How many points is a One in a Krillion answer?

100. Deep Cut is 85, Rare 60, Schooler 30, Too Clever 15 and Plankton 10.

What is a good Krillion score?

The game rates 251–350 as Rare, 351–449 as Deep cut and 450 or more as One in a krillion. The maximum is 700.

What does “Too Clever” mean in Krillion?

In the developer's words, it is the go-to “obscure” pick that isn't actually that obscure. It scores 15 points.

Krillion scoring: tiers, points and what a good score is

Every accepted answer in Krillion belongs to one of six tiers, and the tier sets the points: 10, 15, 30, 60, 85 or 100. Seven prompts make a maximum of 700.

What are the tiers and points?

Tier	Points	The game’s description
One in a Krillion	100	The designated gem. Trench floor.
Deep Cut	85	True obscurity. Few go this deep.
Rare	60	Genuinely uncommon. Nice pull.
Schooler	30	Solid — swims with the school.
Too Clever	15	A famously “obscure” pick. Everyone reaches for it.
Plankton	10	The answer everyone blurts out.

What is a good score?

The day’s total is rated like this:

Total	Verdict
450+	One in a krillion. Trench dweller. Absurd.
351–449	Deep cut. Seriously deep dive.
251–350	Rare. Below the thermocline. Nice.
151–250	Schooler. Swimming with the school.
0–150	Plankton. Plenty of ocean left down there.

Seven deep cuts make 595. Reaching 450 takes an average of about 64 a prompt, so mostly Rare or better.

How is rarity decided?

From the official FAQ: some prompts are scored mechanically against a database of how often a word or name turns up in what people read and write. Others get “a combination of LLM assistance and a manual pass”.

Nothing in the FAQ says the score depends on what other players typed that day.

Why is the Krillion answer sometimes less rare than a deep cut?

Because it is not chosen on rarity alone. The developer:

Past a certain depth, rarity gets hard to disambiguate, so the gem is also picked for fun factor: the answer that’s most satisfying to land on.

85. On November 14, 2026, 109,024 of the 110,840 accepted answers were deep cuts. The hard part is thinking of one in 25 seconds.

Why was my answer not accepted?

The developer’s answer: “Because I didn’t find it, sorry.” Answer sets are built by merging datasets such as dictionaries, wiki pages and encyclopedias, and a fair answer is sometimes missing from the sources.