Unpacking Is Not Packing in Reverse
A packing list is a bet placed before the fact. You don't know if it will rain, so you pack the shell. You don't know if you'll want to run, so you pack the shoes. Every line on the list is a small wager against an unresolved future, and the list itself is just the collected record of those wagers.
Unpacking looks like the same task run backward, but it isn't. By the time you're standing over the open suitcase, the future you were betting against has already happened. It rained or it didn't. You ran or you didn't. The uncertainty that made packing hard is gone, replaced by a fact. So it seems like unpacking should be easy: just look at what you used, keep that, cut the rest.
Except the fact doesn't settle the question it looks like it settles. Say you packed the rain shell and it rained. Fine, the wager paid off, but that tells you almost nothing about whether to pack it next time, because you don't know if it would have rained anyway on some baseline version of this trip, or whether this particular destination just runs wet in this season, or whether you got unlucky in a way that won't repeat. One outcome is not a track record. Say instead you packed the shell and it didn't rain. That's even less informative, because the absence of rain doesn't mean the shell was a bad idea, it means this specific roll of the dice came up dry. You can't tell, from a sample size of one trip, whether you overpacked or whether you got away with something.
This is the actual reason there's no such thing as an unpacking checklist, and it isn't for lack of trying to write one. A checklist compresses judgment into a procedure, and that compression only works when the underlying decision is stable across instances, when the same input reliably produces the same right answer. Packing questions are semi-stable in exactly that way: you can look at a forecast, a duration, an itinerary, and make a decent-enough guess about what a person going there will generally need. That's why packing lists are a whole genre online, endlessly reproduced, copied from trip to trip with minor edits. The judgment they encode barely changes.
Unpacking's implicit question, keep this in the go-bag or cut it, is unstable by construction, because it depends on whether this trip's particular need showed up, and that varies every time. You can't write a stable procedure for an unstable target. The checklist would have to say something like "keep the shell if it's going to rain next time," which is not a checklist, it's the original problem again, wearing a hindsight costume.
What people actually do instead is quietly downgrade the whole exercise into vibes. You stand there holding the unused toiletry bag and you don't reason your way to a verdict, you just feel like it was dead weight, or you feel like you'd rather have it than not, and you repack accordingly. That feeling is doing real inferential work, it's a rough integration over every past trip you've half-remembered, but it never gets written down, because writing it down would require admitting there's no rule under it, just an accumulating and slightly unreliable sense of the odds.
The interesting failure mode is the person who tries to fix this with data: logs every item, tracks usage across a dozen trips, actually builds the table. That person can, eventually, produce something checklist-shaped, because they've converted a one-off resolved fact into an actual frequency. But almost nobody does that for a suitcase, because the stakes are too low to justify the bookkeeping and too high to just guess and forget. So the judgment stays soft forever, redone from scratch every time you open the bag, with no artifact to show for the last twenty times you did it.
The shell goes back in the suitcase. You don't know if that was right. You won't know until the next trip, and even then, only a little.