"Coffee dehydrates you anyway"

You may have come to this book believing the opposite of chapter 4. Coffee is a diuretic, everyone knows that, so the two coffees before eleven aren't just failing to count, they're taking fluid out. On that view the dry day is worse than it looks and the coffee is the villain.

The evidence is more specific than the folk version. Maughan and Griffin reviewed the literature on caffeine and fluid balance in 2003. Large single doses of caffeine, at least 250 to 300 milligrams, roughly two or three cups of coffee taken at once, do produce a short-term increase in pee output, but mainly in people who've had no caffeine for days or weeks. In regular tea and coffee drinkers, a profound tolerance develops and the effect is much diminished. Killer, Blannin and Jeukendrup's three-day study in chapter 4 is the direct test. In habitual drinkers, four cups a day of coffee and the same volume of water produced the same total body water and the same 24-hour pee volume. If you drink coffee every day, it isn't drying you out.

You may then notice that the NHS dehydration page says to avoid caffeinated drinks and alcohol if you have symptoms of dehydration, because both will dehydrate you, while the NHS fluids page says tea and coffee count. Both pages are right in their place. If you're already dehydrated, from being sick or from heat, the safe advice is water or rehydration solution and nothing that might add to the loss. On an ordinary day for an ordinary coffee drinker, the coffee counts. This book is about ordinary days. Chapter 20 is about the others.

So coffee isn't the villain of the dry day. The dry day's problem isn't what the coffee takes out. It's that the coffee stops at eleven and nothing follows it, and that the third one at three is drunk for the lift rather than the fluid, which brings its own cost in the evening. The second coffee, the earlier book, covers that cost in full. Here I'll say only that a glass of water at three is the one drink that addresses the flat hour without charging you for it at bedtime.

There's a fair objection, which is that if coffee counts, why not just have more coffee. The answer isn't hydration. It's caffeine, and what caffeine after about lunchtime does to the night. Water has no cut-off time. That's its only advantage over coffee, and in the afternoon it's the one that matters.