End Game Move Valuation

Go Resources
1

Hello–

I’m trying to get a better practical understanding of both the philosophy behind & the practical methodology of how to value potential moves in the end game. Does anyone have any resources (You Tube videos, books, etc.) to recommend to better understand swing counting & especially Miai counting, and how to apply them in game situations (examples, etc.)?

I’ve been reading the entries in Sensei’s Library & have found a small handful of videos online that address the topic, but I’m looking for a gentler introduction to the concepts & how to correctly apply them for less experienced players.

Thanks!

3 Likes
2

It’s an interesting topic because I’d played Go on and off for over 10 years and only recently learned in a YouTube video the concept of a “gote” move being worth half of its estimated territory profit due to losing “sente”. The idea is that losing the initiative by playing a gote move is worth it if it earns 2x more points than any sente moves you could play.

Go has many concepts like these which make sense once explained but they’re not trivial to develop on your own, and I don’t know where to find them either. I suspect Chinese books contain them and nobody has bothered to translate them because you have to be good at Chinese, English, and Go to explain properly.

2 Likes
3

Scherm­afbeelding 2026-01-11 om 05.15.22
Scherm­afbeelding 2026-01-11 om 05.15.22762×1034 59.3 KB

6 Likes
4

Isn’t it that the sente is worth double? This is the first time I’ve heard it the other way where the gote is worth half.

3 Likes
5

true, that way makes more sense. Sente is worth double because you get a second move for free, and the second move is usually worth a similar amount to the first move. I probably remembered it wrong.

1 Like
6

English endgame books with the aim of providing a practical understanding of endgame values are:

  • Absolute Counting in the Endgame (O Meien)
  • Basic Endgame Problems Vol. 1 - Gote (Robert Jasiek)
  • Basic Endgame Problems Vol. 2 - Sente (Robert Jasiek)
  • Endgame Problems 1 (Robert Jasiek)
  • Endgame Vol. 2 - Values (Robert Jasiek)
  • Endgame Vol. 3 - Accurate Local Evaluation (Robert Jasiek)
  • Endgame Vol. 4 - Global Move Order (Robert Jasiek)
  • Get Strong at the Endgame (Richard Bozulich)
  • Rational Endgame (Antti Törmänen)
  • The Endgame (Ogawa)

The other relevant English endgame books teach without values or are for experts interested in mathematics. Of the mentioned books, the contents of Get Strong at the Endgame and The Endgame is outdated, contains quite a few mistakes and I would rather recommend to avoid them, unless you really want to read every book out there. These two use, but are a bad explanation of the theory of, traditional endgame theory aka swing counting aka deiri counting. The other books use modern endgame theory aka miai counting aka absolute counting, which is logically consistent so that their contents is almost without mistakes. Endgame Vol. 2 - Values mostly uses modern endgame theory but also briefly explains traditional endgame theory.

Of the books using modern endgame theory, the following provide the most basic explanations or problems: Absolute Counting in the Endgame, Basic Endgame Problems Vol. 1 - Gote, Basic Endgame Problems Vol. 2 - Sente, Endgame Vol. 2 - Values, Rational Endgame. Of these, the two with Problems in their titles offer problems and the value calculations in their answers. The other three books may have a few problems but focus on explaning the basic theory. Rational Endgame does so by focussing on textual explanations. Absolute Counting in the Endgame does so around examples, of which some are global. Endgame Vol. 2 - Values explains - together with examples - the most concepts of basic endgame theory, which means one can learn more while it is harder to learn everything. Endgame values is a demanding subject so reading all books on modern endgame theory is better than reading only one book. One can read these three theory books in this order from easier to more demanding. Eventually, one needs the demanding theory anyway. The two problem books can be read in parallel, before or afterwards.

As a beginner of endgame theory, you would read the less basic Endgame Problems 1 after the two Basic Endgame Problems books. Endgame 3 or 4 are clearly above beginner level.

Endgame books explain very much more than videos or webpages. Sensei’s Library focusses on some mathematical concepts too heavily and is not the best start to learning the basic endgame theory.

My additional advice to less experienced players: do not neglect the even more basic, value-less endgame theory! E.g., with sente before gote you can already advance significantly.

shamisen, it has been difficult to develop almost all of modern endgame theory on my own (with partial input from Bill Spight). I had to do it because it did not exist, in particular not in Asian books. It took me several years of full-time research.

2 Likes
7

Sente worth twice as much as gote expresses it from the view of a gote. Gote worth half as much as sente expresses it from the view of a sente.

In modern endgame theory, the calibration is by numbers of excess plays.

A sente has 1 excess play, which is the sum of one excess play of its gote sequence and of zero excess plays of its sente sequence. (Edit: corrected: it is the gote sequence with the one excess play, of course. Black starts one sequence, White starts the other sequence.) Calibrate by dividing the swing by 1 (or omit this calculation step).

A gote has 2 excess plays, which is the sum of one excess play of its black sequence and of one excess play of its white sequence. Calibrate by dividing the swing by 2.

In traditional endgame theory, the implicit calibration is per two excess plays, as in a gote. To consider a sente, tradition viewed on the swing together with the statement “in sente”, such as “2 points in sente”. Together with the knowledge “worth twice a much as a gote”, we get, for example, “2 points in sente is worth as much as 4 points in gote”.

In modern endgame theory, the example becomes: “The sente has the move value 2. The gote also has the move value 2.” Playing in both local endgames is equally valuable because the move values are equal.

8

Thanks for all of the great suggestions; I really appreciate it! And to add two more possible resources that I found but have not used (just FYI), there appear to be two courses at Go Magic that seem on point:

1 Like