Is this a good swap?

Honestly: from one anchor and a lot of argument. One fun-size Snickers is 10 points, and every other line in the table was placed against that rather than assigned on its own. That is the only way a value table ends up internally consistent — decide what one thing is worth, then ask of everything else whether you would swap it for that.

Three things then do most of the work.

Peanut butter sits at the top. Reese's Cups win national favourite-candy surveys in the United States essentially every year, across polls run by very different organisations with very different methods. When something wins that consistently, putting it first is not an opinion, it is reporting. Everything in the peanut butter family is therefore priced above its chocolate equivalent.

Size is worth more than size. A full-size bar is about 2.6 times the candy of a fun size, and this table prices it at 3. King size is 4.5. The extra is a scarcity premium and it is deliberate: on one night a year, almost every house hands out fun size, so the house that hands out full bars becomes a neighbourhood event and the bar itself becomes a trophy. What a thing trades for and what a thing weighs are different questions, and every real market knows this.

The bottom of the table is not zero by accident. Candy corn is worth 1, not 0. Raisins, loose change and a toothbrush are worth 0. The distinction matters: something worth 1 can still be bundled into a swap to close a gap, which is funnier and more useful than excluding it. A toothbrush is worth nothing tonight and quite a lot of value over a lifetime, which is a different table entirely.

Two things have to be true at once, and most arguments about candy swaps are really an argument about which one is being used.

The first is equivalent value — the two sides are worth about the same. That is what the table computes, and it is the version adults instinctively reach for, because it is the version that sounds fair.

The second is both sides ending up better off. This is the version that actually explains why swaps happen at all. If you hate coconut and I hate liquorice, an Almond Joy for a Good & Plenty is a swap where both of us gain, even though on paper one side is worth more points than the other. Nobody was cheated. Value moved anyway. This is not a childhood quirk; it is the whole reason trade exists, and a Halloween bucket is the clearest demonstration of it most people will ever encounter.

A calculator that only knows the first definition will confidently call a swap that suits both people unfair. That is why the tastes row is here, and why it is not a bonus feature. Tick what you do not eat and the table drops it to a quarter of book value on your side — enough to change the answer, not enough to pretend it is worthless. The gap between the plain total and the total with tastes applied is, roughly, the amount of value the swap is creating out of nothing.

Telling two people that a swap is unfair ends the conversation. Telling them exactly what would fix it continues it, which is the outcome everybody standing over the bucket actually wants.

So when the answer is not fair, the page searches the whole table for the single item — at any size — that most nearly closes the gap, and it only offers it if adding that one item would land the swap inside the fair band. "They need to add a Dum-Dum" on a forty-point gap is worse than saying nothing, so on a gap that large it says so plainly and tells you to try a different swap.

It prefers the smallest ask that works. A fun-size piece beats a full-size one, and a closer fit beats a further one, because an offer that is easy to accept is the point. That line is not the page being clever; it is the page being useful in the ninety seconds before somebody walks off with the bucket.

Candy swapping arrives with its own folk law, and most of it is surprisingly durable.

• Three-for-one on anything, four-for-one on the good stuff. The most widely repeated exchange rate there is. It works because it encodes a real asymmetry — the top of the table really is worth several times the bottom — and fails whenever somebody tries to pay four Dum-Dums for a full-size anything.
• Full size counts double. Universal, and undervalued. This table says triple.
• No swapping back. A sound rule with no economic content whatsoever. It exists to stop the negotiation from becoming the evening.
• Unwrapped is unswappable. Not a valuation rule at all — a safety rule wearing a valuation rule's clothes, and worth keeping for exactly that reason.
• Everything is negotiable until someone cries. The actual governing principle. This site exists to postpone that moment.

The one folk rule this table rejects is counting pieces. Quantity and value are not the same thing, and a bucket sorted by count is a bucket being swapped badly.

Better than most things marketed as one, for a reason that has nothing to do with candy: the stakes are real to the child and trivial to everybody else. Every concept that makes markets hard to teach is sitting there in the bucket already.

Scarcity is the full-size bar. Subjective value is the coconut argument. A medium of exchange is whatever the table's peanut-butter currency turns out to be that year. Price discovery is two eight-year-olds finding out what a Kit Kat actually goes for on 31 October. Mutual gain is the swap where both of them walk away pleased and an adult, watching, cannot work out who won.

The one lesson to be careful with is the one this page is best at delivering: that there is an objective, correct price. There is not. There is a table, the table is a starting point for an argument, and the argument is the part where the learning happens. If a child ends the evening quoting the value chart at their sibling, something has gone slightly wrong.

It is American. The list is the haul from a United States doorstep, because that is where trick-or-treating at this scale happens, and a reader in the UK, Ireland, Canada, Mexico or Japan will find brands missing that they consider fundamental. Other languages of this page have their own list rather than a translated one, for exactly this reason.

It ignores condition. A Kit Kat that has been in a warm pocket for two hours is not the same asset as one that has not, and the table has no opinion about this.

It ignores getting bored. In a real bucket, the tenth fun-size Snickers is worth noticeably less than the first, because you have run out of appetite before you have run out of Snickers. The table prices every copy the same. This is wrong and it is wrong on purpose — the alternative is a tool that argues with you about your own marginal utility while a seven-year-old waits.

And it ignores the single largest factor in any real trade, which is who wants it more. The “do you like it” row is a rough stand-in for that and nothing more. If the page says no and both of you want to do it anyway, do it. The table is not the law. It is just the only impartial party in the room.

Nothing you do here leaves your browser. The baskets, the totals and the answer are all worked out on this page — there is no account, no server holding your swap, and nothing is recorded anywhere, including the swap number, which is made up from the swap itself as a joke about paperwork.

The one thing that persists is the list of candy you ticked as "I don't eat this", stored in this browser on this device so you do not have to tick it again next time. Clearing your browser data removes it. Using a different browser or device means it was never there. The privacy page covers the rest of the site.