Endgame Theory

Endgame theory is the branch of chess knowledge concerned with positions in which few pieces remain.

6 frameworks, from 1749 to today · Chess · Games

still a working view ended · each leads to its entry below

Overview

Endgame theory is the branch of chess knowledge concerned with positions in which few pieces remain1. Its central question is deceptively narrow: given a small set of pieces on the board, can the side with an advantage force a win, can the defender hold a draw, and what is the exact procedure either way? What makes the field distinctive is that, unlike opening or middlegame theory, its answers can often be proved. A king and rook against a lone king is a win; a king and two knights against a lone king is not. Endgame theory is where chess comes closest to mathematics, and its history is the story of how players, composers, and eventually machines turned vague intuition about such positions into exact knowledge.

Philidor and the First Systematic Endgame Thought

Before the eighteenth century, endgame knowledge in Europe circulated as scattered practical maxims. The decisive change came with François-André Danican Philidor, the French composer and chess player whose Analyse du jeu des échecs (1749) was the most influential chess book of its century. Philidor's lasting contribution to the endgame was his insistence that pawns, not pieces, were the soul of chess: the structure of the pawn chain determined whether a position was winning or losing, and the endgame was where that truth became visible. His famous maxim that pawns are "the soul of chess" was not a slogan but a working principle. He analysed rook-and-pawn endings in detail and showed, for example, that a rook on the seventh rank behind enemy pawns could be decisive. Philidor also demonstrated the technique of the "Philidor position" in rook endings, a drawing method in which the defending rook occupies the third rank to block the enemy king2. His work established that the endgame could be studied as a system rather than a collection of tricks.

The Classical Manual Tradition

The nineteenth century produced the great manuals that codified endgame technique for generations. The German master Johann Berger's Theorie und Praxis der Endspiele (1890) was the first comprehensive treatise devoted entirely to the endgame, and it set a pattern: systematic classification of material combinations, with rules for each. The tradition reached its peak in the early twentieth century with the Russian-born, French-based theorist André Chéron, whose multi-volume Traité complet d'échecs (1920s–1930s) and later Lehr- und Handbuch der Endspiele (1952–1970) became the standard reference. Chéron's method was exhaustive: for each material balance he established whether the position was a win or a draw and gave the winning or drawing procedure. This classical manual tradition treated the endgame as a body of settled fact to be catalogued, and its authority shaped how players studied the phase for most of the twentieth century.

The Composed Endgame Study Tradition

Alongside practical manuals, a parallel tradition of composed endgame studies developed. A study is a position constructed by a composer, not reached in a game, that poses a precise task—usually "White to move and win" or "White to move and draw"—with a unique solution3. The study tradition grew out of the nineteenth-century problem movement, but its endgame branch became a distinct art form. The Czech composer and theorist Josef Kling, working in London in the 1840s and 1850s, produced some of the earliest systematic studies, and the genre flourished in the late nineteenth and early twentieth centuries across Europe. The Russian school of study composition, associated with figures such as Alexei Troitzky, whose work on two knights against pawns remains fundamental, and the brothers Vasily and Mikhail Platov, treated studies as both art and a source of practical truth: a well-composed study often revealed a resource that practical players had missed. The study tradition fed directly back into practical theory, because composers were forced to establish exact win/draw verdicts for material combinations that manuals had treated loosely.

The Positional School and Endgame Strategy

The early twentieth century saw the endgame absorbed into the broader positional school of chess thought. Siegbert Tarrasch, the German master whose Die moderne Schachpartie (1912) and other works shaped a generation, taught that endgame play was governed by general principles—the activity of the king, the creation of a passed pawn, the principle of two weaknesses—rather than by memorized procedures alone. Tarrasch's emphasis on the king as a fighting piece in the endgame, and on the importance of centralizing it early, became standard doctrine. This strategic approach complemented the manual tradition: the manuals told you what was winning, the positional school told you how to steer toward winning positions. The two coexisted, and most strong players drew on both.

