A silly question: What's the highest obtainable score in a game of Go?

Japanese, Chinese and Korean. :slight_smile:

(But Chinese make a distinction depending on the type of repetition, and cannot inflate score like this anyway.)

This actually happened to me not long ago. The opponent (who was behind) tried such repetition (Japanese rules). I explained to him why that wouldn’t work, and that I would stop repeating after a while, but he was not sure so we agreed to play 30-40 cycles.

Interestingly, this was in a low d game (on another server), and I remember how it surprised me that players are aware of the “no win no lose” rule (his words) - even before understanding cases like sending-2-returning-1.

2 Likes

Could anyone explain this concept?
Every one seems to understand it…and still it’s the first time I see this expression

It is a seki with some repetitive potential, which beginners often confuse with true repetitive shapes.

https://senseis.xmp.net/?SendingTwoReturningOne
(beware that not all of this page is correct)

The reason this cannot be used for persistent repetition is that the player forcing the cycle loses one more stone with each iteration. So after a few dozen loops the other side wins even by sacrificing.

I think it would be stalling, which is not allowed under OGS’ Terms of Service, even if the rule set allows it.

The one on post 18 of this topic.

White plays A (A1), black captures 2 stones by playing at A3, white captures the 1 black stone by playing A2. black passes, and we are back to original starting point (white could play at A again and the cycle repeats)

This is prohibited within area scoring but japanese rules does allow it, since it does change the score tally (black captures 2 stones whle white only captures 1, so black will gain a point everytime this cycle repeats)

1 Like

This is a popular misbelief. In reality Japanese scoring was invented much earlier than today’s Chinese scoring. History went like this: stone scoring → territory scoring → area scoring. (I remember I was also surprised when learned about this.)

Well actually before Japanese territory scoring was invented the Tang dynasty had a system that was area scoring (with group tax) done by territory counting. It was effectively a system similar to AGA rules, with its own mechanism to make territory counting come out to the same result as area scoring (with group tax). But when the Japanese borrowed the game of Go they “forgot” that mechanism and hence modern territory scoring was born.

I was wondering if there is a minimum winning score, or the smallest amount of score a player needs to have a guaranteed victory? How would this even be calculated?

In chinese rules, as all is counted the answer is easy. More as half of the board, pondered by komi

1 Like

“Smallest amount of score a player needs to have a a guaranteed victory” does not exist for Japanese rules (since the opponents score can be arbitrarily large thanks to captures). For Chinese rules, the answer is as @Groin said above: for a 19x19 board with 7.5 komi this means having an area of 185 points on the board.

It’s sligthly more interesting to consider what is the minimal score that could theoretically win. Let’s start with Japanese rules.

If we count prisoners as negative points as is usually done at the end of OTB games, arbitrarily large negative scores can win. So let’s not do that.

If we don’t use komi, a winning score must then be larger than 0. It’s easy to make a player have exactly 2 points - simply fill the board with a single group having just two eyes. But how do we make just one point?

This would work if Japanese rules counted territory in seki, but they don’t:
image

If Japanese rules used superko, we could have black play the first move in the middle of the board and then set up a position like this where white is alive with just one eye:
image
(but Japanese rules don’t use superko)
(also this actually requires white to make a capture, increasing her score)

For some reason I considered both these solutions before realizing that we can just give one capture to one of the players!

So the answer for Japanese rules without komi is exactly 1 point (from a capture), and no territory on the board (like in the seki example above). If we give white 0.5 komi, we can skip the capture and bring this down to 0.5 points. And if we allow any komi value, we could pick an arbitrarily small positive real number.

Now, what’s the smallest score that can win in Chinese rules, no komi? The trick here is to maximize the number of neutral points on the board. For instance, the area score on this board is W 151, B 150:


This could certainly be improved further - I stole this seki example from Sensei’s and just rearranged the lower left corner to make the score more even.

If we don’t require the game to be “correctly scored” we can do this:
image
Black has 2 points, white has 1 point, rest of the board is neutral.

In summary, the question is pretty meaningless, and the answers are not very clever :smiley:

5 Likes

Still a good rules practice.

3 Likes

Under Japanese rules, imagine a 19x19 board almost completely filled 360 stones of one color, with just one point left empty.

While it would be possible for the other color to play another stone to remove all of those stones, that player could also choose to pass and go to scoring instead.

At the scoring phase, I think one could actually argue that all of those stones, with only one eye, are alive. Of course, the opponent could clearly remove those stones with hypothetical continuation of play, but that would result in a mostly empty board, where both players would have the ability to play more stones that should live. Hence, a 360-stone group with one eye would actually be considered alive, if both players pass.

This is just due to a technicality of the definition of alive under the Japanese rules:

Stones are said to be “alive” if they cannot be captured by the opponent, or if capturing them would enable a new stone to be played that the opponent could not capture.

4 Likes