5x5 Lights Out Solution: Chase Method and Correction Table

5x5 Lights Out Solution: Chase Method and Correction Table

Created Aug 23, 2026 Updated Aug 23, 2026

The 5x5 board is the default size on Lights Out and the one most people mean when they picture the classic version of this puzzle. It is also a genuinely trickier size than it looks: of all the ways you could theoretically light up a 5x5 grid, only exactly one in four can actually be cleared by any sequence of presses at all. The good news is that every board the game itself deals you through the New Game button falls inside that solvable quarter, guaranteed โ€” and this guide gives the exact row-by-row method and correction table to clear it.

What follows is the same chase method used across every board size on this site, worked out specifically for 5x5: how many neighbors each tile has, the correction table for the top row, a full worked example, and the exact math behind that one-in-four figure. For the general method behind this table, see the Lights Out solution algorithm; for the smallest, fully solvable board size, see the 3x3 Lights Out solution.

The 5x5 Grid and Its Neighbors

A 5x5 board has four corner tiles with two neighbors each, twelve edge tiles (three per side, excluding corners) with three neighbors each, and nine interior tiles with the full four neighbors. Pressing a corner toggles three lights, pressing an edge tile toggles four, and pressing an interior tile toggles five. With twenty-five tiles total, the Lights Out board at this size is large enough that chasing individual lit tiles instead of working row by row gets unmanageable fast โ€” the row-by-row method matters more here than it does on the smaller 3x3 board.

Step 1: Chase Down Without Touching the Top Row

Leave the top row exactly as the puzzle gave it to you. For every tile lit in the top row, press the tile directly beneath it in row two. Then look at row two as it now stands and, for every tile still lit, press the tile beneath it in row three. Continue the same way through row three into row four, and row four into row five. By the end, rows one through four should be completely dark, and whatever remains lit in row five โ€” the bottom row โ€” is your leftover pattern.

Step 2: The Top-Row Correction Table

A 5x5 board has 32 possible ways to press the top row, but because the puzzle's underlying structure has a couple of built-in redundancies at this size, those 32 choices only ever produce eight distinct leftover patterns in the bottom row โ€” not 32. That number is not a coincidence: it is exactly why only one in four of all possible 5x5 patterns is solvable in the first place, and it means every leftover pattern you will ever actually see from a real, solvable board matches one of the eight rows below.

Lit in the bottom row after Step 1Press these tiles in the top row
Nothing lit (board already clear)Nothing โ€” you're done
1st, 2nd, and 3rd tiles2nd tile only
2nd and 4th tiles1st, 2nd, and 3rd tiles
1st, 3rd, and 4th tiles1st and 3rd tiles
1st and 5th tiles1st and 2nd tiles
2nd, 3rd, and 5th tiles1st tile only
1st, 2nd, 4th, and 5th tiles3rd tile only
3rd, 4th, and 5th tiles2nd and 3rd tiles

Tiles are counted left to right, 1st being the leftmost column and 5th the rightmost. If a real, playable board ever leaves a leftover pattern in the bottom row that is not one of these eight, something was pressed off-plan earlier in the chase โ€” recheck rows one through four before assuming the board itself is broken.

Step 3: Press the Corrected Row and Chase Again

Return to the board's original starting state, not the state left over from Step 1. Press the top-row tiles the table calls for first. Then repeat the same forced chase: press row two beneath every top-row tile still lit, row three beneath every row-two tile still lit, and so on down to row five. A solvable board โ€” which, again, is any board this game actually deals you โ€” finishes completely dark.

Worked Example

Take a scrambled 5x5 board and run Step 1 without touching the top row. After chasing all the way down, suppose the bottom row ends up with its 2nd, 3rd, and 5th tiles lit and the rest dark. Checking the table, that leftover pattern calls for pressing only the 1st tile of the top row.

Go back to the original board and press just that one top-left tile. Chasing down from there โ€” pressing row two beneath whatever is lit in the corrected row one, row three beneath row two, row four beneath row three, and row five beneath row four โ€” clears every row in turn, including the last one. The specific tiles pressed differ from board to board, but the pattern is always the same: one top-row press chosen from the table, followed by a forced, mechanical sweep down through the other four rows with no further decisions to make.

The One-in-Four Rule, and Why It Doesn't Matter for New Game

Testing every one of the roughly 33.5 million possible ways to light up a 5x5 grid confirms the exact fraction: exactly 8,388,608 of them, or one in four, can be cleared by any sequence of presses. The other three in four are simply impossible, no matter how skilled the solver or how many moves are tried.

That statistic only applies to patterns built by hand or picked arbitrarily, though โ€” for example, a pattern copied from a photo or assembled tile by tile to look a certain way. The board Lights Out actually gives you when you press New Game is built differently: the game simulates real random button presses starting from a fully dark board, and any board reached that way is, by construction, always inside the solvable quarter. Running 50,000 simulated New Game boards through the chase method and table above solved every single one of them โ€” a 100 percent success rate, exactly as the math predicts.

Reading the Board Before You Start Pressing

Before running Step 1 on a fresh 5x5 board, it helps to just count: how many tiles are lit, and roughly how they cluster. A board with lights scattered evenly across all five rows takes the same number of steps to chase as one with a tight cluster in a corner, but a cluster near the top row is worth a second look, since a top-row correction that happens to cancel out several of those lights at once can make the second pass noticeably shorter than the first. None of this changes the method itself โ€” Steps 1 through 3 apply exactly the same way regardless of how the initial lights are arranged โ€” but a quick scan before the first press helps catch a misclick early, before it propagates all the way down four more rows.

Frequently Asked Questions

Is every 5x5 Lights Out board solvable?

No โ€” only one in four of all mathematically possible 5x5 patterns can be cleared. Every board generated by the game's own New Game button is solvable, though, since it is built from real simulated button presses rather than an arbitrary pattern.

Why does the correction table only have eight rows instead of 32?

Because the 32 possible top-row presses only produce eight distinct outcomes in the bottom row at this size โ€” several different top-row choices land on the same leftover pattern, so the table only needs one entry per distinct outcome.

What if I built my own 5x5 pattern and it won't solve?

That is expected roughly three times out of four for a pattern that wasn't generated by real button presses. Try New Game for a board guaranteed to be solvable, or build one yourself by pressing random tiles from an already-clear board.

Is 5x5 harder than 3x3 and 6x6?

In one specific way, yes: 5x5 is the only size on Lights Out where a randomly built pattern is more likely to be unsolvable than solvable. Every 3x3 and 6x6 pattern is solvable regardless of how it was built.

Do I need to memorize all eight table rows?

No. Keep this page open and match your leftover pattern against the table as you play โ€” the method works the same whether you have it memorized or not.

Related Guides