Go and mathematics

The game of Go is one of the most popular games in the world. As a result of its elegant and simple rules, the game has long been an inspiration for mathematical research. Shen Kuo, an 11th century Chinese scholar, estimated in his Dream Pool Essays that the number of possible board positions is around 10¹⁷². In more recent years, research of the game by John H. Conway led to the development of the surreal numbers and contributed to development of combinatorial game theory (with Go Infinitesimals^[1] being a specific example of its use in Go).

Computational complexity

Generalized Go is played on n × n boards, and the computational complexity of determining the winner in a given position of generalized Go depends crucially on the ko rules.

Go is “almost” in PSPACE, since in normal play, moves are not reversible, and it is only through capture that there is the possibility of the repeating patterns necessary for a harder complexity.

Without ko

Without ko, Go is PSPACE-hard.^[2] This is proved by reducing True Quantified Boolean Formula, which is known to be PSPACE-complete, to generalized geography, to planar generalized geography, to planar generalized geography with maximum degree 3, finally to Go positions.

Go with superko is not known to be in PSPACE. Though actual games seem never to last longer than $n^{2}$ moves, in general it is not known if there were a polynomial bound on the length of Go games. If there were, Go would be PSPACE-complete. As it currently stands, it might be PSPACE-complete, EXPTIME-complete, or even EXPSPACE-complete.

Japanese ko rule

Japanese ko rules state that only the basic ko, that is, a move that reverts the board to the situation one move previously, is forbidden. Longer repetitive situations are allowed, thus potentially allowing a game to loop forever, such as the triple ko, where there are three kos at the same time, allowing a cycle of 12 moves.

With Japanese ko rules, Go is EXPTIME-complete.^[3]

Superko rule

The superko rule (also called the positional superko rule) states that a repetition of any board position that has previously occurred is forbidden. This is the ko rule used in most Chinese and US rulesets.

It is an open problem what the complexity class of Go is under superko rule. Though Go with Japanese ko rule is EXPTIME-complete, both the lower and the upper bounds of Robson’s EXPTIME-completeness proof^[3] break when the superko rule is added.

It is known that it is at least PSPACE-hard, since the proof in^[2] of the PSPACE-hardness of Go does not rely on the ko rule, or lack of the ko rule. It is also known that Go is in EXPSPACE.^[4]

Robson^[4] showed that if the superko rule, that is, “no previous position may ever be recreated”, is added to certain two-player games that are EXPTIME-complete, then the new games would be EXPSPACE-complete. Intuitively, this is because an exponential amount of space is required even to determine the legal moves from a position, because the game history leading up to a position could be exponentially long.

As a result, superko variants (moves that repeat a previous board position are not allowed) of generalized chess and checkers are EXPSPACE-complete, since generalized chess^[5] and checkers^[6] are EXPTIME-complete. However, this result does not apply to Go.^[4]

Complexity of certain Go configurations

A Go endgame begins when the board is divided into areas that are isolated from all other local areas by living stones, such that each local area has a polynomial size canonical game tree. In the language of combinatorial game theory, it happens when a Go game decomposes into a sum of subgames with polynomial size canonical game trees.

With that definition, Go endgames are PSPACE-hard.^[7]

This is proven by converting the Quantified Boolean Formula problem, which is PSPACE-complete, into a sum of small (with polynomial size canonical game trees) Go subgames. Note that the paper does not prove that Go endgames are in PSPACE, so they might not be PSPACE-complete.

Determining which side wins a ladder capturing race is PSPACE-complete, whether Japanese ko rule or superko rule is in place.^[8] This is proven by simulating QBF, known to be PSPACE-complete, with ladders that bounce around the board like light beams.

Legal positions

Since each location on the board can be either empty, black, or white, there are a total of 3^n² possible board positions on a square board with length n; however not all of them are legal. Tromp and Farnebäck derived a recursive formula for legal positions $L(m,n)$ of a rectangle board with length m and n.^[9] The exact number of $L(19,19)$ was obtained in 2016.^[10] They also find an asymptotic formula $L\approx AB^{m+n}C^{mn}$ , where $A\approx 0.8506399258457145$ , $B\approx 0.96553505933837387$ and $C\approx 2.975734192043357249381$ . It has been estimated that the observable universe contains around 10⁸⁰ atoms, far fewer than the number of possible legal positions of regular board size (m=n=19). As the board gets larger, the percentage of the positions that are legal decreases.

