Roughly a third of the ALLEGEDLY* puzzle bank is not scandal at all. Same sourcing rule, same standard of proof, opposite valence: documented acts of generosity, absurd competence, or quiet decency that never got the coverage the disasters did.
They are also, consistently, the hardest puzzles in the game. Not because the people are more obscure - several are among the most famous humans alive - but because of what players do the moment they finish reading.
Give someone a redacted story about a famous person and their first guess is almost always somebody with a bad reputation. Not the person the details point to. The person the genre points to.
This happens even when the story contains no wrongdoing whatsoever. A case about someone quietly funding something extraordinary still draws guesses aimed at the reliably scandalous, because the frame - redacted name, breathless tone, newsprint styling - has already told the reader what kind of story this is. The frame overrides the text.
What makes this visible is the guess history. When the answer is someone who did something genuinely good, the list of wrong guesses sitting above it reads as an accidental confession about how the reader thinks. Nobody set out to accuse those people of anything. The reflex did it for them.
A good redaction is accurate but incomplete: the troubled leading man, the heavyweight champion, the baby-faced pop idol. Each of those is true of the subject. But readers do not treat a description as a filter - they treat it as a character, and characters come pre-loaded with plots.
"Troubled leading man" narrows the field honestly. It also imports an entire expected storyline, and once that storyline is running, contrary evidence in the text gets read as detail rather than contradiction. The most common way to lose a Good News Desk case is to notice the absence of wrongdoing and assume it is coming later in the sentence.
The single most useful habit for these puzzles: when a story is conspicuously short on wrongdoing, that is not a missing detail. That is the puzzle telling you which desk it came from, and the entire candidate list flips.
Scandal cases are dense with specifics that narrow fast - a charge, a venue, a fine, a date. Good News cases tend to be lighter on that scaffolding, because the underlying event was never processed by a court system that generates paperwork. If you are three paragraphs in and nobody has been arrested, stop hunting for the crime and start asking who would plausibly have done something remarkable and not made it about themselves. That question has a much shorter answer list than it feels like.
There is a straightforward design answer and a less comfortable one.
The design answer: a game made only of disasters gets monotonous fast. Once every case is a crime, the crime stops being information - it is just the ambient condition, and every puzzle collapses into "which famous person is the bad one today." Mixing in cases where the answer is admirable keeps the reader genuinely uncertain about what kind of story they are reading, which is the only thing standing between this and a quiz.
The less comfortable answer: a game built entirely on documented misbehaviour, played daily, does something to how you think about famous people. Not dramatically. Just a gentle, repeated nudge toward the belief that the interesting fact about anyone well-known is the worst thing they ever did. The Good News Desk is a deliberate correction to that - and the fact that those cases are the hardest ones is itself the most interesting result the game produces.
It would be easy to relax the sourcing rule for nice stories. Nobody sues over being called generous, and the legal exposure that makes a scandal case dangerous does not exist here.
They are held to the same standard anyway: a documented act, verifiable, on the public record. Partly because a puzzle whose answer might not be true is a broken puzzle regardless of whether it flatters the subject. Mostly because inventing flattering stories about real people is the same act as inventing damaging ones, and the fact that one feels harmless is not a reason to do it. The full rule is in our editorial standards.
New case every midnight. Today's edition is here - and if it seems suspiciously short on wrongdoing, you now know what that means.