01What a tie-break actually is
Most quiz questions have a right answer and wrong answers. The tie-break has gradations. Someone is closest; everyone else is somewhere behind them. That single structural difference changes everything about how the question is written, and most setters treat it as an afterthought. They shouldn't.
A tie-break is deployed when two or more teams have finished level, or occasionally as a tiebreaker round in its own right. It asks for a number: a year, a quantity, a distance, a population, a price. The team that comes nearest without any prior knowledge of the answer wins. No buzzers, no conferring, no running total — just a figure on a scrap of paper, handed up, read out. The drama is simple and immediate.
The craft question is deceptively small: which number?
02Why the number matters more than the topic
A bad tie-break number is one that clusters. Ask for the population of the United Kingdom and most teams will converge within a few million of each other, because everybody has a rough sense of the answer. The question rewards nothing, decides nothing cleanly, and if two teams happen to write the same figure you are back where you started. A good number spreads the guesses out, so that genuine reasoning — or a lucky instinct — produces a clear winner.
The ideal figure is one that people can estimate by thinking carefully but cannot recall precisely. The number of steps in the Eiffel Tower. The year a particular treaty was signed. The weight of a blue whale in kilograms. These are questions where someone who knows roughly how to think about the subject will edge out someone guessing blind, but neither can simply retrieve the answer from memory. That is the sweet spot: inferable, not knowable.
Equally, the number must not be so obscure that guesses scatter randomly across several orders of magnitude. If a team writes 400 and another writes 4,000,000, closest-to wins, but the question has produced noise rather than judgment. A setter testing a tie-break should ask: if ten reasonably well-informed people attempted this, would their guesses form a sensible spread across a plausible range? If yes, the question is working.
There is a secondary trap — the answer that sounds specific but admits argument. A question about the "current" population of a city, or a record "as of today," is unusable in a printed or pre-written set, since those figures move. The tie-break needs a fixed, verifiable answer that the host can read out without caveat. Facts with a shelf life are a hazard everywhere in quiz writing, but they are fatal in a tie-break, where the stated answer is the final arbiter and there is no rubric for "close enough on a moving target."
That single structural difference changes everything about how the question is written, and most setters treat it as an afterthought.
03The number nobody disputes
When the host announces the answer, the room needs to accept it instantly. There is no time, in that moment of tension, for a team to say "well, that depends how you measure it." The tie-break question must therefore have exactly one defensible number attached to it. Not a range. Not an approximation. A fact.
This is why setters who write good tie-breaks tend to work backwards: find the verified figure first, then construct the question around it. The number is the foundation; the wording is the superstructure. Reverse the order and you risk discovering, at the worst possible moment, that the figure you chose is contested, dated or genuinely unknown.
A well-made tie-break looks improvised. It isn't.
Absolute Trivia is an independent publication about the craft of writing quiz questions. It is not a quiz service, question bank or league.