Pub quiz scoring systems

Flat points, speed bonuses, jokers, doubling rounds and tie-breaks: what each one does to the shape of a night, which to use, and which to leave alone.

Last reviewed

Flat scoring

One point per correct answer, ten points a round, sixty points a night. It is the default for good reason: everyone understands it instantly, it needs no explanation, and nobody can argue about the arithmetic.

Its weakness is that it produces ties and it produces runaway leaders. A table that is good at general knowledge stays six points clear from round two and nothing in the format can catch them, which is dull for everybody including them.

Use it as your base and add exactly one mechanic on top, not three.

Weighted questions

Not every question is worth one point. A harder question in the same round can be worth two or five, and a closing tie-breaker can be worth considerably more.

This is the least intrusive way to stop a runaway leader, because it does not add a rule, it just means the last round can still change the result. A round where the final question is worth five is a round nobody has won until it is over.

The danger is overdoing it. If the last question is worth more than a whole round, the first fifty questions were decoration and the room works that out.

Speed bonuses

Answer faster, score more. It only works with something keeping time, which is why it is rare on paper and common in software.

The effect on a room is real: it stops tables conferring for the full window on every question and it keeps the standings moving, because two tables that both got eight out of ten did not get the same score.

The thing to get right is the floor. A speed system where a late correct answer scores nothing turns the quiz into a reaction test and punishes the table sharing one phone between six. A late correct answer should always beat a fast wrong one. In PubTrivia's time-weighted mode an answer at the buzzer is still worth a fifth of the question, which is the design that keeps knowledge ahead of reflexes.

Do not use it on a charity night or in a club with an older crowd. Those rooms are slower for reasons that have nothing to do with knowing the answer.

Jokers and doubling rounds

Each team nominates one round, before the quiz or before that round, to count double. It is the classic catch-up mechanic and it is genuinely good.

What it does is give every table a decision, which is more engaging than a scoring rule has any right to be. A table that plays its joker on the music round and then watches the music round go badly has had a whole story happen to them.

Two rules make it work: it must be declared before the round starts, and it cannot be the last round. Otherwise it stops being a gamble.

Bonus points, and using them sparingly

A point for the best team name. A point for spotting the connection. A point for the wooden spoon.

They are fine in small numbers and they wreck a quiz in large ones. Once the bonus points outweigh a round, the room stops believing the leaderboard reflects the quiz.

The rule of thumb: all bonuses combined should be worth less than one round.

Handling ties

Decide before the night, announce it if it comes up, and do not invent it on the spot.

A single tie-break question is the best answer. Something with a numerical answer where nearest wins, so it cannot itself tie: how many steps in the Eiffel Tower, how many countries border Brazil.

Counting back through rounds works and is boring. Splitting the prize works and is unsatisfying for everyone. A sudden-death question in front of the room is the only version anybody enjoys.

Negative marking

A wrong answer costs a point. Used across a whole quiz it is miserable: tables stop answering anything they are not sure of, which removes the guessing that makes a quiz fun.

Used for exactly one round, announced clearly, with the option to pass, it is one of the best rounds you can run. The difference is that it is a choice rather than a condition.