Krillion Lab / Core strategy
Balance confidence and rarity
Every answer in a rare-answer game is a small bet. A safe answer pays a little and almost never fails. A rare answer pays a lot and sometimes fails completely. The players who score best are not the ones who know the most obscure facts — they are the ones who judge the bet well.
The simple maths
Think of it as expected value. An answer worth 10 points that you are 95% sure fits is worth about 9.5 points on average. An answer worth 60 that you are 70% sure fits is worth about 42. An answer worth 100 that you are only 30% sure fits is worth about 30 — still more than the safe 10, but only if your 30% is honest.
The trap is not the maths. The trap is how sure you feel. A name that sounds right is not the same as a name you can check. Before you reach, ask: can I say why this fits the prompt, in one sentence? If you cannot, your real confidence is lower than it feels.
When to reach
Reach when the answer passes the one-sentence test and the clock allows it. If you have ten seconds left and a rarer candidate you can defend, take it — the safe answer will still be there if you change your mind, but the rare one will not wait.
Bank when the clock is short or the fit is shaky. With five seconds left, submit the answer you know. A submitted 10 beats an unsubmitted 100 every time, because the unsubmitted one scores zero.
Worked example
Prompt: Name a metal.
- Iron is the anchor: safe, common, about 10 points. You would only submit it if the clock was nearly gone.
- Tungsten is the stretch: a real metal, used in light-bulb filaments, less obvious than iron or copper. You can defend it in one sentence, so it is a good reach — roughly the 30 to 60 range.
- Rhodium is the rare one: a real precious metal, used in catalytic converters. If you know that sentence, it is a strong reach, roughly the 85 to 100 range. If you only half-remember the name and cannot say what it is used for, your confidence is not 70% — it is closer to 30%, and the bet is poor.
Notice what changed between tungsten and rhodium: not the rarity, but whether you could defend the fit. That is the whole decision.
The two mistakes
The first mistake is reaching on a feeling. A word that sounds scientific, foreign, or old is not automatically a valid member of the category. Sounding rare and being rare are different things, and the game only pays for the second.
The second mistake is never reaching. Some players submit the anchor every time and wonder why their depth never improves. If your anchor is worth 10 and you never risk a stretch, your ceiling is 70 points a game — and the players above you are simply the ones who reached twice.
Test yourself
Prompt: Name a play by Shakespeare.
Hamlet is the anchor. Before you submit it, ask the question: can I defend a rarer one in one sentence? Cymbeline is a late romance by Shakespeare, rarely staged — that sentence is your defence, and it is a good reach. If you cannot name the sentence, bank the anchor. If you can, take the deeper dive.
Next: Read the prompt like a checklist · Back to Guides