Queens Strategy Guide

Why Queens rewards method over speed

Every Queens board has exactly one solution, and that solution is always reachable without guessing. This single fact should shape how you play. If a placement is correct, something already on the grid proves it; if nothing proves it, the placement is a guess, and a guess on a single-solution puzzle will either be right by luck or quietly poison the next ten moves. Strong solvers are not faster at seeing answers — they are more disciplined about refusing to place a queen they cannot justify.

The practical consequence is that most of your time should be spent eliminating squares rather than placing queens. A queen is the conclusion of an argument. The argument itself is made out of X marks: squares you have proved cannot hold a queen. Players who mark thoroughly find that queens start appearing on their own, because eventually some region or some line is left with a single open square and the conclusion becomes unavoidable.

The two questions that drive every solve

Almost all progress in Queens comes from alternating between two questions. The first is the region question: does any colour now have exactly one open square left? If so, its queen goes there, because every region must contain exactly one queen. The second is the line question: does any row or column now have exactly one open square left? If so, the same conclusion follows for the same reason.

These two questions look similar but they surface different information, and players who only ask one of them stall constantly. A region can be reduced to one square long before any row is, and vice versa. The habit worth building is deliberate alternation: after each queen you place, run the region sweep, then run the line sweep, then return to the region sweep. Each placement changes the answers to both, so a single pass is never enough.

When both sweeps come back empty, you are not stuck — you are simply not finished eliminating. That is the point at which the more advanced patterns below earn their keep.

Start with the most constrained region

The opening move of a good solve is almost never a queen. It is a decision about where to look. The most productive place is the smallest region on the board, because a region's queen count is fixed at one regardless of its size: a colour with three squares has three candidate positions, while a colour with fourteen has fourteen. Information is concentrated where the options are fewest.

Better still is a region that collapses onto a single line. If a colour lies entirely within one row, its queen must be in that row, which means no other colour may use that row at all — you can immediately cross out every square in it that belongs to a different colour. The same applies to a colour confined to one column. These regions are free moves, and a board that has two or three of them will fall apart quickly once you cash them in.

When no region collapses that neatly, test candidates instead. Take the smallest region, imagine a queen on each of its squares in turn, and apply the eliminations that would follow: the full row, the full column and the eight surrounding squares. If any of those trial placements would leave another region with nowhere to go, that candidate is dead and can be crossed out permanently. Killing even two candidates in the smallest region usually tightens the whole board.

Count the colours on each line

A quick diagnostic that most players never learn: walk along each row and count how many distinct colours it touches, then do the same for each column. The lines touched by the fewest colours are where the next forced move is most likely to hide. If a row meets only two colours, one of those two must own it, and every other colour on the board is competing for the remaining rows — a substantial constraint that costs seconds to spot.

This also tells you where not to look. A row crossed by six or seven different colours is carrying very little information and will usually be among the last lines to resolve. Directing your attention by colour count rather than scanning top to bottom is one of the cheapest speed improvements available, especially on larger grids where a full scan is expensive.

The pairing argument

This is the single most valuable advanced technique in Queens, and it generalises further than most players realise. Suppose two different regions can each only fit inside the same two rows. Between them, those two regions must occupy both of those rows — one each, in some order. It does not matter which way round. What matters is the consequence: every other region on the board is now locked out of both rows entirely, and you can cross out a great deal of the grid in one stroke.

The argument works identically for columns. It also scales: three regions confined to the same three rows take one row each and exclude everyone else from all three; four regions across four rows do the same. In the literature of other logic puzzles this pattern is sometimes called a naked set, and recognising it is what converts a slow, grinding solve into a rapid cascade.

The inverse is equally useful. If some set of rows can only be served by an equal number of regions, those regions are committed to those rows and cannot appear anywhere else. Learning to see the pattern from both directions roughly doubles how often you spot it.

Mark the diagonals, every time

The no-touching rule is what makes Queens different from a simple Latin-square puzzle, and the diagonal half of it is the most frequently forgotten detail in the entire game. When you place a queen, it eliminates its full row, its full column, and all eight squares immediately surrounding it — including the four diagonal neighbours. Three of those eight are usually eliminated anyway by the row and column rules, but the diagonals are not, and they are the ones that get skipped.

If a board suddenly seems to have no available move, the overwhelmingly likely explanation is a missing diagonal mark rather than a deep chain of logic you have failed to see. Before hunting for something clever, audit every queen already on the grid and confirm its complete ring of eliminations is recorded. This one check resolves more stalled solves than any technique on this page.

Adapting to larger grids

On 7x7 and 8x8 boards you can afford to re-scan the entire grid after every placement, and the simplicity of doing so outweighs the small cost. From 9x9 upward, that stops being true: a full sweep becomes slow enough that it dominates your solving time, and most of it is wasted on parts of the board where nothing has changed.

The fix is to work locally. A newly placed queen only creates new information in the rows, columns and regions it actually touches, so check those first. Fall back to a complete scan only when the local pass yields nothing. On 11x11 and 12x12 grids it helps further to divide the board into quadrants and treat them semi-independently, since most deductions at that scale are regional rather than global.

Large boards also punish memory. With 144 squares in play, trying to hold partial eliminations in your head rather than marking them on the grid is the main reason solves go wrong. Mark everything; trust the board, not your recollection of it.

A repeatable routine

Put together, the method looks like this. First, read the board before touching it: note the region sizes, find the smallest, and check whether any region is confined to a single row or column. Second, cash in any free moves those confined regions give you. Third, count colours per line and mark the tightest rows and columns as places to watch.

Then enter the main loop. Place any queen that is forced, mark its full ring of eliminations including diagonals, run the region sweep, run the line sweep, and repeat. When both sweeps come back empty, look for a pairing argument among the narrow regions before considering anything else. When you genuinely cannot find a move, re-audit the existing queens for missing marks rather than guessing.

This routine is slower than intuition for the first few boards and considerably faster after that. It also scales: the same sequence that solves a 7x7 grid in a minute will solve a 12x12 grid without modification, which is not true of guess-and-check. Work through the puzzle library with it and the improvement is usually obvious within a dozen boards.

Common mistakes worth unlearning

Three habits account for most avoidable errors. The first is placing a plausible queen — one that looks right but has no stated justification. On a single-solution board this is always a gamble, and unwinding it later costs far more than the deduction would have. The second is marking lazily, particularly skipping the diagonals, which hides forced moves and produces the illusion of a stuck board.

The third is scanning in a fixed pattern out of habit — always left to right, always top to bottom — regardless of where the board has actually changed. Information in Queens is local and uneven. Letting the last placement tell you where to look next, rather than starting each sweep from the top-left corner, is a small change of habit that pays off on every board you solve.