Why did my game become so close?

otenki · Post by **otenki** » Tue May 05, 2020 2:01 am

I have not been playing a lot lately but just doing some tsumego. Now back to playing on tygem.
In the end I won this game but I thought this game was way closer than I imagined it.
It felt to me like black made some slow moves in the beginning but then gradually got back closer the whole game.

Any pointers for me? (i'm white)

Thanks in advance!

Knotwilg · Post by **Knotwilg** » Tue May 05, 2020 11:08 am

I'll first make a no-LZ analysis. Then check with LZ.

When White finishes the local sequence at 28, the thickness outweighs the territory and Black has a useless stone.

For 30 I would not 3-3 but either approach at R6 or play on the left side to build, for example C9, or, if you really want to play AI-like, C5.

So, when 35 gets to play on the left side, Black's sphere of influence looks better than White's.

38 allows a cap on the attacking stone. Let' reverse the order. You play 38, White 39, would you then invade at 37, inviting 40. I think White should jump out.

White next has to spend many moves to create a living group, while Black gets territory and influence. The attack at 48 looks like an overplay.

52 looks slow again. Why not jump out and continue the attack?

66 White is playing envious go. The White group at the bottom still has aji, so I'd rather play C5 or C7 or C11 or at the top, P16

84 I think Black can fight here.

90 Black could play the ladder now

146 and subsequent, O3 looks bigger

170 yout ake O3 and for me the outcome is clear now

SoDesuNe · Post by **SoDesuNe** » Tue May 05, 2020 11:12 am

I think White played inconsistent in the opening and then chose to make a weak group. Over the course of settling this group, White gave away a lot of points to Black. But Black also did not find a good way to attack and later played really small and often gote moves, which allowed White to win the game in yose.

Some comments (EGF 3k ; )):

feels strange. Immeadiately settling after playing

light and wide, doesn't seem very consistent.

is nice.

builds some juicy center facing structure.

I'd have approached normally, seems to work better globally. I'd also thought about attaching D4.

Now it feels like all of White's initial play got countered.

would also have been my move.

I don't like at all ^^ Why not simply jump? Because of the press? If you would have jumped, maybe you could have cut? In my opinion White gets behind here.

is nice.

feels heavy.

seems flexible but if Black just cuts through, I don't see a good result for White.

I think, here Black starts to play a bit lax?

Up until now I think this is a reasonable result for White, giving that you tenukid once. Especially Black's three keimas in the middle are not doing much.

is just giving Black a solid corner.

Again I think White just gave points away here.

is slow. This at D6?

B105 has to be at J11?

B109 I guess bots would play attachment here. You don't enter the chinese normally anymore. Up until now I actually like Black and by my count Black is ahead in fix points and has more potential than White. After living horribly with B127, taking gote every so often and playing really small endgame: White won.

Knotwilg · Post by **Knotwilg** » Tue May 05, 2020 11:27 am

Now with LZ

When White finishes the local sequence at 28, the thickness outweighs the territory and Black has a useless stone.

For 30 I would not 3-3 but either approach at R6 or play on the left side to build, for example C9, or, if you really want to play AI-like, C5.

So, when 35 gets to play on the left side, Black's sphere of influence looks better than White's.

38 allows a cap on the attacking stone. Let' reverse the order. You play 38, White 39, would you then invade at 37, inviting 40. I think White should jump out.

LZ thinks that's all fine but she dislikes 42. M6 is 10% better

White next has to spend many moves to create a living group, while Black gets territory and influence. The attack at 48 looks like an overplay.

52 looks slow again. Why not jump out and continue the attack?

LZ thinks I'm talking nonsense. You're up by 98% already

66 White is playing envious go. The White group at the bottom still has aji, so I'd rather play C5 or C7 or C11 or at the top, P16

nope, LZ agrees with you

84 I think Black can fight here.

nope says LZ, but she thinks White has lost 20% in the driving sequence. PLaying 78 directly at 80 would have been better (1-2-3 priciple)

90 Black could play the ladder now

LZ agrees. Black would be back in the game with 40-60. However Black chooses another course and by 114 White is in the lead by 80% again

146 and subsequent, O3 looks bigger.

LZ wants to atari first. Anyhow you are ahead.

170 yout ake O3 and for me the outcome is clear now

For LZ the outcome has been clear ever since 150.

All in all, only the sequence 78-90 did some damage, otherwise you controlled the whole game.

