Secret Santa When Couples Are Involved
Add exclusions and the draw can stop having any valid answer. Here is where that line sits.
Couple exclusions turn a gift exchange from a shuffle into a constraint problem, and the surprise is which direction the difficulty runs: couples alone almost never make it unsolvable, while a small group with a few overlapping exclusions can have no valid draw at all.
Excluding couples from drawing each other turns a shuffle into a constraint problem, and the thing everyone expects to go wrong mostly does not: couples alone almost never make a draw impossible.
What does make it impossible is a group of people who between them have too few permissible recipients — and that is worth being able to recognise, because no sequence of re-rolls will find an answer when it happens.
Where the line is
The intuition most people have is that couples make it hard, and they are wrong. Two couples, three couples, four couples — all have valid arrangements. Six people in three couples can be assigned as two three-cycles, so nobody draws a partner and nobody draws themselves.
There is exactly one all-couples case that fails, and it is the trivial one: a single couple on their own. Each person's only possible recipient is the other, and there is nowhere else to go.
What genuinely breaks a draw is a bottleneck — several people funnelled onto the same recipient. The smallest case needs four participants and three rules: if Ana, Ben and Cleo are all barred from one another, every one of them has to draw Dev, and Dev can only receive from one person. No assignment exists however long you search.
Couples against group size
- 6 people, 3 couples: no valid arrangement exists.
- 7 people, 3 couples: solvable, and found within a few attempts.
- 8 people, 4 couples: solvable, though the valid fraction is small.
- 10 people, 5 couples: solvable comfortably.
How to tell in advance
The rough rule is that trouble arrives when excluded pairs make up a large fraction of the group. If more than about half the participants are in an excluded pair, expect the search to struggle; if the group is entirely couples and small, expect it to fail.
The tool reports the attempt count when it succeeds, and that number is the early warning. A valid arrangement found on the second attempt means the constraints are loose. One found on the four-thousandth means you are one exclusion away from having no answer at all.
Watching that number over successive years is genuinely useful for a recurring exchange, because the exclusion list tends to grow — new couples form, and people add last year's pairings — and the point where it becomes unsolvable arrives without warning otherwise.
What to do when there is no answer
The options are to remove an exclusion, add a person, or split the exchange. Removing an exclusion is usually easiest, and the one to remove is whichever person is most constrained — find whoever has the fewest permissible recipients and drop one of their rules, because that is where the shortage lives.
Adding a person works and is often available — a partner who was not going to participate, a colleague from the next desk. It is the cleanest fix because it does not require anyone to give up a preference.
Splitting into two exchanges works for larger groups and changes the event. Two exchanges of six feel different from one of twelve, and the difference is not obviously worse.
- Remove one exclusion — usually the easiest and least contentious fix.
- Add one unconstrained person — cleanest, because nobody gives anything up.
- Split into two exchanges — works for larger groups, and changes the event.
- What does not work: re-rolling. There is no arrangement to find.
Adding last year's pairings on top
The most requested feature for a recurring exchange is not repeating last year's pairings, and it is implemented the same way — list each person's previous recipient as an exclusion. Because exclusions are symmetric, that also prevents the reverse pairing.
Doing both at once compounds fast. A group of ten with four couples and last year's pairings excluded has each person barred from themselves, their partner, whoever they bought for, and whoever bought for them — four exclusions from a field of nine.
That is usually still solvable and it is close enough to the edge that the attempt count is worth watching. If it starts running into the thousands, drop the previous-year constraint and keep the couples one, which is the exclusion people actually care about.
Frequently asked questions
Why does it say there is no valid answer?
Because some group of your people has fewer permissible recipients between them than there are of them. Couples alone rarely do it; a three-way mutual exclusion in a small group does.
Will re-running find one?
No. There is nothing to find. The tool exhausts five thousand attempts precisely so it can say that rather than looping.
How do I fix it?
Remove an exclusion, add one unconstrained person, or split into two exchanges. Adding a person is the cleanest because nobody gives anything up.
How can I tell it is getting close to unsolvable?
Watch the attempt count. Found on the second attempt means the constraints are loose; found on the four-thousandth means you are one exclusion from failure.
Do I need to list an exclusion on both people?
No. Exclusions apply in both directions automatically, because that is what naming a partner means in practice.
Can I also exclude last year's pairings?
Yes, list each person's previous recipient as an exclusion. It compounds with the couples rule, so watch the attempt count.
Which constraint should I drop first if it fails?
The previous-year one. Couples buying for each other is the exclusion people actually care about.
Is eight people in four couples solvable?
Yes, though the valid fraction is small. Ten in five couples is comfortable.
Tools for this job
Gift Exchange with Exclusions
Secret Santa where couples, housemates, or teammates must not draw each other.
Secret Santa Generator
Generate fair Secret Santa pairs — nobody draws their own name, nothing stored.
Random Pair Generator
Split a list into pairs — with an honest answer for what happens on an odd count.
Background reading
Constraint Satisfaction in Fair Draws
What happens when the rules and the randomness disagree — and when there is no answer.
Derangements and the Secret Santa Problem
Why a plain shuffle fails a gift exchange 63% of the time, and what a derangement is.
Bracket and Fixture Scheduling Maths
Byes, seeding, and the circle method that fits a full round-robin into n-1 rounds.