Corresponding Squares

Corresponding squares are pairs of squares in king endgames linked by a strict duty: when the attacking king stands on one, the defending king must stand on the partner square — and whoever is forced to leave the correspondence first loses the fight. The familiar opposition is simply the easiest special case of this deeper theory, which governs every blocked pawn ending.

If zugzwang is the engine of pawn endgames, corresponding squares are the engineering manual. They turn the vague sensation of "my king needs to be on the right square at the right time" into something you can actually compute at the board — and in locked positions, computing it is the difference between a database win and a database draw.

The Idea in One Position

Start with the simplest meaningful example, one I have verified to the last tempo. White: king on e5, pawn on e4. Black: king on e7. The pawn's key squares — the squares whose occupation by White's king wins by force — are d6, e6, and f6.

Now look at the geometry. From e5, White's king attacks all three key squares at once: d6, e6, and f6 are all one step away. Which black squares guard all three? Test them: from d7 the king covers d6 and e6 but not f6; from f7 it covers e6 and f6 but not d6. Only e7 guards all three entry points. So e5 and e7 are corresponding squares — when the white king arrives on e5, the black king must already stand on e7, or a key square falls next move.

And here is the punchline, straight from the definition of correspondence: it matters who arrives last. With White on e5 and Black on e7:

• White to move: draw. The correspondence holds. 1.Kd5 is answered by 1...Kd7 (mirroring to the d-file pair), 1.Kf5 by 1...Kf7, and White never enters a key square. The white king cannot even gain a tempo with the pawn — its own body blocks e4-e5.
• Black to move: White wins. Black must abandon e7, the only square that guards all three doors. After 1...Kd7 2.Kf6, or 1...Kf7 2.Kd6, the king walks into a key square and the pawn promotes by force.

Same squares, opposite results. That is a position of mutual zugzwang, and corresponding-squares theory is precisely the map of where all such positions lie.

Opposition: The Trivial Case

When students meet the opposition — kings facing each other with one square between, the player not to move holding the advantage — they are already using corresponding squares without knowing it. In open king-and-pawn endings the correspondence network is perfectly regular: e5 partners e7, d5 partners d7, f5 partners f7, and so on, each pair separated by the familiar one-square gap. "Take the opposition" is simply the compressed instruction for maintaining these correspondences, and the distant opposition (same line, an odd number of squares between the kings) is the same network extended across the board.

The theory earns its own name the moment the board stops being regular. Put blocked pawns, barriers, and detours in the kings' way, and the tidy mirror geometry bends: a defender may need to answer a king on a5 by standing on c7, or answer two different attacking squares from one post. Opposition is what correspondence looks like on an empty board; corresponding squares are what opposition becomes in traffic.

How to Compute Correspondences in a Blocked Ending

Here is the method I teach, in the order you should apply it at the board. It is slow the first few times and astonishingly mechanical after that.

1. Find the entry points. In a blocked or semi-blocked pawn structure, list the squares where the attacking king would actually break through — the key squares in front of a potential winning pawn, or the squares from which it attacks an unprotectable pawn. Everything else is scenery.
2. Find the defender's guard posts. For each entry point or cluster of entry points, identify the defensive squares that cover them. Where one defensive square covers several entries — like e7 covering d6, e6, and f6 above — you have found a critical correspondence.
3. Work backward through the approach routes. Now label the squares one step further from the action: if the attacker's square X reaches attacking squares A and B, the defender's partner for X must be adjacent to the partners of A and B. Propagate this logic outward, square by square, and the whole board acquires a system of numbered pairs.
4. Count the spare squares. This is where games are decided. If the attacker has more equivalent squares in some region than the defender has partners, the attacker can lose a tempo — walk a little triangle — and force the defender out of correspondence. Triangulation is exactly this: exploiting a surplus of squares to hand the opponent the move. If the network is balanced, the position is a draw and no amount of shuffling will change it.

The most famous illustration is the Lasker–Reichhelm position (composed by world champion Emanuel Lasker with Gustavus Reichhelm, 1901): a fully blocked pawn ending where White to play wins only by a long, exact king maneuver, with the kings dueling across the entire width of the board — squares on the queenside corresponding to squares near the kingside in ways no opposition rule could ever describe. Any single inexact step by White and the win evaporates. It is the standard demonstration that in locked structures, correspondence — not intuition — is the last word.

