A full game of Go with beginner-friendly commentary #2

[Goodspeed]

AKA "The final attempt #2"

Undoubtedly most of you are active Go players by now and would thoroughly enjoy a seemingly never ending thread of screenshots of Go boards with long-winded commentary by some self-important scrub. Those of you who somehow still don't play the game will surely be convinced this time. Anyway I enjoyed doing it last time so why not.
Last thread: viewtopic.php?f=315&t=18319

Go can seem like just another Chess-like game where you're mostly analyzing as many moves ahead as you can, even though it isn't. I figure one way I could shed light on this is by commentating on one of my own ongoing games in a targeted way. That is, not by going into the many variations that were considered each time, but focusing on the broader strategic concepts, the big picture thinking if you will, that goes into each move. The fact that big picture thinking is such an important part of decision making in this game may, for some of you, help lift it from "just another Chess-like game" to the truly unique and, contrary to chess, strategically fascinating gem that it is. We AoE3 players like strategy, don't we?

Since I play a lot of correspondence Go (meaning you get multiple days per move) it is fairly straightforward for me to do this. I can do it as the game is going on, because these games can take months.

In the screenshots, the move order will be shown using numbers, where 1 is the first move played.

Of course, anyone is welcome to ask any questions at any point ITT.

I started a game against my boy funloop, who I'm currently 4-5 against looking to even it up. I'm black, so I get the first move but my opponent gets 6.5 bonus points to compensate.

In the very weird event you missed the memo about Go: viewtopic.php?f=315&t=15086


Glossary

Territory: Map control. Points. The goal of the game. Each empty intersection "surrounded" by your stones counts as one point of territory. Example: This corner would be worth 9 points for black.
Alive: A group is unconditionally alive when it's impossible to surround and kill. If you're interested in how that works, (the beginning of) this video might clarify.
Strong stone/group: A strong stone or group is a stone/group that is not at risk of dying. Strong groups help in fights because they are good to attack from, or to run towards for safety.
Weak stone/group: A weak stone or group is a lonely stone/group with many enemies near it that is at risk of dying, like the white stone in the "invade" example.
Reduce: To move somewhere near your opponent's (potential) territory, often as close to the edge of the territory as possible, in order to reduce his potential to get more.
Invade: To move into your opponent's potential territory in order to reduce it or create your own territory within it. Example: White's move 13 in this position "invades" black's territory. White's stone is weak and can be attacked, but if it survives black will have lost a lot of points.
Secure: To move in or near your own territory to prevent strong invasion or reduction moves.
Influence: The intangible "potential" of any one stone or group of stones, be it in territory or fighting strength. The stronger a group is, the more influence it exerts.
Territorial: A territorial move or play style is focused on gaining territory at the expense of influence. Example: In this variation, black played territorially by taking the corner but allowing white to gain influence.
Settle: To place stones in a way that the area becomes "done" (settled), meaning neither player will find much use playing another move there until much later in the game, when there are no more big moves to play elsewhere.
Sente: Initiative. The player who is forcing the other player to respond to his moves "has sente". They effectively decide where the next area of contention is. Another way of putting it: The player who has the next move where he is not forced to respond to an opponent's move, has sente.
Forcing move: A move that forces a specific response from the opponent in that if it is ignored, they would lose many points or lose the game outright. "Ignoring" here means playing another move elsewhere on the board.

[Goodspeed]

The first 6 moves are standard stuff. Both players take their corners, all of them on the 4-4 points (D16 for example is a 4-4 point because it is the 4th line from the left edge and the 4th line from the top). (5) is a standard approach move, (6) is the most common way to defend the corner efficiently.

(7) is weird. It's typically not recommended to ignore the situation in the bottom left corner, because it's somewhat painful for black if white gets another move there. White playing at A for example would make it so that saving (5) can become costly. However, I learned from the AI that this is actually okay in the opening because there are some ways to still make use of the (5) stone even if we end up sacrificing it, and there are a lot of other big moves still on the board like (7). A benefit of playing this way is that I've been experimenting with it for a while, and our opponent may not be as familiar with the pattern as we are. For example, it's easy for them to overinvest into killing (5).

[Goodspeed]

White followed up in the bottom left. We could have responded at A but we want to keep our investment in that area low. If we invest another move, we are kind of committed to the group and sacrificing it becomes painful.

If our opponent chooses to follow up again he's likely to play at A next, which gives us 2 options. We could still run for safety towards the top, and our opponent will have invested a whopping 4 moves into that corner while we only invested 2 and came out of it alive. We may also choose to ignore him again, putting yet another tough question to him: Should he invest another move to kill the stone for good? If he doesn't, we can still make use of it later.

