Caffeine came off the banned list and nothing terrible happened

A threshold that punished coffee drinkers was replaced by monitoring, and the sky stayed where it was.

Twelve micrograms per millilitre. That was the line, and beneath it caffeine was legal and above it you were a doper, which meant the offence was quantitative in a way almost nothing else on the list was.

It caused exactly the problems a threshold rule always causes. Urinary caffeine concentration varies with body mass, hydration status, genetics and how recently you last went to the toilet, so the same intake could put two athletes on opposite sides of the line. Competitors who used caffeine deliberately learned to dose just under it. Competitors who drank a lot of coffee and happened to be dehydrated after a hot race occasionally found themselves explaining an espresso to a panel.

So a rule intended to catch a performance intervention ended up sorting athletes by their kidneys, and the enforcement had a peculiar shape: the sophisticated user was safest, because the sophisticated user knew the number.

It was removed from the prohibited list and placed on a monitoring programme, where it remains. The consequences are worth looking at honestly, because prohibition arguments are usually settled by prediction rather than evidence. Use went up. It went up a lot, and caffeine is now near-universal in endurance sport, taken in gels and gums and pre-measured doses. Nobody has produced the wave of harm that was forecast.

I count that as a good outcome for a specific reason. The old rule was not preventing use, it was regulating disclosure, and it pushed dosing into a zone where athletes were guessing about their own concentration rather than talking to a nutritionist about a plan. The monitoring programme collects data without creating an offence, which is a genuinely underused regulatory tool.

The interesting question now is where the same logic applies next. Several substances sit on the list mainly because they were on it before, and a monitoring-first approach would tell us whether use patterns actually justify prohibition.

A threshold that can be gamed by the informed and tripped by the unlucky is not a fair rule. It is a test of who read the guidance, and we removed one of those without the sport falling over.