Real life · Regression to the mean
Do Hangover Cures Work? Why Every One of Them Seems To
The short version: every hangover cure seems to work, because you take it at your absolute worst, and from your absolute worst the only way is up. Statisticians call this regression to the mean. The pickle juice gets the credit. Time did the work.
Uncle Gary has a hangover cure. It contains pickle juice, a raw egg, a spoonful of hot sauce and what he calls "a little hair of the dog". He has been making it since 1994.
"Works every time," says Gary. And here's the annoying part: he's right. Every single time he drinks it, he feels better a few hours later.
He would also have felt better a few hours later if he had drunk a glass of water and watched cartoons.
What is going on
Nobody takes a hangover cure on a good morning. You take it at the exact moment you feel worst: lying on the kitchen floor, negotiating with the ceiling. And when something is at its most extreme, the next measurement is very likely to be less extreme, whatever you do.
That's regression to the mean. It isn't magic and it isn't medicine. It's just what happens when you start measuring at the bottom.
It explains a lot of everyday "proof":
- You visit the doctor when your back pain is at its worst. A week later it's better. The doctor's advice looks brilliant, and maybe it was. But it would have looked brilliant anyway.
- A footballer has the season of his life and appears on every magazine cover. The next season is merely good. Fans blame the curse of the magazine cover.
- Kevin bowls a personal best, buys new bowling shoes and then bowls an average game. The shoes get blamed. The shoes were innocent.
How bad you feel the morning after (schematic)
How you would actually test the cure
You need a control group: some bad mornings with the cure, some bad mornings without it, picked at random, and then you compare how you feel at lunchtime. If the cure mornings are clearly better, the cure might work. If they're the same, you've been drinking pickle juice for nothing.
Gary declined to take part in the study, on the grounds that it would require twice as many hangovers.
Where it costs you money
The same trap waits in investing. People buy last year's best fund right after its luckiest year, and sell a stock right after its worst month. Both are measured at an extreme. Both tend to drift back toward normal, and the timing makes the decision look worse than it was.
Anything that is used "when things are at their worst" will look like it works. That's why we test every trading idea against a random stock on the same day.
The moral
If something seems to work every single time, check when you're using it. If the answer is "whenever things are at their absolute worst", you may simply have discovered time.
This is not medical advice either. Drink water. Educational content, not financial advice.