Meanwhile we approached him at (2) and have some strong follow ups there if he ignores us. All said and done it should end up being a pretty even result.

[callentournies]

I admire your vain pursuit

[kaister]

Maybe explain the game of Go in terms that a Fatts Russell Fan like myself or @gibson could understand?

[gibson]

Maybe explain the game of Go in terms that a Fatts Russell Fan like myself or @gibson could understand?

I think that @bwinner could probably help my simple brain understand

[Goodspeed]

Maybe explain the game of Go in terms that a Fatts Russell Fan like myself or @gibson could understand?

When basketball players want to pass the ball it's best if there is no opponent between them and the teammate they want to pass to. It's good to stay "connected" in that way. Similarly in Go, connected stones are stronger

[Goodspeed]

Our opponent did follow up in the bottom left, with the move we were expecting. We chose to ignore it again and followed up in the bottom right. We play (2) to reduce both white's corner and his potential on the bottom side. (3) and (4) are standard continuations of this pattern, but it is surprising that white chose (3) instead of A. The move he chose tends to lead to black getting a little territory on the bottom side, which is also where white was building territory. This way, our opponent is in a way reducing his own potential. But maybe this is fine, or he has other ideas about follow ups that I don't know about.

[princeofcarthage]

Chess

[Goodspeed]

All pretty standard moves, except (5) which I think should have been at A. As played, we have B for later, which is a pretty big move because it nearly forces a response from our opponent, and reduces his corner. I'm not good enough to know when this move becomes bigger than other moves though, so I'm not really sure when to play it.

We respond by strenghtening our right side group with (6) and then (8), and our opponent ends with sente (he gets to play the next move elsewhere).

[Goodspeed]

He chooses to invade our top right corner. Not surprising, AI would very probably approve. We judge that we need to build our wall towards the left, so that we can start building our top side territory. Building it towards the bottom would've also been good, but we would've needed another move to solidify the territory. If we don't play another move there, our opponent can attack our wall AND our right side group while reducing our potential, which didn't seem fun for us.

We didn't want to play another move there because we figured (6) was important. If we allow our opponent to play A, he can attack our group while building his bottom territory, which seems like it might be painful for us.

Now, our opponent gets sente again, and is likely to play something on the top side to reduce our potential there.

[Goodspeed]

He chose to approach our corner instead of playing in the open top side. This gave us the opportunity to play (6) to solidify our top side after defending our corner. Our opponent gets sente again. Now that I'm writing "our opponent gets sente" so often, I'm wondering if maybe giving him sente with L3 was a mistake. It was a pretty passive move.

There are 4 moves we really want to play now:
A is probably big enough to start considering
B is a good way to reduce our opponent's left side. The bonus there is that if we build a strong group there, it could open up the bottom left corner for invasion.
C is a big follow up to the top right pattern which confines white to the corner
D is another way to reduce our opponent's left side potential, but the approach there is to probably sacrifice our lonely stone in the bottom left and reduce his territory from the outside, which gives us opportunities to build something in the (top) middle.

I'm leaning towards playing C if we had sente. But we will probably have to respond to whatever our opponent does.

[Goodspeed]

Our opponent took the move we wanted to play. There go our hopes of a large territory connecting our top and right sides. So be it. Our next project is reducing white's potential on the left. We chose D from the previous post at (2), because it seems like the more flexible move. B from the previous post (C7) would commit us to a specific variation that we don't really like the result of. White would probably get sente and a big wall towards the middle which he should be able to take advantage of by "leaning" on our bottom group and/or top side. We would have to defend, and meanwhile he would be building something big.

They responded with (3), which immediately seems wrong to me because after (4), he can't keep his groups connected. Either (3) or the marked stone will die if he tries to connect with A. I think white should have played at (4).
We expect white to still play A and defend the marked stone, otherwise his entire top left group would be in danger. We then kill (3) and white will probably kill off our lonely C6 stone once and for all. Then we would get sente.

It feels like we're probably ahead now.

[Goodspeed]

Our opponent did indeed play at (1) and after our (2) he chose to defend the bottom side. We sacrifice our (2) stone in order to kill the marked stone. It looks like our opponent is giving up his top left group to further build his bottom left territory, which is probably the right call actually.




It continues with white capturing our stone, and us capturing white's stone. White's marked group in the top left is now dead, left behind enemy lines, but so is our stone in the bottom left. The trade seems good for us but not game-ending by any means. If we are ahead, it's not by much.

