Introduction to the Pyramidal Question

Author: Dwight Wynne

Level: Beginner

This article was originally published through PACE’s now-defunct Freelance Quizbowl University program.


The pyramidal question is the tool of the good question writer. The pyramidal question consists of a series of clues, each pointing to the same answer, arranged in order from most difficult to least difficult. The pyramidal question can thus be viewed as a “pyramid” in one of two ways. The standard representation depicts the question as an inverted pyramid; as more clues are read, the set of answers a player could be considering contracts, until hopefully, only one – the right answer – is left. However, I think of the pyramidal question as more of an upright ziggurat; as more clues are read, the percentage of players who can correctly buzz increases. In this article, we will explore the following questions:

  • What makes a good tossup?
  • Why does the pyramidal question meet each of the requirements for a good tossup better than a short, speed-check question?

Good Tossups Should Be Answered More Often by Players with More Knowledge

Pyramidal tossup (Lily handout)

Tossups are called “pyramidal” because tossups start with hard clues and proceed to increasingly easier clues, allowing players with the deeper knowledge of the topic to buzz in first. Diagram courtesy of Lily Chen’s handout.

This requirement derives directly from the purpose of quizbowl as a knowledge-based competition. Quizbowl is a game in which players demonstrate knowledge. A player with more knowledge of the topic, then, must be more likely to answer the question than a player with less knowledge. Otherwise, there is little to no point in actually asking the questions; the game might as well consist of players pressing buttons as soon as a light goes on.

A good pyramidal question inherently satisfies this requirement. A pyramidal question can be considered a series of short questions, each pointing to the same answer. These short questions are organized in a way such that the hardest question is presented first, with the other questions arranged in order of descending difficulty until you get to the end, which is the easiest question. Thus, a player that can answer a harder question will almost always answer the pyramidal question before a player that can answer an easier question. Because the harder questions typically require greater knowledge to be answered, we can safely conclude that the pyramidal question rewards players with greater knowledge.

What not to do: The Lecture Question

The “lecture” question appears to be pyramidal, because it is as long as a pyramidal question, but it is not. The lecture question includes several lines related to the background behind the answer; in other words, this question reads much like a textbook or lecture with the name of the answer removed. The problem with the lecture question is that the “lecture” clues neither successively taper down the set of possible answer choices nor give additional information that the player may be able to uniquely associate with a particular answer. New question writers should take care to avoid descriptive or background-setting sentences that lack concrete facts; these sentences are useful in a lecture when everyone knows what is being described, but are generally useless in a tossup when everyone is trying to figure out what is being described.

Good Tossups Should Be Answerable by Most Teams

This requirement derives from the concept that quizbowl should be a fun game. No one finds enjoyment in a game in which less than half the tossups are answered. Thus, the question writer shoots for a high target conversion rate; for most tournaments, the benchmark is that the tossup should be answered in 80 to 85 percent of rooms. If a tossup is not answered at a rate comparable to the intended conversion rate, it is usually a poor tossup, regardless of whether or not it fits the other criteria.

The pyramidal question is not inherently an easy question. Like any other question, its difficulty is determined by both the answer and clues selected. However, a one-line question is never easier than a pyramidal question containing the same one line. A question will be answered by the union of the sets of students who recognize and buzz correctly from each clue contained in the question. Since the pyramidal question contains all clues already present in the one-line question, plus additional clues, it must follow that at least as many students can answer the pyramidal question as the one-line question.

It is interesting, then, that many proponents of one-line questions believe that pyramidal questions are “too hard.” There are two simple explanations for this. The first that longer questions and a lack of time limits allow players more time to build up negative impressions of an unanswered question, so players will perceive these questions as harder than the equivalently unanswered one-line question. The second is a corollary that arises from the need of one-line questions to be answered: although a team will always answer more questions in games with pyramidal questions, they will also recognize fewer clues; thus, teams that focus on how many clues they recognize (or nearly equivalently, how long it takes for them to answer the questions) will perceive these questions as harder, even though paradoxically they are answering more of them!

What not to do: The Obelisk Question

Many question writers, even experienced ones, often make the mistake of thinking that if they know the answer from a particular clue, then everyone else will buzz off of that clue. They also make the mistake of thinking that they absolutely must use clues that will stump even the best players. These assumptions lead them to write tossups that have several lines of clues that almost no one will get, followed by a giveaway that seems to be tacked-on as an afterthought. I call these questions “obelisk” questions, because the percentage of players that can buzz stays roughly constant until the end. One suggestion for avoiding this pitfall is to write “from the giveaway up;” that is, consider clues that many players know, then clues that slightly fewer players know, then clues that good players know, then clues that the best players are likely to know. Lead-ins that challenge the best players then get tacked on to the beginning if there is still room in the question.

Good Tossups Should Have the Potential to Educate Most Players

This requirement derives from the concept that quizbowl should both reinforce and stimulate learning. A good question should contain information that a player may not know now, but may find interesting enough to remember. It should also stimulate a player’s intellectual curiosity, such that a player may want to learn more about a specific clue or the answer choice itself.

Every clue in a well-written pyramidal question is designed to both reward the players who have knowledge of the clue and educate the players who do not. Because of the nature of the pyramidal question, only a small percentage of players will answer the question early; even at the end of the question, it is not expected that every player will know the answer. The “extra” lines, then, serve to give additional facts that hopefully both narrow down the possible answer space and pique the player’s interest in the tossup answer. Perhaps players will remember a specific clue that their opponent buzzed from; perhaps they will find an early clue highly memorable; perhaps they will hear the giveaway and remember that they should have remembered which answer went with this clue. In each case, learning is going on, whether a player is first being exposed to the topic, learning a new fact about the topic, or reinforcing a fact once forgotten.

By contrast, the short question provides the potential for education to fewer players. Players who know the one or two clues in the question have the opportunity to learn nothing; while the same can be said for the players who know the opening clue to a pyramidal question, far fewer players answer the opening clue of a pyramidal question than do the opening clue of an equivalent short question.


  • A good tossup satisfies three criteria: it allows players with more knowledge to score more points than players with fewer knowledge, it is answerable by most teams at an event, and it has the potential to educate players who either do not answer the question or use the question as practice.
  • The pyramidal tossup does a better job than the one-line tossup of discriminating among different knowledge levels, because it can be thought of as a series of one-line questions, each successively answerable by players with slightly less knowledge.
  • The pyramidal tossup is always answered at least as often as a one-line tossup containing the same clue.
  • The pyramidal tossup has a much higher potential to educate players than a one-line tossup.

About the Author: Dwight Wynne joined PACE in 2009. He currently writes high school questions for NAQT and HSAPQ.