The daily case is tuned so that a reasonably plugged-in player solves it most days. That is the correct default - a daily game that beats you routinely stops being a habit. But if you are landing it on guess one or two consistently, the game has two harder settings, and they are not just the same puzzle with less room.
Toggle it above the case. You get two strikes instead of three, and the optional magnifying-glass clues are switched off entirely.
The strike reduction is the obvious change. The clue removal is the important one.
In the standard game, a wrong guess is a purchase: you lose a strike but unlock a free clue, which makes deliberately guessing to eliminate a genuinely strong strategy. You can afford to spend a turn on information.
With two strikes, that economy collapses. Spend your first guess on information and you are one mistake from losing with only one clue in hand. Hard Mode rewards solving the redaction cold - working out who the description can honestly apply to before you type anything at all.
Hard Mode keeps its own separate streak, so experimenting with it will not endanger the streak you have been defending on the daily.
After dark, a second case unlocks: a different puzzle from the daily, drawn from further along the schedule, run at two strikes with no optional clues.
It exists for a specific reason. The archive is for people who found the game today and want more right now. The Night Edition is for people who already played today, solved it, and want another - without breaking the one-a-day contract that makes the daily matter. Different puzzle, separate streak, no effect on your main record.
It is generally harder than the daily, and deliberately so. If the daily is calibrated for a broad audience, the Night Edition is calibrated for someone who has already proved they are quicker than that.
Front-load the redaction. In the standard game you can afford to read the whole case, guess, and learn. At two strikes, the descriptor - the champion, the leading man, the idol - has to do most of the work before you commit. Count the people it can honestly describe. If that number is above ten, keep reading rather than guessing.
Date it before you name it. Every case is full of objects that carry a year: a technology, a venue, a broadcast format, a specific sum of money. Establishing the decade first collapses the candidate list far more reliably than free-associating on names.
Do not spend a strike on a hunch. This is the single biggest difference. With three strikes and free clues, a speculative guess is cheap and sometimes correct. With two, a speculative guess is half your game for one clue. Wait until you can name two plausible candidates, then guess the one whose elimination tells you more.
Watch for the Good News Desk. Roughly a third of the bank is about documented good behaviour rather than scandal, and those cases are consistently the hardest because the reflex is always to guess someone with a bad reputation. In Hard Mode that reflex costs you half the game immediately. If a case is conspicuously short on wrongdoing, flip your candidate list before you type.
If you are clearing Hard Mode and the Night Edition comfortably, the practice archive holds every case that is not scheduled soon, at full difficulty and with no effect on any streak. And the Morgue carries every closed case with its answer and the full documented story behind it, which is the only part of this game that is genuinely educational.
Today's edition is here. The toggle is above the case.