[Goodspeed]

White uses sente to play a big move in (1). It seems like a reasonable idea, considering it reduces our top territory and expands white's own potential. However, this stone is not solidly connected to white's D10 group. We can cut them off from each other. Later on, if we manage to get some kind of presence around say G8, this may become relevant.

Our next move was a debate between (2), which we have been wanting to play for a while, and A. We liked A because this expands our territory and greatly reduces the effectiveness of (1). When we didn't play A, white got to play (3) and looks to be building something pretty big. It's still fairly open to reduction though. Our eyes are drawn to G8 again, which reduces from the outside and threatens to efficiently disconnect white (1) from its friends.

But first things first. With (2) we are attacking white's group in the bottom right. After he ignored us, we should be able to strongly attack this group and force white to defend awkwardly. Maybe we can reduce white's potential in the bottom left whilst attacking his bottom right group as it runs into the center.

[Goodspeed]

We played (1) because it looked too good for white if he got to play it. We hoped white would feel inclined to defend his territory and respond, but he ignored it and defended the bottom right. In hindsight (1) may have been a mistake, we should probably have played at white (2). But it was a big move regardless, and it does serve to block white's bottom right group from connecting to its friends on the bottom left, so it can't have been that bad.

We played some forcing moves in the bottom right to strengthen our own groups (both of which are now quite safe) and to prevent white from making a living shape. He can now make 1 eye in the bottom right if he spends another move there, but he would hate to be forced to play such a passive move. Besides, you need 2 eyes to live. So he has no choice but to run to the center with (8).

Now we are at a loss about where to play tbh. There don't seem to be many ways to attack white's group and gain profit from it. Surrounding it entirely is pretty much out of the question, it will only weaken us if we try, so we have to pick a direction to drive the white group into. Nothing seems particularly good.

This tends to be a good time to let it go for now, play elsewhere and then decide how we want to attack later, depending on how the board develops. But where do we play? Something around G8 could be an option. Maybe F13 to expand our top territory, but that seems a bit passive. K13? Same problem. We'd like to follow up on (1) as well, but there don't seem to be strong follow ups to it.

[Goodspeed]

We decided to play another mild attacking move after all, at (1), in order to force white to keep running, preferably in the direction of his left territory. Some forcing exchanges follow. We are happy to have been able to play (7) in sente (forcing a response) because it means our K5 stone is now solidly connected to our living group, and we're also happy to have built a little wall towards the middle with (1), (3) and (5). This could come in quite handy later.

Our attack is over. There is no more move that we can play that will force a response from white. The closest to an attacking move we still have is A, but white doesn't need to respond to it. Therefore we think it would be a little too passive.

We are looking instead at B, C, or D. All of them aim at white's lonely marked stone, which is not connected to his living group on the left. B would threaten to cut immediately, likely forcing a response, and we can then use it to reduce white's territory. We should be able to escape either to our (7) group or to our (1) group.

But cutting the white stone off from its friends is perhaps not as painful as it could be, considering he still has a lot of space towards the middle and top to make life. Maybe we should first reduce that space, while expanding our territory, with C or D.

C is tempting because it blocks white from entering our top side potential territory, but it leaves a weakness at D. If white gets to play there, it could be awkward for us because our left side group is not alive yet. We aren't good enough to properly analyze the following variation, so we'll choose the safer D. White will probably play C but I think we're okay with that. If this becomes the border of our top territory it will be quite a big one indeed, and we still have enough options to reduce any potential white would gain from it.

[Goodspeed]

White instead ignored us and played (2), invading our top side. He threatens to connect to his top right group with Q18, which we can't allow. By blocking it though, we enable white to start attacking our top right group. His goal here is to use the weakness of that group to build strength for his own new weak group. We therefore have to try to strengthen our top right group in a way that also puts pressure on white's corner, so that he has to or at least wants to keep responding there. That way, he won't have time to build strength for his (2) stone and we can go back and attack it later.

We therefore block with (3). I think it forces a response in the corner, because white can't allow our A. If we get A, our group would become very strong and white's corner group very weak. In fact, we'd be threatening to kill the marked stones. Defending 2 weak groups in the same area at the same time should be costly for him.

[Goodspeed]

White ignored us in the corner and played another move to strengthen his lonely stone at (1). Ultimately though, he had to come back and defend his corner and give us sente back. With the following moves, he lost some territory, allowed us to strengthen our top right group, and had to end the variation with a passive move at (7). This all seems pretty good for us.

We would be worried about our right side group below (7) if we hadn't strengthened it before by attacking white's bottom right group, so we're glad we did. This is a great example of how separate parts of the board may not seem connected, but ultimately are.

