Chess and Go: A Comparison

by Richard Bozulich

© Copyright 2015 by Richard Bozulich

All rights reserved according to international law. No part of this work may be reproduced by any mechanical, photographic, or electronic process, nor may it be stored in a retrieval system, transmitted, or otherwise copied for public or private use without written permission from Kiseido Publishing Company.

Kiseido Publishing Company
Chigasaki, Japan

Chess and go are games rich in both strategy and tactics. Because of the relatively small size of the chess board (64 squares), chess is considered to be more of a tactical game than a strategic one, in which gaining a material advantage is the all important first step to the eventual mating of the king. The small size of the chess board does not seem to be conducive to strategic ideas.

Go, on the other hand, with its enormous playing field of 361 points, is generally considered to be a game in which both strategy and tactics are equally represented.

For sure, there are a large number of principles that govern go strategy. With respect to tactics, there are at least 45 different kinds of tesujis that can be used to gain a tactical and, ultimately, a strategic advantage. Many of the strategic and tactical principles of go are encapsulated in a hundred or so go proverbs. In addition, there are a large number of other strategic principles which make up go theory and are instinctively understood by all strong players.

What about chess? What are the principles that guide a player to make sound strategic moves?

When I first started to investigate this, I didn't expect to find too many strategic principles. I was sure that there would be many more chess 'tesujis'. To my surprise, I found more than 60 strategic chess principles, but only 13 chess tesujis.

In spite of the paucity (compared to go) of its tesujis, chess is still a game rich in tactical maneuvers that require deep and accurate analysis. However, due to the small size of the chess board, a small slip in reading is often catastrophic, whereas in go, small mistakes in reading are not necessarily fatal, as compensation can be obtained in other parts of the board.

In a game of chess, there is essentially one battle going on. Only one opening can be played and the opening chosen sets the strategic theme of the game. In go, each corner of the board can feature a different opening (joseki) each with its own strategic theme. Skirmishes also arise on the sides, so there are numerous battles going on simultaneously in different parts of the board, but they are all interconnected and coordinating them into one coherent strategy is what makes go a very difficult and profound game.

There are a number of strategic concepts that exist in go but not in chess. They are analogous to the ones used in decision-making situations in business, geo-politics, or in everyday life. These concepts are dealt with extensively in The Basics of Go Strategy. A few are listed below.

Aji. The concept of aji doesn't seem to exist in chess. Aji refers to the potential that exists in a position. For example, a group in a certain part of the board may be dead, but, if friendly stones are eventually placed in nearby areas, these supposedly dead stones can come back to life. A position containing a defect that can't be immediately exploited is said to have 'bad aji'. The opposing player must be careful not to destroy this bad aji by playing a move that will force his opponent to eliminate this defect. Such a move is referred to as 'aji keshi' (erasure of aji). In other words, we want to keep all options open for as long as possible, or at least until we can find a way to turn the situation to our advantage. This is good strategy on the go board and it is often good strategy in business and everyday life.

Diversification. One of the first things a go beginner learns is to establish a presence in many parts of the board. To play all of your stones in just one area will allow your opponent to build an overwhelming strategic advantage by playing in many parts of the board while your own stones will end up being overconcentrated and working inefficiently. It is also bad strategy to stake out one large territory and rely on it to win the game. As the saying goes, 'Don't put all of your eggs in one basket.' Diversification is a strategy that all professional stock-market investors and venture capitalists use. The analog concept in chess is to develop all of your pieces and to avoid playing a piece twice until all of your pieces are developed.

Thickness. There are a number of principles that govern the use of thick positions in go. The most important one is 'Don't approach thickness.' Since thick positions are strong, there is not much a single stone, which is weak, can do. It is best to stay away from a thick position and stake out a position in virgin territory. As the old adage advises, 'Don't bash your head against a stone wall.' If you are going to start a new business, you do not want to go into a business in which many strong companies are already established. You want to start a business in which there is little or no competition.

Probes. If you know what your opponent's strategy is, you can then devise your own strategy to counter it. In go, probes are played to force your opponent to commit to a strategy. Here is an example.

Dia. 1
This position is from a game between Sakata Eio (White) and Rin Kaiho. Black invades White's corner enclosure by attaching at 1. With this move, Black is forcing White to decide whether he is going to take the territory in the corner or the territory on the top left.

