Every serious puzzle solver has had this experience: you're working through a difficult puzzle, you've exhausted every technique you know, and you genuinely cannot find a logical next step. The natural first question is "am I missing something?" — but sometimes the honest answer is "no, this puzzle is broken." Distinguishing between a puzzle that requires a technique you haven't learned yet and a puzzle that requires an unjustified leap is one of the most important skills in a solver's toolkit. Not just for protecting your confidence, but for ensuring you spend your practice time wisely.
What Makes a Puzzle Hard
A hard puzzle requires you to apply techniques at the edge of your current ability. Every step is logically justified — each placement follows necessarily from the given information, and each elimination is provable. The path to the solution may be long, unintuitive, or require techniques you haven't encountered before. But the information is always present. You might not have the knowledge or pattern recognition to see it yet, but a solver with sufficient expertise can navigate from the given state to the solution through pure logic, step by step.
Hard puzzles are characterized by the feeling of "I must be missing something" combined with the eventual experience of "oh — there it was." The solution, when found or revealed, feels like it was inevitable in retrospect. The information was always there. You just needed to see it.
What Makes a Puzzle Bad
A bad puzzle requires you to guess, assume information not provided, or accept an arbitrary choice at some point in the solution. In Sudoku, a bad puzzle has multiple valid solutions — which means no amount of logical deduction will uniquely resolve the ambiguous cells. In crosswords, a bad puzzle has a square where two different letters produce valid words in both directions — so the solver has no way to determine the correct answer from the clues alone. In logic grid puzzles, a bad puzzle has missing constraints that leave one or more assignments undetermined.
Bad puzzle: You're stuck for 20 minutes. You try every technique. You guess. The guess works. But you'll never know if the other option would also have worked. Satisfaction level: hollow.
How to Tell the Difference
When you're genuinely stuck, the test is whether bifurcation on a two-candidate cell produces a unique outcome. If you try both options for a cell with exactly two candidates and one leads to a contradiction while the other completes the puzzle cleanly, the puzzle is hard but valid — you've found the solution through controlled trial. If both options seem to lead to valid completions, the puzzle likely has multiple solutions and is poorly constructed. If neither option leads anywhere definitive without further guessing, the puzzle may have design flaws in that region.
For crosswords, the test is whether ambiguous squares have crossing clues that unambiguously point to one letter. If two letters both satisfy both crossing clues, the puzzle has a "natick" — an unfair crossing named for a notorious New York Times incident. Good constructors and editors catch these. Bad constructors don't.
What to Do With a Bad Puzzle
The answer isn't to get angry or blame yourself. It's to note the source and adjust your expectations accordingly. Published puzzles from reputable outlets (major newspapers, established puzzle books, competition sets) are almost never genuinely bad — if you can't find a logical path, you're almost certainly missing a technique. Self-published puzzles, app-generated puzzles, and low-quality puzzle books sometimes contain genuine design flaws.
When you encounter what might be a bad puzzle, verify with a solver tool if one is available. If the solver finds multiple solutions or confirms the need for guessing, you've identified a bad puzzle and can move on guilt-free. If the solver shows a unique logical path, you've identified a knowledge gap — and that's valuable information too.