Uberdude · Post by **Uberdude** » Wed May 06, 2020 12:40 am

Impressive game otenki, you've come a long way since we were reviewing your ddk games.
- First corner your opponent made the mistakes and you capitalised on them well to make thick shape.
- M4 then R7 combo showed an admirable advanced idea, a pleasant surprise clicking through. Funny that the other reviewers didn't like it

- k3 was first WTF moment, but you redeemed yourself with crosscut instead of pull back. But what if black extended instead of hane?
- coming out with o7 you justify r7
- but then spoil it with l6 not very useful gote move in middle of the fight.
- 66 I don't want to tenuki, m/N11 so big even if not sente for group strength and then top side potential if you cover black and make him worry about eyes. C6 attach is ok as screening kikashi so long as you take sente.
- driving tesuji was fakesuji as it didn't let you do anything good enough like kill centre to make up for local loss.
- e6 even worse, saving the 2 junk stones. Black gave you chance to H10. Here we can see these sequences would have made chunkier instead of potentially ovetconcentated centre of you already had M11 to stop black m10 reducing in sente hane.
- oh you are Tygem 5d now, that'd be why

Knotwilg · Post by **Knotwilg** » Wed May 06, 2020 1:44 am

Uberdude wrote: - M4 then R7 combo showed an admirable advanced idea, a pleasant surprise clicking through. Funny that the other reviewers didn't like it

Well, I'm happy that my weak understanding of Go at least serves your amusement.

In my no-LZ review, I thought White's play was inconsistent and used a kind of tewari to show that. LZ did approve and so do you. Is R7 then induced by M4? What would happen if R7 were played directly, omitting the M4-K4 exchange? Would Black take M4 for himself? That looks too close.

I hate to reduce the smirking potential here, but would love to understand the idea.

otenki · Post by **otenki** » Wed May 06, 2020 2:18 am

Thanks for your review, see my inline comments:

SoDesuNe wrote:I think White played inconsistent in the opening and then chose to make a weak group. Over the course of settling this group, White gave away a lot of points to Black. But Black also did not find a good way to attack and later played really small and often gote moves, which allowed White to win the game in yose.

Some comments (EGF 3k ; )):

feels strange. Immeadiately settling after playing light and wide, doesn't seem very consistent.
-> yeah i got ahead here I think.

is nice.

->
builds some juicy center facing structure.
-> I assumed black would at least G16 instead of capture.

I'd have approached normally, seems to work better globally. I'd also thought about attaching D4.

-> I personally don't like D4 because a pro once showed me d5 as an anwser.
Ever since that I almost never play that. Probably it is playable though.
Aproach can work I guess, matter of taste ?

Now it feels like all of White's initial play got countered.
-> True but at the same time it was not the biggest move on the board right now.
If we follow simple logic of corners first he should have aproached my 3-4 at the top ?

would also have been my move.
->

I don't like at all ^^ Why not simply jump? Because of the press? If you would have jumped, maybe you could have cut? In my opinion White gets behind here.

-> I think both jump and playing on the right side are almost the same.
I'm preparing for a fight between the two groups on the right.
If I jump in the center and black follows with k6 I still need to play r7 right ?

is nice.
-> It feels like overplay to me, maybe a stronger player can comment here.
-> main point is there are lots of holes and black right side not connected and white has defended the left allready.

feels heavy.

seems flexible but if Black just cuts through, I don't see a good result for White.
-> yeah regretted this move later. I think push might be better to make sure you don't have to make this exchange in a worse situation

I think, here Black starts to play a bit lax?
-> he needs to save his group no ?

Up until now I think this is a reasonable result for White, giving that you tenukid once. Especially Black's three keimas in the middle are not doing much.
-> yes

is just giving Black a solid corner.
-> true, maybe this is not required.

Again I think White just gave points away here.

is slow. This at D6?
-> not sure

B105 has to be at J11?

B109 I guess bots would play attachment here. You don't enter the chinese normally anymore. Up until now I actually like Black and by my count Black is ahead in fix points and has more potential than White. After living horribly with B127, taking gote every so often and playing really small endgame: White won.

jlt · Post by **jlt** » Wed May 06, 2020 2:24 am

@Knotwilg: I got lost in your tewari analysis. The sequence of the game is the following.

Click Here To Show Diagram Code: [go]$$Wc Moves 36 to 39 $$ --------------------------------------- $$ | . . . . . X . . . . . . . . . . . . . | $$ | . X X X X . X O . . . . . . . . . . . | $$ | X . X O X X O O . . O . . . . . . . . | $$ | . X O O O O . . X , . . . . . , O . . | $$ | . O . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . 3 . . | $$ | . . . . . . . . . . . 4 . . . . . O . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . X , . . . . . 2 . 1 . . . X O . . | $$ | . . . . . X . . . . . . . . . X O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

How exactly did you reorder the moves? Could you make a diagram?