When Corresponding Squares Decide Games

Full correspondence analysis is overkill for most practical endings — if the position is open, the square rule and ordinary opposition answer things faster. The theory matters in a specific family of positions, and it pays to recognize them on sight:

• Blocked pawn chains with one entry point on each wing. The defending king must commute between two duties; the attacking king probes at maximum distance. Whether the defense holds is a pure correspondence count, and "he just barely gets back in time" is exactly the calculation the pairs formalize.
• Fortress-like structures. A fortress in a pawn ending survives only if the defender has a guard post answering every probe — that is, only if the correspondence network is complete. Sieges of such positions are won by finding the one probing square the defender cannot answer, or by triangulating where you own a spare square.
• Mutual-zugzwang landmarks. Endgame theory is dotted with positions where whoever moves loses — the trébuchet with its mutually attacked pawns is the tiny classic. Each such landmark is the center of a correspondence system, and precise players maneuver toward these positions with the opponent to move, using the pairs as the route map.
• Reserve tempi change everything. A single spare pawn move — an untouched a2-pawn, say — lets you break any correspondence once: you make the pawn move, and the opponent must leave their post. Before diving into king maneuvers, always count pawn tempi on both sides; a correspondence battle with a tempo in your pocket is a battle already won. This is the same bookkeeping that governs tempo play everywhere in chess, at its most naked.

One practical warning from years of coaching: knights cannot lose a tempo, and neither can a king restricted to two squares. When you compute that your opponent's king has exactly two squares and you have three, the game is over even if it takes fifteen more moves to prove it. When the counts are equal, stop torturing the position — it is a draw, and the honest move is to look for resources elsewhere.

Training It

Corresponding squares are the rare chess skill that is genuinely learnable as an algorithm. Start by re-deriving the example from this article — pawn on e4, find the key squares, prove to yourself that e7 is the unique answer to e5 — then graduate to blocked positions with two entry points, where you must build a small network of pairs. Our endgame trainer drills the opposition, triangulation, and key-square patterns that make up the correspondence toolkit, against exact play — and exactness is the whole subject here. A student who has computed correspondences by hand five times stops fearing long king endings forever; they have seen that underneath the mystery there is only bookkeeping.

Frequently Asked Questions

Are corresponding squares the same as the opposition?

The opposition is the simplest special case. On an open board the correspondence pairs form the regular mirror pattern that opposition rules describe — kings one square apart, matching files and ranks. In blocked positions the pattern warps, and only the general theory gives correct answers. Every opposition is a correspondence; not every correspondence is an opposition.

How do I actually find corresponding squares during a game?

Work from the target backward: identify the entry squares the attacking king wants, find which defensive squares guard them, then label the approach squares pair by pair. In practice you rarely need more than three or four pairs plus a count of spare tempi. If the position is open rather than blocked, skip the machinery — ordinary opposition and the square rule are faster and sufficient.

What is a position of mutual zugzwang?

A position where whichever side must move worsens its result — White to move only draws, Black to move loses, or similar. Corresponding-squares theory is essentially the science of locating these positions and arriving in them with the opponent to move. Our example (kings on e5 and e7, pawn on e4) is a verified case: the side to move is the side in trouble.

What is the Lasker–Reichhelm position?

A famous 1901 study by Emanuel Lasker and Gustavus Reichhelm: a completely blocked pawn ending in which White to play wins by an exact king maneuver and any deviation only draws. It is the canonical demonstration of corresponding squares, because the winning path cannot be found or explained with opposition rules — the correspondence network spans the whole board and is thoroughly irregular.

Why can't triangulation always break the correspondence?

Triangulation requires a surplus: you need more equivalent squares in your region than your opponent has answers. If the defender's network is complete — one guard post for every probe, with a route between them — the positions of mutual zugzwang all lie with you to move, and losing a tempo is impossible. That is precisely what makes a true fortress: the defense never runs out of correct squares.

Related Terms

• Opposition — the regular, open-board special case of correspondence.
• Key Squares — the entry points that the whole correspondence network is built around.
• Triangulation — the maneuver that exploits a surplus of corresponding squares.
• Zugzwang — the compulsion to move that gives the pairs their meaning.
• Fortress — a defensive structure whose correspondence network has no hole.
• Square Rule — the fast geometric tool for the open positions where correspondence theory is unnecessary.