Dia. 2
If White answers Black 1 with the connection of 2, he indicates his intention to make territory on the top left. Having determined this, Black immediately plays a reducing move with 3, forcing White to defend with 4. He then invades with 5, attacking the marked stone and threatening to slide into White's territory with 'a'.

Dia. 3
Black still has aji in the corner. Namely, he can start a ko there with the moves to 7. This is an annoying ko for White because Black can resolve it by capturing at 'a', threatening to wipe out White's territory on the top left. If White connects at 'a', Black resolves the ko by connecting between 5 and 3 and living outright.

After Black 5 in Dia. 2 White could eliminate this bad aji by making another move in the corner, but at this stage of the game, there are bigger and more urgent issues to attend to. Situations like this where finding the right time for Black to start this ko or for White to prevent it is what makes go such a difficult game strategically.

Dia. 4
Black must not reverse the order of 1 and 3 in Dia. 2. White 2 reinforces his territory on the top left, so it will now be more difficult for Black to invade there. After Black attaches with 3, White secures the corner with 4 to 8 and Black's stones at 5 and 7 will have a very difficult time settling themselves.

Dia. 5
If Black omits the probe of Black 1 in Dia. 1 and immediately invades with 1 in this diagram, White will jump down to 2, then answer Black 3 by expanding his territory on the top left with 4. Now the attachment of Black 5 is no longer a probe, as White has committed himself to making territory on the top left. White answers Black 5 by going for the corner with 6 and White's position in the top left is too strong and Black will have a hard fight to establish a position there.

Dia. 6
In the game in Dia. 1, White answered Black 1 by going for the corner with 2. Black switched to the top left by clamping with 3 and peeping with 5. After White 6, Black extended to 7 and the position at the top was even: Black and White each secured a corner and each ended up with an unsettled group at the top.

Risk management. One can't win by simply eliminating risk (playing safely). If a player feels he is behind, he will try to increase the risk, while the player who feels he is ahead will try to decrease it. In other words, if you feel your chances of winning are more than 70%, you will want to decrease your level of risk. However, if your chances of winning are close to zero, there is no need to be careful and you can take huge risks. Evaluating your chances of winning or losing requires good judgment of the position and this can be quite difficult in go. In chess, it is simpler: A player who has a material advantage will try to exchange pieces to get to the endgame where his opponent will have less chance of staging an upset. In another situation, a player might want to risk playing for a win instead of settling for a draw.

Risk management is strongly intertwined with the personality of the players. Weak players usually avoid risk, if they are aware of the risk at all. Top players have their own default level of risk that they feel comfortable with. A player who likes to make thick positions may seem risk-adverse, but it is also possible that by having thick positions he knows that he can get away with taking risks more easily. So his real personality may not be risk-averse at all.

A lot can be said about the psychological aspects of this issue, but the important thing is to become aware of risk. If your positions don't have any weaknesses — zero risk — your stones may not be working efficiently. A balance must be found between efficiency and risk.

This concept was discussed by Rob van Zeijst throughout Game Three of The 2014 Ten-Game Match between Gu Li and Lee Sedol. This game provided an excellent example of risk management. Lee lost because he prematurely lowered his level of risk.

Go Principles

No one should naively consider these principles as iron-clad rules. Rather, they are only guides to good play. One hundred of the most important principles are analyzed with numerous examples and problems in the following book.
An Encyclopedia of Go Principles,
Other examples of these principles can be found in the following books: 501 Opening Problems,
Attacking and Defending Moyos,
Handicap Go.

Opening Principles

The Order of Playing the Big Opening Points
1. First occupy an empty corner; second, make an enclosure or an a approach move; third, make an extension.

2. Extend three spaces from a two-stone wall.
3. Extend four spaces from a three-stone wall.
4. Establish a position inside your opponent's sphere of influence with a two-space extension.
5. Extend up to five spaces from a corner enclosure.
6. Try to extend at least five spaces from a large-scale wall.
7. When opposing enclosures face each other, the first to play on the central point between them gets the advantage.

Urgent Moves
8. Play urgent moves before big opening moves.
9. Play urgent moves before building or attacking a moyo.
10. Defending your weak stones is urgent.
11. Attack your opponent's weak stones.

