Loading field map
Tesuji theory is the branch of Go (the board game weiqi in Chinese, baduk in Korean) that studies the local tactical moves that achieve a specific objective in the most efficient way possible. The Japanese term tesuji (手筋) literally means "hand technique" or "skill of the hand," and in Go it refers to a move that works—often a move that is surprising, precise, and economical, solving a local problem that ordinary, "natural" moves fail to solve. Tesuji theory is not a separate game from Go, nor a set of rules, but a systematic body of knowledge about recurring tactical patterns and the principles that generate them.
The fundamental question of tesuji theory is deceptively simple: Given a local position where two groups of stones are contesting a boundary, an eye, a connection, or a capture, what is the move that accomplishes the goal with the fewest stones and the least damage to surrounding potential?
This question arises because Go is a game of territorial efficiency. Every stone placed has a cost: it spends a turn that could have been used elsewhere on the board. A tesuji is distinguished from a merely "correct" move by its economy. A correct move may achieve the objective, but a tesuji achieves it while also maximizing future options, minimizing weaknesses, or forcing the opponent into a bad shape. The stakes are concrete: a single tesuji can save a group from death, capture a large group, or turn a losing local fight into a winning one. Conversely, missing a tesuji—or playing a "non-tesuji" that appears similar—can lose the game.
The theory is local in scope, but its consequences are global. A local tactical failure can cascade: a dead group becomes a large territorial loss, which changes the value of all subsequent moves. Thus, tesuji theory sits at the intersection of tactics (immediate local combat) and strategy (global planning), and it is often described as the bridge between the two.
Go originated in China more than two millennia ago, and for most of its history, tactical skill was transmitted through proverbs, commented games, and the study of life-and-death problems (tsumego in Japanese). The concept of a "hand technique" as a distinct category of move emerged gradually. Early Chinese Go literature spoke of "vital points" (yaodian) and "clever moves" (miaoshou), but these were not systematically classified.
The modern concept of tesuji was formalized in Japan during the Edo period (1603–1868), when professional Go players, supported by the shogunate, began to analyze the game with unprecedented rigor. The four great Go houses (Honinbo, Yasui, Inoue, Hayashi) produced extensive commentaries and problem collections. It was in this milieu that the term tesuji came to denote a specific, teachable class of moves, distinct from broader notions of "good play."
The decisive codification occurred in the twentieth century, particularly through the work of Japanese professionals who wrote systematic textbooks. The most influential figure was Go Seigen (Wu Qingyuan), whose revolutionary style in the 1930s–1950s emphasized flexible, efficient local play and demonstrated that many previously accepted "standard" moves were, in fact, inferior to tesuji. His games and writings reshaped how professionals thought about local combat. Around the same time, the publisher and author Toshihiro Shimamura and others produced problem collections that organized tesuji into categories—a process that turned an intuitive art into a learnable discipline.
It is important to note that the historical precursors did not use the term tesuji theory themselves. Chinese players spoke of "vital points," and early Japanese players spoke of "technique" (waza). The modern taxonomy is a retrospective organization of practices that were once transmitted more holistically. The theory, as a named subfield, is a twentieth-century construct built on centuries of accumulated practice.
Tesuji theory is not a single unified doctrine but a set of overlapping approaches that address different aspects of the same problem. These approaches are not rival schools in the sense of mutually exclusive paradigms; rather, they are complementary lenses, each with its own strengths and limitations.
The most common and pedagogically dominant approach organizes tesuji into named categories based on the type of tactical goal they achieve. This is the approach found in most textbooks and problem collections. The major categories include:
The strength of this approach is its clarity and teachability. It provides a vocabulary for discussing moves and a set of recognizable patterns that can be memorized and applied. Its limitation is that it can become a list of tricks without underlying principles. A player who memorizes categories but does not understand why a move works will fail when the position does not match the textbook pattern.
A second approach, often used by stronger players and teachers, emphasizes the underlying principles that generate tesuji rather than the patterns themselves. The central principle is reading: the ability to visualize sequences of moves and their consequences before playing. A tesuji is not a magic move but the endpoint of a correct reading that shows all alternative moves fail.
Within this approach, several heuristics are emphasized:
The strength of this approach is its generality. It explains why a tesuji works, not just what it looks like. Its limitation is that it is harder to teach and requires significant experience to apply. Reading is a skill that must be developed through practice, and principles alone do not tell a player which move to consider first.
