Two-String Kite
Short answer: find a row where one shape fits in exactly two cells, and a column where the same shape fits in exactly two cells, such that one end of each lands in the same box. One of the two free ends must hold the shape — so the single cell where their row and column cross cannot.
What a two-string kite looks like
A kite is two strings — one horizontal, one vertical — tied together inside a box. Each string has two ends. The two ends inside the shared box do the work of connecting them; the two ends outside are what the elimination hangs on.
The argument is short. The shared box can hold the shape only once, so at most one of the two ends inside it is real — meaning at least one of the two free ends must be. The target cell sees both free ends, so it cannot hold the shape either way.
How to use a two-string kite
- Find a row string and a column string. Pick one shape. Find a row where it has exactly two homes, and a column where it also has exactly two. Exactly two in each — a third possibility anywhere breaks the argument.
- Check they share a box. One end of the row string and one end of the column string must sit in the same 3×3 box. They must be different cells; if the strings actually cross at one cell there is no kite.
- Find where the free ends cross. Take the row of the column string’s free end and the column of the row string’s free end. That single cell is the target — it sees both free ends.
- Strike the shape there. Remove the shape from that one cell. A kite gives exactly one elimination, but it often lands somewhere no other technique was looking.
Kites and skyscrapers are worth learning together — both are two strong links joined by a weak one, and the search for either turns up the other. The difference is only which units the strings run along.
Two-string kite vs skyscraper
A skyscraper uses two strings of the same kind — two rows, or two columns — joined because they share a line. A kite uses one of each, joined because they share a box. Both end with the same sentence: one of these two loose ends must hold the shape, so whatever sees both loses it. A skyscraper usually eliminates from two cells; a kite from exactly one, but that one is often far from either string, which is what makes it easy to miss and satisfying to find.
Find Two-String Kite in your own grid
Paste a puzzle and this page will look for Two-String Kite in it — the same pattern shown above, in your grid instead of ours. Nothing is uploaded; it runs in your browser.
Frequently asked questions
Why is it called a kite?
Draw the two strings and the shape suggests a kite with its cross-spars and tail. The shared box is where the spars are tied; the two free ends are the corners that do the work. Like most sudoku names it describes the picture, not the logic.
Does the target cell have to be empty of everything else?
No. The target just has to still list that shape as a candidate — it can hold as many others as it likes. The kite removes one candidate from one cell, and whether that solves the cell depends on what else was there.
What if both ends of one string are in the shared box?
Then that string is entirely inside the box, which makes it a locked candidate — a pointing pair. Use that instead; it is simpler and generally eliminates more.
Do I have to check every shape against every row and column?
In principle yes, which is why kites are usually found opportunistically rather than by exhaustive search. The practical habit is to notice, while hunting X-Wings, that you already know every row and column where a shape appears exactly twice — the kite check is then almost free.