Krillion Rarity Scoring Explained: How to Earn a Deeper Score

Krillion rewards two things in order: a valid answer first, then an answer's editorial rarity. Every accepted response belongs to a 20, 40, 60, 80, or 100-point tier. Familiar answers keep you moving; correct answers that are easier to overlook can send your Daily Dive deeper.

This guide explains what the tiers mean, what they do not mean, and how to make better scoring decisions during each 25-second round. If you are new to the format, begin with the complete how to play Krillion guide.

The five answer tiers

Points	Practical meaning	Best use during a round
20	The most immediate or widely recalled valid answer	Lock it when confidence matters more than upside
40	A familiar alternative that is not always the first response	Choose it when recall is clear and fast
60	A less obvious but still accessible answer	Use it when you can verify every part of the scope
80	An overlooked answer within the reviewed set	Take it when your knowledge is specific, not speculative
100	The deepest editorial rarity tier	Reserve it for an answer you genuinely know

A response outside the reviewed scope receives zero points, no matter how unusual it sounds. The score is therefore not a reward for obscurity alone. It rewards uncommon and correct recall.

Editorial rarity, not a live popularity percentage

Krillion's current scoring tiers are editorial classifications. They express a considered estimate of which valid answers are likely to be obvious and which are likely to be overlooked.

That distinction matters. A label such as 80 points does not mean “only 20% of players chose this” or represent a real-time survey. It means the answer was placed in a higher rarity tier when the reviewed answer set was prepared.

Clear wording protects the integrity of the game: players can understand the scoring model without being led to believe that Krillion is publishing live statistical data it does not currently collect.

How a question becomes scoreable

A strong open-answer question needs more than a prompt. It needs a boundary that can be checked consistently.

Consider: Name one official language of the United Nations. The United Nations identifies six official languages—Arabic, Chinese, English, French, Russian, and Spanish—in its multilingualism resources. That creates a closed, auditable answer set.

Krillion then needs to:

define the scope in plain language;
identify every accepted entity in that scope;
add reasonable aliases and formatting variants;
assign a rarity tier to each accepted answer; and
attach a short explanation so the player understands the result.

This process is why precision matters. “A UN language” could be interpreted broadly; “one official language of the United Nations” establishes a specific list.

Why rare is not always better

Imagine that you immediately know a 20-point answer. You suspect there is a rarer alternative, but you cannot fully recall its name. Should you keep searching?

Use expected value, not pride:

A locked 20-point answer is worth 20.
A guessed 80-point answer is worth 80 only if it is valid and recognized.
A timeout or answer outside the scope is worth 0.

You do not need a formula during play. Ask two questions instead:

Can I state the rarer answer exactly?
Can I explain why it meets every qualifier in the prompt?

If either answer is no, lock the safer response. Deep scores are built from accurate decisions across seven rounds, not one dramatic guess.

Examples of scope changing the score decision

Planets with rings

Saturn is the immediate answer, but NASA documents ring systems around Jupiter, Uranus, and Neptune as well. If the prompt asks for any planet in our solar system with rings, all four can be eligible. If it asks for the planet with the most visually prominent rings, the scope would be different.

Countries crossed by the Equator

The phrase “whose land is crossed by the Equator” excludes a country when the line passes only through its territorial waters. A player who ignores land may retrieve a geographically related answer that does not satisfy the rule.

Films directed by a named creator

“A Studio Ghibli feature directed by Hayao Miyazaki” requires both conditions: it must be a Studio Ghibli feature and Miyazaki must be credited as director. A film connected to him in another role does not automatically qualify. The official Studio Ghibli works catalogue is the kind of primary reference used to establish that boundary.

These examples reveal the core skill: rarity scoring measures recall inside a defined set, not free association around a topic.

How points become depth

Your seven answer scores are added together and translated into a depth zone:

Sunlight, 250 m: 0–219 points
Twilight, 500 m: 220–359 points
Deep Sea, 750 m: 360–499 points
Abyssal, 1,000 m: 500 points or more

Depth does not change the value of an individual answer. It gives the final total an easy-to-read identity. A player reaches the Abyssal zone by averaging a little over 71 points across seven questions, so consistent 60- and 80-point recall matters more than relying on a single top-tier answer.

Three ways to improve your score

Build answer clusters

When you learn a fact, store it as part of a group. Do not remember only Saturn; connect it with Jupiter, Uranus, and Neptune as the four ringed planets. Do not remember only Apollo 11; connect it with the six Apollo missions that completed crewed lunar landings.

Closed lists are powerful because each familiar item becomes a cue for the less obvious ones.

Separate recall from validation

First generate candidates, then check them against the wording. Mixing those steps can cause you to reject a good answer too early or lock a related answer that misses one qualifier.

A compact loop works well:

recall two or three candidates;
test each candidate against the scope;
choose the least obvious candidate you can defend; and
lock before time pressure disrupts the decision.

Review misses by type

Not every zero has the same cause. After a dive, ask whether you:

did not know the relevant fact;
knew it but could not retrieve it in time;
overlooked a qualifier;
used an unsupported alias; or
traded a safe answer for an uncertain rare one.

Each cause suggests a different improvement. Knowledge gaps need study; retrieval gaps need practice; scope errors need slower reading.

A fair way to think about a “good” score

There is no single total that proves someone is good at trivia. A Daily Dive samples seven categories, and category mix affects every player differently. A useful personal benchmark is therefore process-based:

How many answers were valid?
How often did you correctly identify the scope?
Did you find a second candidate before locking?
Did a riskier choice improve the score or create a zero?

Track those decisions over multiple dives. The total is the headline, but the answer log tells you how to improve.