What the analyzer measures
The analyzer enumerates every legal origin for the selected piece, simulates the placement and any simultaneous clears, then recalculates the shared board-health formula. Ranked results show the resulting health, its change from the current board, and the number of completed units.
The six visible inputs are move availability, largest empty rectangle, open lanes, hardest-piece fit, clear potential, and isolated holes. The weighted score is clamped from 0 to 100. It favors measured survival space, not maximum immediate points.
Four reproducible positions
Isolated-hole trap
Toggle blocks around one empty cell and compare a move that closes the fourth side. The isolated-hole count rises, long lanes can split, and health normally falls unless the placement creates an immediate clear.
Long-bar emergency
Create a crowded board with one contiguous five-cell row segment. Select Long bar and confirm that its fit count is scarce even when the total number of empty cells looks generous.
Multi-clear tradeoff
Build a row and column that share one missing cell. A single piece can clear both. Compare the health after removal with a lower-scoring placement that protects a different difficult shape.
How to interpret a ranking
The first result is the highest under Block Master's published flexibility heuristic. It is not guaranteed to maximize the eventual score because future piece batches are not exhaustively searched. If two results are close, consider immediate score, visible piece order, and whether a planned clear accepts several possible finishers.
Practice the underlying patterns in the authored exercise library, or read the exact formula on the board health methodology page.
Component-by-component interpretation
Move availability compares the number of current legal origins with the number available to the same offered shapes on an empty board. Largest open rectangle finds the biggest all-empty rectangle and divides its area by 81. Open lanes count rows and columns containing a contiguous empty run of at least five cells. These three signals describe space from different angles.
Hardest-piece fit takes the lowest normalized fit ratio among the offered pieces, preventing a very flexible single cell from hiding a blocked long bar. Clear potential divides legal moves that complete a unit by all legal moves. Isolated-hole count finds empty cells surrounded by occupied orthogonal neighbors and applies a visible fixed penalty.
Analyzer edge cases
An empty editor is valid and provides a baseline. A fully occupied editor may return no legal placement, which is a result rather than an error. A piece with no fit displays a direct message instead of an empty table or invalid number. Resetting the board removes all editor cells and recalculates the ranking immediately.
A candidate can complete cells and then clear them, so its final health must be calculated from the resolved board rather than the momentarily placed board. This matters most at intersections. The simulation shares the exact placement and clear functions used by the playable game, keeping the method consistent.
How ranking ties should be used
When several placements produce the same rounded health, the list order is not a hidden claim that the first is superior. Inspect their clears, hole counts, remaining piece fits, and immediate points. Equivalent measured flexibility may support several reasonable moves, especially on an open board.
The analyzer currently compares the selected piece at every legal origin. It does not promise an exhaustive optimal search across unknown future batches. For tray-order training, the authored practice states explain enabling moves and accepted alternatives. For reward comparison, the scoring guide reproduces each score term separately.
Privacy and reproducibility
The editor state is processed in the browser and is not uploaded. There is no screenshot recognition, account history, or private board database. To share a position, describe filled coordinates or recreate the cells manually. This simple model keeps the result reproducible and avoids claiming certainty from an image parser that the first release does not include.
A useful comparison records both the baseline and result. If a move changes health from 54 to 61, note which component caused the seven-point rise. A clear may improve the rectangle while leaving hardest-piece fit unchanged, or it may remove a hole but break a lane. Component evidence makes a ranking explainable and gives the player a specific pattern to test again.