Hofstadter's Law

You've probably drawn up a schedule on the first day of summer break. You write down that if you do two pages of the workbook a day, you'll be done in ten days. But even though you work steadily every day, you get stuck on hard problems and have to redo the ones you got wrong. It takes fifteen days.

Next summer, you feel wiser and fix the plan. Last time you ran late, so this time you pad it to twenty days. And this time it takes twenty-five.

Hofstadter's Law sums up this loop in one sentence. Everything takes longer than you expect, and even a plan that already allows for delays runs late again.

The first half sounds like old, familiar nagging. The reason people keep quoting this law is the second half. It points out that even when you prepare, it's hard to escape.

Think about the day that pops into your head the moment you make a plan. You sit at your desk and breeze through the workbook. You're picturing only the one perfect day with nothing in the way, and then counting up the days.

Real days don't go like that. You tear your room apart looking for the workbook, a friend texts you, and you have to fix an early section you did wrong. These bits of time almost never come to mind when you draw up a schedule.

When jobs are chained together, the slips get bigger. To write a book report, you need the book first. If the library copy is checked out, everything scheduled after it gets pushed back. When the front slips, the back slips with it, and if the pushed-back work collides with other commitments, a one-day delay can balloon into several days.

So the gap between "how long it takes when everything goes well" and "how long it really takes" keeps widening. Most of us build our schedules from the first one.

To be more precise, this isn't a formula measured with a ruler. It's a line that the American cognitive scientist Douglas Hofstadter wrote in one of his books, a witty saying that captures something many people have lived through.

The charm of the line is that it points back at itself. Stretching your homework plan to twenty days is still just a new plan. You made it by imagining the smooth-sailing days again, so the same trap follows right along.

That's why simply doubling your buffer doesn't help much. What does help is writing down how many days the same homework actually took last time, then using that record as the baseline for your next plan.

That said, it doesn't mean everything always runs late. For short, familiar tasks you've done many times, your estimates are pretty accurate. The slips get noticeably bigger with things you're doing for the first time or jobs that involve many people.