Your Morning Coffee Lasts Longer Than You Think: A Lot Longer
Source: LessWrong | Published April 8, 2026
Every caffeine calculator on the internet uses the same model. You drink coffee, caffeine enters your bloodstream, and its effects decay with a 5-hour half-life. Drink a cup at 8am and by 6pm there is almost nothing left. That is the story we tell ourselves so we can have that afternoon espresso without guilt.
It is probably wrong.
A post on LessWrong by user kman walks through the actual biochemistry. The tl;dr: caffeine gets metabolized into paraxanthine, which also blocks adenosine receptors and has its own 3-5 hour half-life. Most circulating caffeine (over 80%), turns into paraxanthine before it gets cleared. And paraxanthine is roughly as potent as caffeine at the receptors that make you feel alert.
This means the effective concentration curve is not a simple exponential decay. It is a two-stage system. Caffeine hits, then paraxanthine takes over and keeps going.
What This Actually Means
If you drink 200mg of caffeine at 8am, the simple model says you are down to negligible levels by early evening. The two-stage model says you still have meaningful adenosine blockade going into the night.
There is an interactive simulator linked in the post that lets you play with the numbers. I tried it with my own habits (two cups before noon) and the curve shows elevated concentration past 10pm. That tracks with how I actually feel when I am honest about it.
The Paraxanthine Option
The post also introduces something I had never heard of: paraxanthine supplements. You can buy 100mg capsules. The idea is that since paraxanthine is the metabolite rather than the parent compound, it clears faster with no downstream active metabolites. So you get the wakefulness without the lingering tail.
The author tried it and reports:
- 100mg in the late afternoon gives energy without affecting sleep at 10pm
- 200mg in the morning feels stronger than 100mg caffeine
- Effects peak within an hour and noticeably decline after 2-3 hours
- Some energy crash in the morning dose but not with afternoon/evening
That shorter duration is the whole point. If caffeine’s effects actually persist for 12+ hours due to paraxanthine, then a stimulant that cuts that in half is useful for anyone who wants to drink coffee past noon and still sleep.
Why Nobody Talks About This
The metabolism of caffeine into paraxanthine has been known since the early 1980s. Paraxanthine’s adenosine receptor antagonism has been documented for just as long. Yet every caffeine calculator, every wellness influencer, every sleep doctor repeats the 5-hour half-life model as if it is settled fact.
The author calls this a “civilizational inadequacy”, something well understood in the literature that never made it into common knowledge. If the most widely used psychoactive substance on earth is this poorly understood by the people who consume it daily, there is probably more low-hanging fruit in cognitive enhancement than we realize.
I do not know if paraxanthine would work for me the way it did for the author. But I do know that the story I told myself about caffeine being out of my system by bedtime was never quite true, and I am glad someone finally checked. Maybe it’s better not to drink coffee at all, or to have it only once in a while.
Crepi il lupo! 🐺