That's a Totally Normal Thing to Say! (Trivia Extravaganza 2026)

David Imel introduces a rotating two-person team setup where both teammates score if correct, but if they miss, the “odd man out” gets the point—changing incentives and encouraging strategy (including joking about trying not to answer).

In Keyword Search, teams can’t answer directly; they must propose three search terms that reveal the answer on Google’s first page while avoiding words from the question/answer. The round showcases how hard “findable” knowledge becomes under constraints (e.g., Apple Vision Pro was repeatedly hard to surface without banned terms).

The quote game centers on identifying who said out-of-context lines and offers extra points for tweeting them verbatim; the group debates how risky/awkward it would be to tweet CEO quotes without context and largely declines.

In the Taboo-style segment, one player gets a secret term via Slack and must give three clue-words (not the term) for a single guess. The round highlights both ambiguity (e.g., “Arm” guesses gone wrong) and how extremely “obvious” references can be (e.g., “Worst product ever / Humane AI Pin” getting instantly solved).

Reddit Feud uses sentiment analysis of comment frequency (not upvotes) to build a top-10 board; the hosts argue briefly about whether ignoring upvotes reflects Reddit accurately, but proceed because frequency better “distills” the hive mind.