Board size n×n	3^n²	Percent legal	$L$ (legal positions) (A094777)^[11]
1 × 1	3	33.33%	1
2 × 2	81	70.37%	57
3 × 3	19,683	64.40%	12,675
4 × 4	43,046,721	56.49%	24,318,165
5 × 5	847,288,609,443	48.90%	414,295,148,741
9 × 9	4.43426488243 × 10³⁸	23.44%	1.03919148791 × 10³⁸
13 × 13	4.30023359390 × 10⁸⁰	8.66%	3.72497923077 × 10⁷⁹
19 × 19	1.74089650659 × 10¹⁷²	1.20%	2.08168199382 × 10¹⁷⁰

Game tree complexity

The computer scientist Victor Allis notes that typical games between experts last about 150 moves, with an average of about 250 choices per move, suggesting a game-tree complexity of 10³⁶⁰.^[12] For the number of theoretically possible games, including games impossible to play in practice, Tromp and Farnebäck give lower and upper bounds of 10^10⁴⁸ and 10^10¹⁷¹ respectively.^[9] The lower bound was improved to a googolplex by Walraet and Tromp.^[13] The most commonly quoted number for the number of possible games, 10⁷⁰⁰^[14] is derived from a simple permutation of 361 moves or 361! = 10⁷⁶⁸. Another common derivation is to assume N intersections and L longest game for N^L total games. For example, 400 moves, as seen in some professional games, would be one out of 361⁴⁰⁰ or 1 × 10¹⁰²³ possible games.

The total number of possible games is a function both of the size of the board and the number of moves played. While most games last less than 400 or even 200 moves, many more are possible.

Game size	Board size N (intersections)	N!	Average game length L	N^L	Maximum game length (# of moves)	Lower limit of games	Upper limit of games
2 × 2	4	24	3	64		386,356,909,593^[15]	386,356,909,593
3 × 3	9	3.6×10⁵	5	5.9×10⁴
4 × 4	16	2.1×10¹³	9	6.9×10¹⁰
5 × 5	25	1.6×10²⁵	15	9.3×10²⁰
9 × 9	81	5.8×10¹²⁰	45	7.6×10⁸⁵
13 × 13	169	4.3×10³⁰⁴	90	3.2×10²⁰⁰
19 × 19	361	1.0×10⁷⁶⁸	200	3×10⁵¹¹	10⁴⁸	10^10⁴⁸	10^10¹⁷¹
21 × 21	441	2.5×10⁹⁷⁶	250	1.3×10⁶⁶¹

The total number of possible games can be estimated from the board size in a number of ways, some more rigorous than others. The simplest, a permutation of the board size, (N)_L, fails to include illegal captures and positions. Taking N as the board size (19 × 19 = 361) and L as the longest game, N^L forms an upper limit. A more accurate limit is presented in the Tromp/Farnebäck paper.

Longest game L (19 × 19)	(N)_L	Lower limit of games	Upper limit of games	Notes
1	361	361	361	White resigns after first move, 361 ignoring all symmetry including y = x else (distances from corner) 10×10−10=90 90/2=45 +10 (adding back x = y points of symmetry) = 55.
2	722		721	Black resigns after white's first move, 721 ignoring all symmetry including y=x else 19×19−19=342 342/2=171 +19 = 190 − 1 = 189.
50	2.1×10¹²⁶		7.5×10¹²⁷
100	1.4×10²⁴⁹		5.6×10²⁵⁵
150	6.4×10³⁶⁷		4.2×10³⁸³
200	1.9×10⁴⁸¹		3.2×10⁵¹¹
250	8.2×10⁵⁸⁷		2.4×10⁶³⁹
300	2.8×10⁶⁸⁴		7.8×10⁷⁶⁶
350	3.6×10⁷⁶⁰		1.3×10⁸⁹⁵
361	1.4×10⁷⁶⁸		1.8×10⁹²³	Longest game using 181 black and 180 white stones
411	n/a		1.3×10¹⁰⁵¹	Longest professional game^[16]
500	n/a		5.7×10¹²⁷⁸
1000	n/a		3.2×10²⁵⁵⁷
47 million	n/a		10^10⁸	361³ moves
10⁴⁸	n/a	10^10⁴⁸	10^10¹⁷¹	Longest game

10⁷⁰⁰ is thus an overestimate of the number of possible games that can be played in 200 moves and an underestimate of the number of games that can be played in 361 moves. Since there are about 31 million seconds in a year, it would take about 2+1⁄4 years, playing 16 hours a day at one move per second, to play 47 million moves.