The Soviet Endgame School

The most systematic endgame pedagogy of the twentieth century emerged in the Soviet Union. The Soviet chess school, built on state support for chess education from the 1920s onward, treated the endgame as a core subject of training rather than an afterthought. The central figure was Nikolai Grigoriev, a Soviet master and composer whose studies and analyses in the 1920s and 1930s established precise rules for king-and-pawn endings—the "Grigoriev position" and his work on the opposition and the breakthrough remain standard. Grigoriev's approach was to reduce each ending to its essential geometry, showing exactly when a king could catch a pawn or when a passed pawn could promote. This exactness was then built into Soviet training methods, which emphasized endgame technique as a measurable skill. The Soviet school's influence spread worldwide through its literature and through the players it trained, and it made endgame study a standard part of serious chess education everywhere.

Tablebases: The Endgame Becomes Computable

The decisive rupture in endgame theory came with the computer. In the 1970s and 1980s, researchers began asking whether the endgame could be solved by exhaustive computation rather than analysis. The key figure was Ken Thompson, the American computer scientist who, working at Bell Labs, produced the first endgame tablebases—databases that store the exact win/draw/loss verdict for every legal position with a given set of pieces. A tablebase works by retrograde analysis: starting from positions where checkmate or stalemate is immediate, the computer works backward through all legal moves to determine, for every position, whether the side to move can force a win or a draw. Thompson's early tablebases covered three and four pieces, and by the 1990s five-piece tablebases were complete. The results were sometimes surprising: certain material combinations long thought drawn were proved won, and some "obvious" wins turned out to be draws. The tablebase did not just confirm theory; it corrected it.

The tablebase tradition continued with the seven-piece databases completed in the 2010s, and with the development of distance-to-mate and distance-to-zeroing metrics that measure not just the verdict but the optimal procedure. The practical effect on endgame theory was profound: for any position with few enough pieces, the question "is this winning?" now has a definitive answer, and human analysis is measured against it. The tablebase also changed how players study: endgame training software can now present exact verdicts and optimal lines for positions that manuals had treated only approximately.

The Present: Exact Knowledge and Its Limits

Today endgame theory operates on two levels. For positions within tablebase range—roughly seven pieces or fewer—the verdict is exact and computable, and the field's literature increasingly cites tablebase results as authoritative. For positions with more pieces, the classical methods still apply: the manual tradition's classifications, the positional school's strategic principles, and the Soviet school's precise king-and-pawn rules remain the working tools of practical players. The open questions are those the tablebase cannot yet reach: how to evaluate and steer complex multi-piece endings where exact computation is infeasible, and how to translate tablebase verdicts into human-understandable plans. The history of endgame theory is thus a history of increasing exactness—from Philidor's principles, through the manuals and the Soviet school, to the tablebase—with the frontier now lying where computation ends and judgment must take over.

References

  1. Chess endgame, Wikipedia
  2. Philidor position, Wikipedia
  3. Endgame study, Wikipedia

The frameworks of Endgame Theory

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1749-1850

    Philidor's Pawn-Structure Endgame System

    How do pawn structure and rooks decide endgames?

    Fly to its tower in Nousfarer

  2. 1840-Present
  3. 1890-1980

    Classical Manual Tradition

    Which material balances are won or drawn?

    Fly to its tower in Nousfarer

  4. 1900-1950

    Positional School Endgame Strategy

    What general principles guide endgame play?

    Fly to its tower in Nousfarer

  5. 1920-Present

    Soviet Endgame School

    What exact rules govern king-and-pawn endings?

    Fly to its tower in Nousfarer

  6. 1977-Present

    Endgame tablebase

    Is every small position exactly solvable by computation?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Philidor's Pawn-Structure Endgame Sys…Endgame studyClassical Manual TraditionPositional School Endgame StrategySoviet Endgame SchoolEndgame tablebase

Beyond the field

Where Endgame Theory's ideas came from in other fields, and what they led to there: the strongest 12 of 15 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Endgame Theory is a planet in Nousfarer

Its city has a tower for each of these 6 frameworks. Land, walk in, and climb one concept at a time.

Land on Endgame Theory

Related fields

Fields its ideas are linked to

More in Chess

See where Endgame Theory sits on the Map of Human Knowledge