Caffeine half-life and performance

After seeing the post this morning “How much caffeine on 100mi gravel” I went down a rabbit hole and have a question that I think merits its own topic.

I saw a paper that suggested that caffeine has an approximate half-life in the body of 5.7hrs. So by my math (degree in physics so I’m pretty sure I got this right) that means there is a k value of 0.1216 for caffeine when using hours and we can do some interesting calculations.

Let’s assume a racer has a cup of coffee (100mg) at 7am before a race. If that race starts at 9am, they will still have 78mg of caffeine in their system when the gun goes off. Then if they take a gel with 20mg of caffeine every hour after the start of the race for a 5 hour race, they will end up with 102mg of caffeine in their system at the end of the race.

So, does having 102mg of caffeine in the system mean the racer’s body will experience the same effect of that caffeine as when they took most of that in back earlier in the day? Or does this latent caffeine that’s been in the body for many hours end up losing its effect? Or, finally does the physiological impact of racing mean the caffeine gets cleared faster from the body than when just sitting at a desk typing at a computer so the racer would have less than 102mg?


I don’t know the exact label, but there is a gene that controls the rate of caffeine metabolism, and there are fast metabolizers and slow metabolizers. I have the fast metabolizer gene, but at least in my experience I’m very sensitive to caffeine. I would expect the serum half lives to be different in the two groups.

1 Like

This would depend on the kinetics and mechanism of caffeine excretion. If freely filtered in the kidneys, the elevated HR and systolic BP would lead to elevated GFR and quicker losses. (You’d have increased rate of circulation through the kidneys and also increased pressure at the glomerulus).

If there is enzyme mediated breakdown of caffeine, it could be a multitude of possibilities based on the rate limiting component.

Since caffeine must act at extracellular receptors, my assumption without looking into more research, is that the receptors will become desensitized overtime and a larger intake will be needed. If this happens acutely during a race or more chronically, I’m not sure. Different drugs lead to desensitization at different rates.

(BS in Biochem and recent MD grad). If I’m wrong, let me know!


This is a pharmacology problem. Caffeine is primarily broken down by Cytochrome P450, much like everything else, where metabolites are then eliminated through the kidneys. It follows first order kinetics, so that keeps things simple. Your rate of caffeine breakdown is less dependent on your heart rate, blood pressure or anything like that as the metabolism of it is mediated by enzymes in the liver.

The effect of caffeine on the other hand…is more variable.

1 Like

The Wikipedia article on Caffeine is fairly complete. Take away is that half life in healthy adults ranges from 3 to 7 hrs. You can read the article for all the detail you want.

Congratulations on your recent MD. I graduated from medical school 39 years ago

1 Like

It’s this one.

However there are many other factors asides from genetics that can affect the effects. My view on it is that knowing your baseline is a good starting point if you’re looking to do some experimenting.

I’m using this as my main takeaway. From somebody who doesn’t do chemistry or biology, I’m coming away with the idea that the effect of the caffeine is going to be greater when I first have it, even if later on in the day the total amount of caffeine in my body is more than 100mg.

I’ll just have to keep experimenting and see what works for me.

This is a chart I created a while back to understand impact of different dosing strategies on caffeine concentration in the body for a long race. It assumes a 4 hr half life.

3 mg/kg of bodyweight has been shown to be the minimum dose need to get a performance benefit.

So the big cup of coffee at the start of the race may not be the best approach.

The scenario I didn’t show on this, but is probably the best approach, is to start taking caffeine half way through, so you get the boost at the end where it is most needed.

Strategy #1: 400mg of caffeine all at once e.g. four 100mg caffeine tablets at the beginning of the race.

Strategy #2: 200mg of caffeine up front, and three more 100mg tablets at 2, 4 and 6 hours.

Strategy #3: 250mg of caffeine up front, and another 30mg every hour thereafter.

70kg rider assumed.


Cool chart. Have you executed this to see if the chart holds true to what you feel on the bike?

I’ve done the approach shown in the red line - 200mg upfront, then 100mg every couple of hours.

I’ve found I feel great for the first few hours, but then fade in the latter half, despite continuing to take caffeine.

I think the caffeine at the beginning lowers RPE too much, so I’ve gone out too hard at the beginning.

As I mentioned above, I think a better strategy is to hold off on caffeine until the latter half of a race.

1 Like

Hmmm… just curious question regarding your prefered Strategy #2… I do believe the time between using caffeine pill and peak uptake is around 90 minutes. In your graph you peaked your intake within 15 min it looks like if I see correct? In fluid form in say caffeine shots uptake time is around 30 min, give or take 10 min… or chewing gum is also a way how to quickest absorb caffeine… :thinking:

Thanks for this post. Been a game changer for me. It led to me searching the raw data on my 23andme profile and I found the gene. I’m A/C which means I’m a fast metaboliser and that explains a lot. Used to wonder when we started going to coffee shops as older teens why my friends would love it and I’d get all jittery. I started ordering hot chocolate instead!

I do drink coffee now, lots of it, and it does make me anxious but I burn it off on rides and to be honest actually love getting jacked up on caffeine and nailing a hard session. I do avoid caffeine socially otherwise though.

Great to finally understand why. Always thought it wasn’t just… me.

I takes 40-80 mins to absorb. For simplified modeling, I assumed instant.

I wouldn’t describe #2 as my preferred strategy, as I found it leads to me pushing too hard early in the race when caffeine makes the RPE feel low. Delaying caffeine consumption towards the later stages of the race I think will work better for me.

Does caffeine get absorbed faster when you exercise? I am curious about this from a sleep impact point of view. If I have extra coffee late in the afternoon because I am doing a hard workout after work will it impact my sleep less than if I have the same amount of coffee and don’t exercise?

1 Like

Someone more smarterer than I should do an adrenaline half-life thread.

That exercise/sleep comment reminded me of how, during exercise, the level of adrenaline is stable but the reabsorption rate slows down, resulting in more adrenaline in the system. (I could be remembering that process completely wrong.)

Are there substances etc which negate/prolong the utilisation of caffeine? :man_shrugging:

Dont know of ”absorbed” is the word I would use, but you will metabolize caffeine faster during excerise. There were some study, 60 min rest or 60 min excersice was compared. With rest, it was around 4 hours ,’with excerise it was around 2.5 hours. (Time to half the amount)… but as others had mentioned, there are basically 3 gen types that responds differently where one geno-type actually performed worse in a time trail test.

Simple answer:

A study of the literature shows there might be a slight difference between exercise and not, but probably not a meaningful one.

Having the coffee whether you exercise or not will probably impact your sleep. It can either affect your ability to fall asleep or your sleep quality. Even if you are the type that can sleep after having two cups of coffee after dinner, the data suggests that your deep sleep/REM cycles may be affected. Does that have a meaningful impact? It’s complicated.

The best test is to just try it out. Go with a week without the coffee and a week with the coffee in the afternoon, then assess your sleep quality.

That 1991 paper wasn’t the best because they didn’t control for cigarette smokers, menstruation cycle amongst other things. All of which can affect caffeine metabolism. Subsequent studies didn’t show a significant difference between exercise or non exercise. Also, their version of “exercise” was 30% VO2 max which is better than nothing, but we’re more interested in something like 70% VO2 max efforts. There isn’t a whole lot of research in this area though so we don’t have much to work with.



Agree… Probably the individual metabolism and gene setup has bigger impact, so doing a study like that would need to take a lot of parameters into account…

Yup, and those studies show that your recent behavior has an even bigger effect on rate of metabolism. The frequency in which you consume coffee or other drugs (eg. cigarettes) will have a greater effect than your genes.

1 Like