The 90/90 rule and other lies we tell ourselves

The internet is full of rules for deciding what clothes to keep, and every single one of them is a partial truth wearing the costume of a law. Let's go through the greatest hits — what each gets right, where it lies to you — and end with the only test we've found that survives contact with a real closet.

The 90/90 rule

"If you haven't worn it in 90 days and won't wear it in the next 90, let it go."

What it gets right: recent behavior predicts future behavior far better than intentions do. The lie: it was invented by people who apparently don't experience winter. Run 90/90 in July and you'll donate every coat you own; run it in January and lose your entire summer wardrobe. Patch: make it seasonal — haven't worn it this season, won't wear it next same-season. That's really a 365-day rule, which is less catchy but more honest.

The hanger flip

"Flip all your hangers backward; flip each one forward when you wear the item. After six months, whatever's still backward goes."

What it gets right: it converts memory into physical evidence, and evidence is hard to argue with. The lie: it takes six months, which means it's not a cleanout method — it's a way to schedule a future argument with yourself. And folded clothes, shoes, and everything on the chair are outside its jurisdiction entirely. Use it as a audit, not a method.

"Does it spark joy?"

What it gets right: more than people admit. The physical-contact part is real — holding an item produces a faster, more honest signal than looking at it. The lie: joy is the wrong bar for half a wardrobe. No one's plain white undershirt sparks joy. Workhorse clothes should be judged on service, not sparkle. The patch: joy for the expressive stuff, usage for the basics.

The one-in-one-out rule

What it gets right: it's the only rule here that prevents the problem instead of treating it. The lie: it assumes a steady state you haven't reached yet. One-in-one-out maintains a right-sized closet; it cannot create one. It's the rule you adopt after the cleanout, not instead of it.

The only test that actually works

Every rule above is trying to approximate the same underlying question, so skip the approximation and ask it directly:

"If this weren't already mine, would I take it today — for free — from a friend's cleanout page?"

It works because it removes the two forces that corrupt every keep decision: sunk cost (you already paid for it, which is irrelevant) and status quo bias (keeping feels like no decision, which is false). Framed as a fresh acquisition, your honest taste shows up instantly. You'd take the flannel. You would not take the pinchy shoes. You knew that before you finished reading the question.

And everything that fails the test has somewhere good to go — a friend who would take it today. That's the whole premise of the method: the exit is as important as the verdict. Sort with the tree in sell, give, or donate, and let the rules be what they always were — training wheels for the one question that matters.