12. Play on the junction of two opposing moyos.
13. Expand your moyo while reducing your opponent's.
14. Expand your a moyo with double-wing extensions.
15. When expanding a large-scale moyo, play on the fourth line.
16. Reduce your opponent's moyo with a shoulder hit.
17. Reduce a moyo by playing into it no farther than its outer rim.
18. Don't let your opponent make a moyo while he is reducing yours.
19. Be willing to transfer a moyo from one part of the board to another.

Answering a 3–3 Point Invasion
20. Against a 3–3 point invasion, block on the side that makes the biggest moyo or territory.
21. Against a 3–3 point invasion, defend the territory you have invested in.

Miscellaneous Opening Principles
22. Confine your opponent's stones.
23. Don't allow your stones to be confined.
24. Separate your opponent's stones into weak groups, then attack.
25. Maintain a balance between the third and fourth lines.
26. Reinforce widely spaced stones.
27. There's no ko in the opening.

Middle-Game Principles

Sabaki and Light Moves
28. Attach to make sabaki.
29. Sacrificing some stones may be necessary to make sabaki.
30. Escape lightly into the center with heavy stones.

31. Don't approach thickness.
32. Don't play in the direction of thickness.
33. Don't use thickness to make territory.
34. Use your thickness to attack.
35. Ponnuki is worth 30 points.
36. The tortoise shell is worth 60 points.

Crawling and Pushing
37. Don't crawl along the second line.
38. Don't crawl along the third line.
39. Make a wall by pushing along the fourth line.
40. Crawl along the fourth line.
41. Don't push on the fifth line.
42. Don't push on the sixth line.
43. The fourth line is the line of victory, the third line is the line of defeat.

44. Attack one group by leaning against another.
45. Attack two groups at the same time.
46. Attack while strengthening your weak stones.
47. Map out territory while attacking.
48. Attack with a knight's move.
49. Attack with a cap.
50. Answer a cap with a knight's move.
51. A cap together with a knight's move are a good attacking combination.
52. Make your opponent's stones heavy, then attack.
53. Rob your opponent's stones of their base, then attack.
54. Secure a base for your stones.
55. Don't strengthen your strong stones.
56. Don't make your opponent's weak stones strong.

Creating a Shortage of Liberties
57. Play a hane at the head of two stones.
58. Play a hane at the head of three stones.
59. A group that is short of liberties is in danger of being captured.
60. Play at the center of three stones.

61. Attach across the knight's move.
62. Don't cut a knight's move attachment.
63. Attach against the stronger stone.
64. Against an attachment, play a hane.

The One-Space Jump
65. A one-space jump is never a bad move.
66. Escape with a one-space jump.

67. Even a fool will connect against a peep.
68. Don't eliminate a cut with a peep.
69. Destroy an eye with a peep.
70. Don't play two peeps against a bamboo joint.

71. Capture the cutting stone.
72. When caught in a crosscut, extend.

The Endgame
73. The monkey jump is worth 8 points.
74. Be the first to play a double-sente sequence.

Miscellaneous Principles
75. The turn in the center is a big move.
76. Increase the sacrifice to two stones.
77. Sacrifice cumbersome stones.
78. In a symmetrical position, play on the central point.
79. If you don't have a good move, play elsewhere.
80. An approach-move ko that needs two moves to turn it into a direct ko is not a ko.
81. My opponent's vital point is also my vital point.

Life and Death

82. In a capturing race, a group with one eye beats a group with no eyes.
83. There's death in a hane.
84. A three-space eye has three libertties.
85. A four-space eye has five liberties.
86. A five-space eye has eight liberties
87. A six-space eye has 12 liberties.
88. In the corner, six live, four die.
89. On the side, eight live, seven die.
90. The comb formation is alive.
91. Three crows in the corner always live.
92. There's a brilliant move on the 2–1 point.

