly and sleep deprived (both groups are methyl constrained).
And on me!
I have MTHFR T/T homozygous genes which makes me methylfolate production run at 30% of that of a normal person.
What's your evidence that it works well on the elderly, sleep deprived, or yourself?
Just sharing a mechanistic hypothesis for you to consider as to why some report cognitive benefits, I don't have any interesting data beyond what you already shared.
OK, fair enough!
Have you evaluated the use of large single doses of Creatine in cases of sleep deprivation? Even if modest continued supplementation doesn’t yield extra cognitive benefits— being able to mega-dose away some of the cognitive loss from lack of sleep seems hugely valuable (especially given that creatine is far easier to get than Modafinil).
Anecdotally, the megadoses have helped me when sleep deprived.
Also— anectdotally, ~10-15g per day continued supplementation I notice no particular increases in my processing abilities, but i do seem to have more sustained cognitive energy. I can work on hard-ish things seemingly far longer before my brain burns out.
Either way, it also makes working out easier and more intensive— so that probably also has an indirect advantage to boosting cognitive benefits if you can do better HIIT.
Or or or, it is simply that I remember to supplement creatine only when I am in a hypomanic and already energized state, which with me atleast, is the very likely cause.
I'm not aware of any experiments that tested anything like that, so I can only speculate. Mechanistically, you could argue that a large one-off dose might work better for the brain because it won't have time to downregulate cerebral production of creatine. On the other hand, transport between the body and brain seems quite limited. Regarding if it gives you the ability to work longer, I suppose it's possible, but the problem is that thinking hard just doesn't change energy usage to any significant degree. (And that the analogy to exercise isn't promising; creatine doesn't help (much) with endurance exercise.)
https://www.nature.com/articles/s41598-024-54249-9
Study i read. Was convincing enough to me…
Interesting, thank you! I'd be interested too in differences between people that take creatine and people that don't, in the long term, for lifting.
I suspect there's just not much we can say with confidence, just because we don't understand the mechanisms to say much with confidence and because (AFAIK) there are no long-term RCTs: There actually *are* a bunch of creatine RCTs that ran for 1-2 years, but they all seem to be in older adults from the general population, not weight lifters.
I've been running some self-experiments that use my own reaction time as a proxy for cognitive function, following a protocol developed by the late psychologist Seth Roberts. As a long-time vegan, I take in very little dietary creatine, so I expected creatine monohydrate to give me a noticeable boost in processing speed. It didn't. Supplementing had no measurable effect on my speed.
The graph below captures the result. Lower values are better, since they represent faster reaction times:
https://www.self-experiments.org/wp-content/uploads/2026/07/metformin-creatine-graph-1.png
The same graph shows an earlier finding that surprised me: metformin made me faster. I don't have a full write-up of the creatine experiment yet, but I wanted to share the data here.
Ack, several times when writing this post I thought to myself that I should mention your results. My apologies for forgetting!
You can find my other self-experiments here: https://www.self-experiments.org/
If anyone is interested in collaborating on this kind of work, I'd be glad to hear from you. There's a contact email on the site. These projects tend to be more fun and productive with more than one person involved.
> As far as I can tell, the list of normal nutrients that have been proven to increase strength is: protein, creatine, beta-alanine, the end.
A note on supplementation in general:
It's often assumed that supplements that provide short term increases in strength will eventually lead to longer terms increases in strength—because short term increases in workout and training volume should in theory translate to more volume stacked over time, which in theory translates to more adaptation. But this is a mere assumption, and I think it’s generally not well-founded: it’s not that hard to push into injury, burnout, or general exhaustion if you just do too much training volume, and you don’t need a supplement to do so. For example, it's well documented that caffeine can increase performance in long distance running, but runners don't supplement caffeine before all of their runs because they aren't limited by the amount of running they do, but how well they can recover and adapt to the running they actually do.
To me, it seems like the supplements that provide long-term benefits are those that assist in *recovery*, which is where all the gains are made anyway. For example, more protein means more muscle protein synthesis. Many anabolic steroids also work on the same principle, granting greater adaptations for the same amount of training stimulus (instead of allowing the user to dump more stimulus onto their body). And of course, the most anabolic natural supplement of all is sleep.
These points in no way refute anything you've said in this article, I just think it's important to note.
Beautiful.
To risk piercing the veil a bit here (I know, a touch uncouth, but I just can't help myself):
Does Dynomight take creatine?
Any reason why or why not?
Waaaaaay too risky to answer! I can't win!
wait, would the cleanest test be sleep-deprived/vegetarian/older people, with brain creatine measured before and after, instead of just asking whether random healthy people feel sharper?
Yes! In fact several experts specifically call for this kind of experiment. For example, the McMorris review (https://doi.org/10.1016/j.bbr.2024.114982) says:
> Probably the most important factor for future research is the necessity to measure brain creatine content before and after supplementation.
Ah nice, that’s the missing endpoint. I’d really want baseline brain creatine in the model too, because a null average could just be mixing actual brain responders with people whose brains barely changed.
Yeah, absolutely. I need to see scatterplots of Δ(brain creatine levels) vs. Δ(cognitive tests). However, I guess this would be pretty expensive due to the difficulty of measuring (really, estimating) brain creatine levels.
Yeah, even a noisy MRS substudy would be worth a lot. Baseline + Δbrain creatine would at least tell you whether the cognitive null is biology, psychometrics, or averaging over non-uptakers.
I understand the thought. A problem is that there's really two different questions, and both are interesting: (1) does a random health person see a benefit (the big payout, if proven), (2) does ANYONE show a benefit (the easier proof, but far less important demonstration). Depending on what your mandate/interest in, I could justify wanting to bias towards either one of those questions.
Yeah, fair split. I was thinking of the enriched group as a cheap mechanism test: if sleep-deprived/vegetarian/older people with low-ish brain creatine don't budge, I get much less excited about paying for the random-healthy-adult RCT.