Notes

^ "Go Infinitesimals at Sensei's Library". senseis.xmp.net. Retrieved 2022-02-10.
^ ^a ^b Lichtenstein, David; Sipser, Michael (April 1980). "Go Is Polynomial-Space Hard" (PDF). Journal of the ACM. 27 (2): 393–401. doi:10.1145/322186.322201. S2CID 29498352.
^ ^a ^b Robson, John (1983). "The complexity of Go". Proceedings of the IFIP 9th World Computer Congress on Information Processing: 413–417.
^ ^a ^b ^c Robson, J (1984). "Combinatorial games with exponential space complete decision problems". Mathematical Foundations of Computer Science 1984. Lecture Notes in Computer Science. Vol. 176. pp. 498–506. doi:10.1007/BFb0030333. ISBN 978-3-540-13372-8. {{cite book}}: |journal= ignored (help)
^ Aviezri Fraenkel and D. Lichtenstein (1981). "Computing a perfect strategy for n×n chess requires time exponential in n". J. Comb. Theory A. 31 (2): 199–214. doi:10.1016/0097-3165(81)90016-9.
^ J. M. Robson (1984). "N by N checkers is Exptime complete". SIAM Journal on Computing. 13 (2): 252–267. doi:10.1137/0213018.
^ Wolfe, David (2002). Nowakowski, Richard J. (ed.). "Go endgames are PSPACE-hard" (PDF). More Games of No Chance, Mathematical Sciences Research Institute Publications 42: 125–136. Archived from the original (PDF) on 2017-08-10. Retrieved 2016-07-09.
^ Crâşmaru, Marcel; Tromp, John (2000). "Ladders Are PSPACE-Complete". Computers and Games. Lecture Notes in Computer Science. Vol. 2063. Springer. pp. 241–249. CiteSeerX 10.1.1.24.4665. doi:10.1007/3-540-45579-5_16. ISBN 978-3-540-43080-3.
^ ^a ^b Tromp, J; Farnebäck, G (2007), "Combinatorics of Go", Computers and Games, Lecture Notes in Computer Science, vol. 4630, Springer, Berlin, Heidelberg, pp. 84–99, doi:10.1007/978-3-540-75538-8_8, ISBN 978-3-540-75537-1
^ https://tromp.github.io/go/legal.html 208 168 199 381 979 984 699 478 633 344 862 770 286 522 453 884 530 548 425 639 456 820 927 419 612 738 015 378 525 648 451 698 519 643 907 259 916 015 628 128 546 089 888 314 427 129 715 319 317 557 736 620 397 247 064 840 935
^ "Combinatorics of Go" (PDF). github.io. Retrieved 17 June 2023.
^ Allis 1994
^ Walraet, M; Tromp, J (2016), "A Googolplex of Go Games", Computers and Games, Lecture Notes in Computer Science, vol. 10068, Springer, Berlin, Heidelberg, pp. 191–201, doi:10.1007/978-3-319-50935-8_18, ISBN 978-3-319-50934-1
^ "Home - American Go Association". www.usgo.org. Retrieved 17 June 2023.
^ Tromp 1999
^ "Statistics on the length of a go game".

References

AGA. "Top Ten Reasons to Play Go".
Allis, Victor (1994). Searching for Solutions in Games and Artificial Intelligence (PDF). Ph.D. Thesis, University of Limburg, Maastricht, The Netherlands. ISBN 978-90-900748-8-7.
Hearn, Robert A. (2006). "Games, Puzzles, and Computation" (PDF). [Ph.D. Thesis, MIT.]
Johnson, George (1997-07-29). "To Test a Powerful Computer, Play an Ancient Game". New York Times.
Papadimitriou, Christos (1994), Computational Complexity, Addison Wesley.
Tromp, John (1999). "Number of 2×2 games with positional superko".
Tromp, John (2016). "Number of legal Go positions (up to 19×19)".
Tromp, John; Farnebäck, Gunnar (2007). "Combinatorics of Go".

External links

Go and Mathematics
Number of possible outcomes of a game - article at Sensei's Library

[1] "Go Infinitesimals at Sensei's Library". senseis.xmp.net. Retrieved 2022-02-10.

[sipser1980-2] Lichtenstein, David; Sipser, Michael (April 1980). "Go Is Polynomial-Space Hard" (PDF). Journal of the ACM. 27 (2): 393–401. doi:10.1145/322186.322201. S2CID 29498352.

[robson1983-3] Robson, John (1983). "The complexity of Go". Proceedings of the IFIP 9th World Computer Congress on Information Processing: 413–417.

[robson1984-4] Robson, J (1984). "Combinatorial games with exponential space complete decision problems". Mathematical Foundations of Computer Science 1984. Lecture Notes in Computer Science. Vol. 176. pp. 498–506. doi:10.1007/BFb0030333. ISBN 978-3-540-13372-8. {{cite book}}: |journal= ignored (help)

[Fraenkel1981-5] Aviezri Fraenkel and D. Lichtenstein (1981). "Computing a perfect strategy for n×n chess requires time exponential in n". J. Comb. Theory A. 31 (2): 199–214. doi:10.1016/0097-3165(81)90016-9.