A third approach, closely related to the principle-based approach but distinct in emphasis, focuses on the concept of shape (katachi). This approach holds that tesuji are often moves that create "good shape"—efficient, flexible, and resilient arrangements of stones—while avoiding "bad shape"—inefficient, heavy, or easily attacked arrangements.
Good shape is characterized by:
Bad shape includes empty triangles (three stones forming an L-shape with a wasted move), heavy groups (groups with too many stones for their function), and dumplings (overconcentrated stones).
The shape-based approach argues that many tesuji are simply the moves that create good shape in a given position. For example, the bamboo joint is a tesuji because it is the most efficient way to connect two groups while maintaining flexibility. This approach is particularly influential in modern professional Go, where shape intuition is considered a core skill.
Its limitation is that shape is a heuristic, not a law. A move that creates "bad shape" can sometimes be a tesuji if it achieves a critical objective. The shape-based approach must be supplemented by reading to determine when the heuristic applies.
A fourth, more recent approach applies combinatorial game theory (CGT) to Go positions. This approach, developed primarily by mathematicians and computer scientists in the late twentieth century, treats local positions as independent games that can be summed to determine the value of a move. In this framework, a tesuji is a move that maximizes the value of a local game, often by creating a ko (a repeated-capture situation) or by forcing the opponent into a disadvantageous exchange.
The combinatorial approach has been most successful in analyzing endgame positions (the final phase of the game) and certain types of life-and-death problems. It has produced rigorous results about the value of specific moves and has influenced computer Go programs. However, its applicability to the middle game—where positions are too complex for exact analysis—remains limited. It is best understood as a complement to the pattern- and principle-based approaches, providing mathematical precision where the position is small enough to analyze exactly.
These four approaches are not rivals but layers of understanding. The pattern-based approach provides the vocabulary and the initial recognition. The principle-based approach explains why the patterns work and extends them to novel positions. The shape-based approach provides a global aesthetic and heuristic framework that guides which patterns to consider. The combinatorial approach provides exact analysis for small positions and a theoretical foundation for why some moves are better than others.
In practice, a strong player uses all four simultaneously. They recognize a pattern, read the sequence to verify it, evaluate the resulting shape, and—if the position is small enough—check the value computationally. The approaches are taught in sequence: beginners learn patterns, intermediates learn principles, advanced players refine shape intuition, and only a small minority engage with combinatorial analysis.
The contemporary landscape of tesuji theory is shaped by two major developments: the rise of computer Go and the globalization of the game.
The advent of strong computer Go programs, particularly AlphaGo (2016) and its successors, has transformed the study of tesuji. These programs, based on deep neural networks and Monte Carlo tree search, have discovered moves that deviate from traditional tesuji patterns but are nonetheless effective. This has led to a re-evaluation of some classical teachings: some moves once considered tesuji have been shown to be suboptimal, while new tesuji have been discovered.
However, the impact is more nuanced than a simple "replacement" of human theory. Computer analysis has confirmed the vast majority of classical tesuji—the patterns are correct, though the reasons for their correctness are sometimes different from what humans believed. The programs have also revealed that some "tesuji" are context-dependent: a move that is a tesuji in one position may be inferior in another, and the computer's ability to evaluate global context has shown that local efficiency is not always the deciding factor.
The current state is one of dialogue. Professional players use computer analysis to check and refine their understanding, while computer scientists study human tesuji to improve program design. The theory is not obsolete; it is being refined.
Go has spread from East Asia to the rest of the world, and tesuji theory has been translated and adapted into many languages. The Japanese terminology remains dominant—tesuji, tsumego, sente, gote, ko are used internationally—but there is a growing body of literature in English, Chinese, and Korean that presents the theory in local terms. This globalization has led to a standardization of the pattern taxonomy, though regional differences in emphasis persist. Chinese Go, for example, has traditionally placed more emphasis on fighting and tesuji, while Japanese Go has emphasized shape and flow. These are tendencies, not hard boundaries, and modern professional play has largely blended them.
Despite these changes, the core of tesuji theory remains stable. The central problem—finding the most efficient local move—is unchanged. The major categories of tesuji are still taught and used. The principles of reading, vital points, and shape remain the foundation of tactical skill. What has changed is the depth of analysis and the tools available.
The durable present landscape is thus one of continuity and refinement. Tesuji theory is a mature subfield with a well-established body of knowledge, a clear pedagogical structure, and a vibrant research frontier where human intuition and machine analysis interact. It is not a static canon but a living discipline that continues to evolve as the game itself evolves.