Practiceby Lab 10 Get started ›
Crossfire

When should I stop drinking coffee: the same time as everyone else, or a time that depends on me?

What the experts say

Michael J. BreusPhD · clinical psychologist (NPPES taxonomy 103TC0700X, Arizona licence 3760, record last updated 26 Nov 2012) · Diplomate of the American Board of Sleep Medicine, specialty Psychology, certified 17 Apr 2000, a lifetime credential with no expiry and no recertification · stated Fellowship of the American Academy of Sleep Medicine, a voluntary association, not verified

It depends on your chronotype. Four types, four windows, from 8am to 10am for a Lion up to 11am to 2pm for a Wolf.

What it rests onHis per-type schedules place caffeine inside each type's stated productive block rather than at a fixed clock time: Lion 8 to 10am, Dolphin 8:30 to 10am, Bear 9:30 to 11am, Wolf 11am to 2pm. Two of the four give an explicit cortisol mechanism for the opening, and the Wolf page states the curfew as an instruction rather than leaving it in a list: 'Avoid any caffeine after 2 p.m. to avoid interfering with your sleep schedule.' His general-population article gives the same instruction without a type, keep caffeine to the morning and early afternoon, on the grounds that noticeable effects last up to seven hours. Three of his four windows close earlier than Walker's noon and one runs two hours past it. No trial is cited for any of the four windows, and the four chronotype pages carry no citations at all.

Source: sleepdoctor.com ↗
Matthew WalkerPhD · neuroscientist

Around noon, for everyone.

What it rests onHis reasoning is pharmacokinetic rather than behavioural: caffeine's half-life is five to six hours and its quarter-life roughly ten to twelve hours in an average adult, so a mid-afternoon coffee is still measurably present at midnight. The show notes for the source episode state it flatly, 'Finally, try to abstain from coffee from around noon and avoid drinking alcohol in the evening', with no exception, no chronotype and no mention of individual variation anywhere in the surrounding notes. He restates it in two later episodes and in the Huberman guest series.

Source: themattwalkerpodcast.buzzsprout.com ↗

Overview

This is a disagreement about population wearing the clothes of a disagreement about substance, and reading either man's headline rather than his table manufactures a fight that mostly is not there.

Both want a curfew, both argue it from how long caffeine lasts, and for three of Breus's four types his curfew is STRICTER than Walker's. The genuine dispute is confined to one group, the Wolf, roughly 15 percent of the population on Breus's own figures, who is told to drink coffee for two hours after Walker's cutoff. Neither man cites a trial for his number, and their two duration figures are not the same measurement: Walker gives a half-life and a quarter-life, Breus gives seven hours of noticeable effect.

Where they agree

Both hold that there should be a caffeine curfew, both argue it from the persistence of caffeine rather than from habit or dependence, both treat afternoon caffeine as a sleep problem rather than a health problem, and neither tells anyone to quit.

What would settle it

A trial that stratified a fixed-curfew rule against a chronotype-matched rule on measured sleep outcomes, in people typed by a published and validated instrument. Nothing of that shape exists, and Breus's instrument is neither published as a method nor validated, so the comparison cannot currently be run even in principle.

Practice does not pick a winner here

You do, and your call is dated, kept, and yours to change. Follow the people you already trust and what they publish turns into a short daily practice.

Get started

Free to sign up and use. We do not ask for a card.