[robson1984checkers-6] J. M. Robson (1984). "N by N checkers is Exptime complete". SIAM Journal on Computing. 13 (2): 252–267. doi:10.1137/0213018.

[7] Wolfe, David (2002). Nowakowski, Richard J. (ed.). "Go endgames are PSPACE-hard" (PDF). More Games of No Chance, Mathematical Sciences Research Institute Publications 42: 125–136. Archived from the original (PDF) on 2017-08-10. Retrieved 2016-07-09.

[8] Crâşmaru, Marcel; Tromp, John (2000). "Ladders Are PSPACE-Complete". Computers and Games. Lecture Notes in Computer Science. Vol. 2063. Springer. pp. 241–249. CiteSeerX 10.1.1.24.4665. doi:10.1007/3-540-45579-5_16. ISBN 978-3-540-43080-3.

[Tromp_and_Farnebäck_2007-9] Tromp, J; Farnebäck, G (2007), "Combinatorics of Go", Computers and Games, Lecture Notes in Computer Science, vol. 4630, Springer, Berlin, Heidelberg, pp. 84–99, doi:10.1007/978-3-540-75538-8_8, ISBN 978-3-540-75537-1

[10] ttps://tromp.github.io/go/legal.html 208 168 199 381 979 984 699 478 633 344 862 770 286 522 453 884 530 548 425 639 456 820 927 419 612 738 015 378 525 648 451 698 519 643 907 259 916 015 628 128 546 089 888 314 427 129 715 319 317 557 736 620 397 247 064 840 935

[11] "Combinatorics of Go" (PDF). github.io. Retrieved 17 June 2023.

[12] Allis 1994

[13] Walraet, M; Tromp, J (2016), "A Googolplex of Go Games", Computers and Games, Lecture Notes in Computer Science, vol. 10068, Springer, Berlin, Heidelberg, pp. 191–201, doi:10.1007/978-3-319-50935-8_18, ISBN 978-3-319-50934-1

[top_ten-14] "Home - American Go Association". www.usgo.org. Retrieved 17 June 2023.

[15] Tromp 1999

[16] "Statistics on the length of a go game".

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

v t e Go
Overview	Handicaps Komi Rules
Equipment	Bowls Goban Katsura Kaya Stones Clamshell Slate Yunzi
Terms	Aji Atari Board positions Dame Divine move Double hane Eyes Gote, sente and tenuki Hane Hayago Jigo Joseki Kakari Keima Kiai Kikashi Ko Komi Korigatachi Kosumi Ladder Liberty Miai Monkey jump Moyo Myoushu Nakade Nerai Myoushu Peep Pincer Probe Sabaki Seki Sente Shape Shoulder hit Tesuji Thickness Yose
Strategy and tactics	Capturing race Fuseki Chinese Kobayashi Shinfuseki Shusaku Jōseki Nadare Taisha Ko fight Ladder Life and death Mirror Go Opening theory Proverbs Shape Empty triangle Ponnuki Tenuki Tsumego
History	Classic of Arts Dunhuang Go Manual Emperor Yao Four Go houses Four arts Hoensha 9 Pin Zhi Oskar Korschelt Oshirogo Players European players Female players Nihon Ki-in Hall of Fame Professional handicaps
Competition	Go professional Ranks and ratings Dan Kyū Honorary titles Jubango Title holders Tournaments
Games and matches	AlphaGo vs. Fan Hui AlphaGo vs. Ke Jie AlphaGo vs. Lee Sedol Atomic bomb game Blood-vomiting game Ear-reddening game The Game of the Century Kamakura jubango Lee's broken ladder game
Art and media	AlphaGo The Divine Move The Girl Who Played Go The Go Master The Go Player Go World Hikaru no Go Igo Hatsuyōron Long Ode to Watching Weiqi The MANIAC The Master of Go Ranka Sensei's Library Shibumi The Surrounding Game The Weiqi Devil
Computers	Computer Go UEC Cup Engines AlphaGo AlphaGo Master AlphaGo Zero AlphaZero Crazy Stone Darkforest Fine Art GNU Go KataGo Leela Leela Zero Zen Future of Go Summit Monte Carlo tree search Smart Game Format Servers KGS Go Server Pandanet Tygem
Organizations	American Go Association Australian Go Association British Go Association China China Qiyuan Chinese Weiqi Association Hong Kong Go Association European Go Federation French Federation of Go International Go Federation Irish Go Association Japan All Japan Student Go Federation Kansai Ki-in Nihon Ki-in Korea Korea Baduk Association Myongji University Mind Sports Organisation New Zealand Go Society Singapore Weiqi Association Taiwan Chi Yuan Culture Foundation
Other	Benson's algorithm Game record (kifu) Games played with Go equipment Go and mathematics Variants Batoo Capture go Sygo
Go portal Category