Caffeine after 2pm: the half-life argument

Caffeine is the most useful legal drug available to someone in a calorie deficit. It blunts appetite slightly, it makes a training session in week three feel like something you can do, and it is genuinely performance-enhancing.

It also has a half-life long enough that the afternoon dose is a night-time dose. That is the whole argument, and it's arithmetic rather than opinion.

What a half-life actually means

Caffeine's half-life in healthy adults averages somewhere around five to six hours. That means half of what you took is still circulating after that time — not gone, halved.

Run it forward from a 4pm coffee. Around 10pm you still have half of it in you. Around 4am you have a quarter. A single large afternoon coffee is not a small residue by bedtime; it's a meaningful fraction of the dose you deliberately took to feel alert.

Move the same coffee to 10am and the picture inverts. By 10pm you're down to roughly an eighth, which is the level at which most people's sleep architecture stops noticing. Nothing about the drink changed. Only the clock did.

It also stacks. Three coffees across a working day don't clear independently — each one is still partly present when the next arrives, so the level you're carrying into the evening is higher than any single dose would suggest.

The evidence people find hardest to believe

Controlled work in this area has given people a large dose of caffeine — around the amount in a big takeaway coffee — at bedtime, three hours before bed, and six hours before bed, and measured what happened to their sleep. The six-hour dose still disrupted it.

The detail that matters more than the disruption is this: the participants largely didn't notice. Their own ratings of how they slept didn't track what the measurements showed. Self-report is not a reliable instrument here, which is exactly why so many people are certain that coffee doesn't affect them.

It's worth being clear about what the disruption looks like, because it isn't usually insomnia. It's a slightly longer time to fall asleep, slightly less deep sleep, and slightly more brief wakings you don't remember. A night that felt fine and produced less recovery than it should have.

Why your tolerance isn't the defence you think it is

How fast you clear caffeine varies a lot between people, mostly down to liver enzyme activity, and that variation is largely genetic. Some people are fast metabolisers and genuinely get away with more. Some clear it far more slowly than average, and hormonal contraceptives can slow it further.

But tolerance to the feeling and tolerance to the effect are different things. What habitual drinkers lose is the jittery, obviously-caffeinated sensation. What they don't lose is the effect on how quickly they fall asleep and how much deep sleep they get. You stop feeling it long before you stop being affected by it.

“You build tolerance to noticing caffeine. You do not build tolerance to what it does to your sleep.”

The cut-off, and how to actually keep it

The program's rule is nothing caffeinated after 2pm. That gives you roughly eight to nine hours before a reasonable bedtime, which drops most people to a fraction small enough to stop mattering. If you know you're a slow metaboliser, or you're pregnant, move it earlier.

The failure mode isn't discipline. It's that the afternoon coffee was never really about caffeine — it was a break, a walk, a warm thing to hold. Replace the ritual and the drug takes care of itself.

• Set a phone alarm labelled last coffee at 1:45pm for the first week. After that you won't need it
• Swap the 4pm cup for a decaf or a herbal tea — the break, the walk and the warm mug survive intact
• Check the hidden sources: green tea, cola, pre-workout, some painkillers and dark chocolate all carry a dose
• If you train in the evening, use the pre-workout earlier or not at all. It is the single most common cause of a 1am ceiling stare
• Front-load instead. Most of your daily caffeine before noon costs you nothing and buys the same alertness

What this is worth to the month

Sleeping badly during a deficit is expensive in a specific way. Short sleep pushes appetite up and makes training feel harder at the same load, and both of those are already elevated by eating less and training more. The 2pm rule is one of the cheapest protections available against that stack.

It also compounds. A late coffee costs you sleep, poor sleep makes the next afternoon harder, and the obvious fix at 4pm the following day is another coffee. Breaking that loop at the front is far easier than fighting it at the end.

And none of this asks you to give up coffee. Nothing in the program does. The morning dose stays exactly where it is, including the one before training, where caffeine is genuinely useful for output. You are moving the same drug earlier in the day, not deleting it.

If you're pregnant, managing an anxiety or heart condition, or taking medication that interacts with caffeine, the right daily amount is a question for your clinician rather than for a training article.