The 45 Basic Go Tesujis

1. atari — A threat to capture a stone or a group of stones on the next move.
2. atekomi — A move touching two opposing stones placed diagonally, thereby forming a right triangle with them.
3. de — Pushing in between two stones one space apart or pushing in against a knight's move, threatening to split that shape.
4. degiri — A push-in move followed by a cut.
5. fukurami — Bulge. A hane that expands a one-space jump.
6. guzumi — move that forms the apex of an empty triangle.
7. hai — Crawl. A move that crawls along the second line.
8. hana-zuke — An attachment at the protruding head of a group. Literally, nose attachment.
9. hane — A diagonal move that bends around an opposing stone.
10. hanedashi — A hane which threatens to split the opponent's position.
11. hanekomi — A hane which wedges in between two enemy stones. Note that the word wedge is used to describe both hanekomi and warikomi.
12. hara-zuke — Belly attachment.
13. hasami-tsuke — A clamp.
14. hekomi — Dip.
15. hiki — A move that draws back to an allied stone.
16. horikomi — A throw-in. A sacrifice tactic used to reduce the liberties or eyes of enemy stones.
17. kado — A point diagonally above or below an enemy stone.
18. kake — A pressing (literally, covering) move.
19. kake-tsugi — A diagonal connection.
20. keima — A knight's move.
21. kiri — A cut.
22. kosumi — A diagonal move.
23. kosumi-tsuke — A diagonal attachment.
24. magari — A turning move.
25. narabi — A solid extension in which neither the stone extended from nor the extension touches other stones.
26. nidan bane — A two-step hane.
27. ni-no-ichi — The 2–1 point. This point is often the vital point for the life or death of a group in the corner.
28. nobi — A solid extension.
29. nozoki — A peep. A move that threatens a cut.
30. nuki — A capture.
31. ogeima — A large knight's move.
32. oki — A placement move. A move played in the middle of the enemy's eye space.
33. osae — A blocking move.
34. oshi — A pushing move.
35. sagari — Descending toward the edge of the board.
36. sashikomi — Thrusting in. A move that touches three enemy stones.
37. shita-tsuke — An underneath attachment.
38. tobi — A one-space jump.
39. tobi-tsuke — A one-space jump that is also an attachment.
40. tsuke — An attachment.
41. tsukekoshi — A move that attaches across a knight's move.
42. tsuki-atari — Bumping into an enemy stone.
43. tsugi — A solid connection.
44. warikomi — A wedge played into a one-space jump. In contrast to a hanekomi, a warikomi is a wedge that has no allied stones adjacent to it, diagonally, horizontally, or vertically.
45. yoko-tsuke — A side attachment.

All of these tesujis are covered in the following Kiseido publications:

A Survey of the Basic Tesujis and 501 Tesuji Problems.

Chess Principles

Pawns and pawn chains
1. A good pawn-chain shape from the player's perspective is the shape of a triangle ▲, with the apex pointing upward.
2. Pawn chains are strong. Avoid splitting them into more than two separate chains.
3. Isolated pawns can be a liability in the middle game. Avoid them until the end game.
4. The isolated queen pawn can be an advantage in the middle game. Since the queen bishop's pawn and the king's pawn are missing, your pieces will have more room to maneuver, but it can be a disadvantage in the endgame.
5. Exchange an isolated queen pawn before the endgame.
6. Avoid creating doubled pawns.
7. Avoid creating hanging pawns.
8. Avoid creating a backward pawn
9. Don't advance a backward pawn.
10. A strong center pawn position makes a good base from which to launch an attack.
12. Unless you are planning a king-side attack, avoid prematurely advancing the pawns in front of your castled king until the endgame.
13. A passed pawn can give you an advantage by requiring your opponent to use a piece to block its advance.
14. In the endgame, passed pawns on the rook's file are weaker than passed pawns on other files, as it is easier for the opponent's king to block it.
15. When you have a pawn majority on one side, the unopposed pawn is the candidate to become the passed pawn. The other pawns are to be regarded as supports. (Nimzovich)
16. Block a passed pawn by placing a piece directly in front of it. (Nimzovich)
17. When trying to queen a passed pawn, put your rook behind it.
18. Avoid putting your rook in front of a passed pawn.
19. If you are trying to stop a passed pawn from queening, put your rook behind the pawn, attacking it from the rear.
20. Do not put your rook in front of a passed pawn unless you have no choice.
21. Passed pawns should advance.
22. Opposed pawns, where there is an enemy pawn on the same file, should often be held back.
23. In the endgame, the advancing pawn must must stay in close contact with its allies.
24. Never play to win a pawn if your development is unfinished.