With our groups safe, we can now come back and severely attack white's (1) group. The question is how. We may not be able to kill it, but we should get some nice profit out of it or at the very least come out of it with sente, allowing us to play A.

[Goodspeed]

We played (1) to block the white group from running into our territory and force him to run to the center, growing our top left in the process. We also ended with sente and got to play the big move we wanted to play at (5).

Before playing this variation, we had to make damn sure our marked group in the top right would still be okay after white A. It's close but it should be able to make life in the corner, unless I missed something. The importance of strenghening that group before going back to attack white's group is now obvious. It allowed us to play a variation that gains us a lot of profit in the top left, as well as sente.

[chris1089]

Enjoyable read. Thank you for making the effort. Looking forward to seeing how this plays out, particularly if white can get a large territory in the centre-left without giving you too much.

Go is vastly superior to chess.

[Goodspeed]

White extended with (1), defending his stone and his bottom left territory. We now face a tough choice. B is a very tempting move here, but it can lead to a complicated position and therefore carries some risk. If we're ahead, we should avoid making things complicated and finish the game peacefully. If we're behind, we should actively seek out complexity.

The reason B could lead to complexity is our opponent might choose to play A afterwards, which cuts our stones off from each other. It would probably lead to both his A stone and our B group running towards the right, where both sides will find friends (in our case the N9 group, in their case the N13 group). But there would be opportunities to cut, and the groups we'd be running to aren't necessarily that strong either, so there's some uncertainty there. A bonus for us though is that white's bottom right group is still not strictly alive, and will come under pressure if we build more influence towards it.

If we play A on the other hand, our opponent will play B and we would both peacefully build our own territories. The question is: Would we be giving our opponent too much there? Should we go for the complexity or keep it simple?

It becomes important to figure out who is actually ahead. Below is an estimate.



Squares are our points, circles white points. It comes to 72 for us, 56 for white. Taking into account the 6.5 bonus points white gets for going second, we are ahead by an estimated 9.5 points. However, white still has some potential in the bottom left that we weren't able to count as points yet, but he will be able to convert into at least some points. On the other hand, it is our move next.

All said and done, I think we're only barely ahead. Considering B seems to me, ignoring the risk aspect, to be the better move, and considering it's definitely the most exciting move, we'll play that. YOLO

[Goodspeed]

So we played (1). White asked us politely to prove that we are alive in the top right, which we did, then went back to play (6). This is an interesting move. We suspect white intends to attack our right side group. We think we're okay there, so we went back and followed up on our (1) by playing (7). I don't think white can allow us to play A, capturing the marked stones, so surely he will have to respond. Of course, if he fancies his chances attacking our right side group, he can go for that any time and we will have to respond. But then he better end that variation with sente, or we get A.

If our opponent fixes his A weakness, we will play B. This move solidifies our territory, protects against white C ideas, and creates another friend for our right side group to run to.

[Goodspeed]

White indeed responded with (1), and we played the move we described in the previous post. White then cut off our marked stone with (3), which kills it. You may think: If it was that easy to kill, then why did we play it in the first place? Why not play (2) immediately?
First of all, it forced a response from white in (1), which is not a big move, so that means it could never have been much of a loss for us. Second, it meant (4) threatened to capture (3), forcing a response again. Lastly, in the current game position, the presence of the marked stone makes D extra painful for white.

We played (6) to prevent the cut at A. This is probably an overly defensive move, but I think we're ahead so we played it safe. White then cut off our entry to his territory with (7), and we pushed into it from the other side with (8). White can't afford to let us play D, so he will likely play it next (or E which serves the same purpose).

If we can find a way to play B, that would be great. This cuts off white's entire bottom right group. That group will then have to find a way to make life, which is probably possible, but as it struggles for life we can make profit and probably end with sente. If we do, we will play C and claim victory. By our count, there's not much white can do after that.

[Goodspeed]

A series of forcing moves follows. Our (2), (4), (6), and (8) stones are all dead. We played them to enable pushing through white's territory border with (10) and (12). This threatens to kill the marked stone, which would be big, or otherwise keep reducing white in the direction of B. White may have to spend an extra move to kill (2) and (4) once and for all, or he may have to choose between that and saving the marked stone. It seems like a good result for us ...

... except there's a problem. We may have inadvertently created a win condition for white by playing this variation with our weakness at A still open. Allowing white (9) made our group below A weaker, and white A became a whole lot more painful as a result. Hopefully, white doesn't spot this, but he probably will. Someone in his position would be looking for moves exactly like this.