"I need it because I sleep so badly"

Some readers will say the order is wrong. The coffee isn't what's spoiling your sleep. You've always slept badly, and the coffee is how you cope with the days. Take it away and you'll be a bad sleeper without a crutch.

I want to take this seriously, because there are people for whom it's true, and the preface said so. Insomnia, sleep apnoea, pain, a baby, shift work and a good many other things ruin sleep on their own. If you have one of those, the coffee is downstream of it, and the thing to fix is the thing itself, with a GP if needed. This book will still help a little, because coffee makes any night worse, but it isn't the main answer.

For most readers, though, I'd question the order, and here's why. Go back to what the research says the coffee does to a night. It makes sleep shallower, more broken and shorter, and it reduces the deep part. It does this even from a morning dose, and much more from an afternoon or evening one. So a person who drinks coffee through the day is a person with a caffeine-damaged night, whatever else is going on. Their morning is a shortfall. Their afternoon is a slump. Their answer to both is more coffee, which damages the next night. That's a circle, and from inside the circle it feels exactly like being a bad sleeper who needs coffee.

You can't tell, from inside, which came first. What you can do is notice that the circle only has one place where you can put your hand on it. You can't decide to sleep more deeply. You can decide when the last cup is. If the last cup moves to before two, the night gets a little better, and the shortfall the next morning is a little smaller, and the circle turns the other way.

There's a reasonable test, and it's the plan in chapters 17 and 18. Bring the caffeine down over two or three weeks, move the cut-off back, and watch the morning score from your notes. If it climbs, the coffee was part of the problem. If after a month at a lower level with an early cut-off you're sleeping just as badly and just as tired, then your bad sleep has another cause, and you've learned something worth taking to a GP, and you can go back to drinking coffee at any hour you like, since it wasn't the coffee.

I'd expect the first result for most people. Not because I doubt your sleep, but because the mechanism is well established and the cut-off you've been keeping is inside the window that the evidence says matters.