Knights and Bishops
25. Develop knights before bishops. Knights have a short range so they should be brought out first.
26. Develop your bishops early, as your pawn chains may block its development and hence its movement.
27. Knights are more useful than bishops in closed positions.
28. Bishops are more useful than knights when there are many open diagonals.
29. In an open position, two bishops are a big advantage.
30. Knights are more efficient when placed in the center of the board and least efficient on the rook files. 'A knight on the rim is grim.'
31. A supported knight on the d6 or e6 square which can't be driven away by your opponent's pawns can give you a positional advantage.
32. In the endgame, knights are weaker than bishops in stopping the opponent's pawns from advancing.
33. In the end game, when your opponent threatens to advance pawns on both the left and the right flanks, a knight will be helpless.
34. In the end game, a bishop is better than a knight when your opponent has pawns on both the left and right flanks.
35. Bishops are more useful than knights in the end game.
36. In the endgame, two bishops will generally defeat a bishop and a knight.
37. If the bishop controls the long diagonal towards your opponent's castled king, an attack on the king side is possible.
38. In the endgame, a bishop and some linked pawns can support a pawn advance or it can delay your opponent's pawn advance.
39. In the endgame, a pair of bishops is more advantageous than a pair of knight's.

40. Rooks work most efficiently on open files.
41. Two rooks on same open file provide opportunities to attack.
42. Two rooks on the 7th or 8th rank have great potential to mate or to gain material.
43. In the endgame, two rooks in an open position are often stronger than a queen, especially if most of the minor pieces and pawns have been traded off.
44. In endings where you have King, Rook and Pawn against a King and Rook, place your rook behind the pawn and your king next to the pawn.

45. A queen is most effective in combination with a rook or a minor piece.
46. Avoid developing the queen during the opening, as it will probably be harassed by the opponent's minor pieces and cause a loss of tempo.

47. Usually it is best to castle early, particularly when castling on the king side. Castle for a specific reason. Usually to protect the king from immediate attack while developing the rook. But absent any threat it is often advantageous to continue development and leave your options open (aji) as to which side to castle or forgo castling altogether.
48. When the kings are castled on opposite sides, you can attack by advancing your pawns toward the opponent's king.
49. When the kings are castled on the same side, do not start an attack on the king too early; maneuver on the other side instead.
50. Kings should advance into the center after the major pieces are removed from the board.
51. Keep the king near your pawn chains to support their advance.
52. In king and pawn endings, the king should stay ahead of the pawns.
53. In the endgame, provide a shelter for the king in case it is attacked.

Exchanges should be made with a particular objective in mind.

54. Exchange to gain a tempo or to prevent the opponent from gaining a tempo.
55. Exchange to seize or open a file.
56. Exchange to eliminate a defending piece.
57. Exchange to avoid losing a tempo by retreating.
58. Exchange to eliminate weak pawns.
59. Exchange if you have a material advantage.
60. Exchange an inactive piece for one of your opponent's active pieces.

General Principles
61. Develop your pieces quickly. Every piece should be developed using only one move.
62. In the opening, avoid moving the same piece twice.
63. Every pawn move must be regarded as a loss of tempo unless it helps to build or support the center or attacks the enemy's center.
"In the opening, one or two pawn moves, no more." (Lasker)
64. In general, pawn moves on the flanks result in a loss of tempo.
65. Try to control as many of the center d4, d5, e4, and e5 squares as you can.
66. Control the center with pieces instead of pawns. (Nimzovich)
The hyper-modern strategy for doing this is to develop the bishop on the second rank of the adjacent knight file — the fianchettoed bishop. These are the ideal squares to place the bishops, as each one is placed on their longest diagonals, controlling the maximum number of squares.
67. Delay direct occupation of the center with pawns and undermine the opponent's pawns, using a fianchettoed bishop.
68. Avoid exchanging fianchettoed bishops, as the squares they were protecting will become weak.
69. Overprotect pieces and pawns under attack. (Nimzovich)
70. Don't attack until you have developed all of your pieces.
71. When the center is locked, start an attack on the flanks.
72. If two pieces are attacking your position, drive off one of those pieces by advancing a pawn.
73. Play moves with more than one meaning. For example, a pawn move may attack a piece while opening a diagonal for your bishop.

Chess Tesujis

1. The pin
2. The skewer
3. Discovered check
4. Double check
5. The see-saw
6. Double attacks (forks executed by pawns and other pieces)
7. Removing the guard
8. Protection by interposing a piece.
9. Zwischenzug (an intermediate move)
10. Smothered mate
11. Perpetual check
12. Zugzwang (compulsion to move. Unlike in go, passing is not allowed in chess)
13. Triangulation