top of page
Writer's pictureTakumigo

Introducing Three Methods to Analyze Board Positions (Part 1 of 2)

Updated: Sep 21

*This article was updated for length and clarity on certain cultural references. For the original article, please refer to https://www.1200igosharing.blog/en/post/analyzing-methods-in-go


This article was inspired by a recent post-game review in Hong Kong between two players who used differing analysis methods of Go board positions. During the review the stronger player made critical remarks on the weaker player’s moves based on traditional Go principles, while the weaker player defended his moves with AI evaluations. So how should one judge a position in Go?


Unlike some scenarios (e.g. life and death problems) that have straightforward solutions and are easy for players to analyze or learn from, there’re many more scenarios with feasible variations from which players would choose based on their subjective understanding and personal experience. Before the introduction of AI, players would attempt to insert objectivity into their move evaluation by adhering to established Go principles and proverbs (e.g. "a ponnuki is worth thirty points."), or by relying on professional commentaries to evaluate board positions. However, professionals can also make subjective assessments, attributing them to their “style”. AI helped provide players with a more objective assessment, or so we think.


Often it is difficult for amateur players to make sense of AI suggestions and apply them properly in their games. This is especially apparent in the middle and late stages of the game, where the AI’s best moves can change significantly from the slightest variation. Unaware of AI’s peculiarities, some players would memorize a few AI patterns without having learnt the traditional analysis methods and become frustrated when their AI patterns failed them.


Due to the profound nature of Go, I believe it is important to use both AI and Traditional analysis methods as they are great complements to each other for learning.

In the spirit of traditional Go learning methods, and as a reference to players who appreciate the hybrid approach, I hereby introduce three common methods for consideration when analyzing board positions.



Method 1: Less is more

Board Position #1



Black to move:  Which move has a higher value, A or B?


~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~


Move B has slightly higher value. The lower right corner territory can be almost secured with just one move.

The fewer moves that are required to enclose a territory, the more efficient they are.

If Black chooses to enclose the upper right corner, White can still invade at A.

Actually, this is somewhat similar to the loosely translated Chinese proverb  "Golden Corner, Silver Side, Grass Belly". It means that fewer moves are needed to surround a corner than moves for the sides, i.e. surrounding the corner is more efficient than the sides. The center (belly), even though its area  is large and vast, needs more stones than the side and corner to surround territories.


Board Position #2:



White to move: A or B?


~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~


Of course, lower side is more efficient as it only needs one stone to prevent Black’s invasion from the lower right corner, whereas the upper side requires 2 moves to do so as it is still open at either A and B (i.e. Miai).


Board Position #3:



Black to move: A or B?


~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~

~~~~~~ANSWER ~~~~~~


Again, a move on the the upper side is more efficient because black’s lower side is open to invasion at either A (shoulder hit) or B (second-line invasion). Unlike the previous examples, "Open on both sides" doesn't only refer to the potential invasion from the corner(s) (i.e. B) - the sides can also include a potential invasion from the center (i.e. A) as well.


Feel free to share screenshots from your recent games and let us know how this method would apply. We will discuss the remaining two methods in the next article. Stay tuned.


written by Michael Cheung and edited by Dan J (danjjman@gmail.com)


114 views0 comments

Comentários


bottom of page