Knotwilg · Post by **Knotwilg** » Wed May 06, 2020 2:28 am

Click Here To Show Diagram Code: [go]$$Wc Moves 36 to 39 $$ --------------------------------------- $$ | . . . . . X . . . . . . . . . . . . . | $$ | . X X X X . X O . . . . . . . . . . . | $$ | X . X O X X O O . . O . . . . . . . . | $$ | . X O O O O . . X , . . . . . , O . . | $$ | . O . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . 1 . . | $$ | . . . . . . . . . . . . . . . . . O . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . X , . . . . . , . 2 . . . X O . . | $$ | . . . . . X . . . . . . . . . X O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

If W1 would Black play B2?

Click Here To Show Diagram Code: [go]$$Wc Moves 36 to 39 $$ --------------------------------------- $$ | . . . . . X . . . . . . . . . . . . . | $$ | . X X X X . X O . . . . . . . . . . . | $$ | X . X O X X O O . . O . . . . . . . . | $$ | . X O O O O . . X , . . . . . , O . . | $$ | . O . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . 1 . . | $$ | . . . . . . . . . . . . . . . . . O . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . X , . . . . . 2 . 3 . . . X O . . | $$ | . . . . . X . . . . . . . . . X O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

If B2, would White play W3?

If yes to both questions, then I understand why Uberdude likes Otenki's ideas.

otenki · Post by **otenki** » Wed May 06, 2020 2:30 am

Knotwilg wrote:Now with LZ
When White finishes the local sequence at 28, the thickness outweighs the territory and Black has a useless stone.

For 30 I would not 3-3 but either approach at R6 or play on the left side to build, for example C9, or, if you really want to play AI-like, C5.

So, when 35 gets to play on the left side, Black's sphere of influence looks better than White's.

38 allows a cap on the attacking stone. Let' reverse the order. You play 38, White 39, would you then invade at 37, inviting 40. I think White should jump out.
LZ thinks that's all fine but she dislikes 42. M6 is 10% better
White next has to spend many moves to create a living group, while Black gets territory and influence. The attack at 48 looks like an overplay.

52 looks slow again. Why not jump out and continue the attack?
LZ thinks I'm talking nonsense. You're up by 98% already
66 White is playing envious go. The White group at the bottom still has aji, so I'd rather play C5 or C7 or C11 or at the top, P16
nope, LZ agrees with you

84 I think Black can fight here.
nope says LZ, but she thinks White has lost 20% in the driving sequence. PLaying 78 directly at 80 would have been better (1-2-3 priciple)
90 Black could play the ladder now
LZ agrees. Black would be back in the game with 40-60. However Black chooses another course and by 114 White is in the lead by 80% again

146 and subsequent, O3 looks bigger.
LZ wants to atari first. Anyhow you are ahead.

170 yout ake O3 and for me the outcome is clear now
For LZ the outcome has been clear ever since 150.

All in all, only the sequence 78-90 did some damage, otherwise you controlled the whole game.

Yeah thank you for the review!
I especially liked your first part since thats the real review, leela can be so hard to interpret.
On move 42, I agree but these kind of things are so hard. I thought that L5 (i assume you mean this instead of m6 since black is allready at that position) exchange would be aji removal so I ommited it. But clearly I should have made this exchange first, letting black decide then and there if he wants to block.
90 Black could play the ladder now ---> I don't understand this comment, maybe I misread but I think there is no ladder here. And if you try to take the stone now I guess its about the same result as the game. only difference is white has a choice to run or not?
I should have played kosumi or keima instead of the driving-nosuji like you mentioned. Bad habbit of me!

Thanks!

otenki · Post by **otenki** » Wed May 06, 2020 2:54 am

Comments inline:

Uberdude wrote:Impressive game otenki, you've come a long way since we were reviewing your ddk games.

-> Thanks, kind of you to say!

- First corner your opponent made the mistakes and you capitalised on them well to make thick shape.

->

- M4 then R7 combo showed an admirable advanced idea, a pleasant surprise clicking through. Funny that the other reviewers didn't like it

->

- k3 was first WTF moment, but you redeemed yourself with crosscut instead of pull back. But what if black extended instead of hane?
- coming out with o7 you justify r7
- but then spoil it with l6 not very useful gote move in middle of the fight.
- 66 I don't want to tenuki, m/N11 so big even if not sente for group strength and then top side potential if you cover black and make him worry about eyes. C6 attach is ok as screening kikashi so long as you take sente.

-> I have to admit I did not really know a good technique to get my group into better shape. Indeed if he extends I have a problem ... I said it before in this thread bug L6 would be so much better. But it requires me to be patient and understand my group will be fine. I need to be able to read/judge the follow up of that situation better in order for me to dare to make that move.
66-> I regretted this afterwards during the game, I should have taken n/m11 like you said.

- driving tesuji was fakesuji as it didn't let you do anything good enough like kill centre to make up for local loss.
--> yeah so bad habbit.

- e6 even worse, saving the 2 junk stones. Black gave you chance to H10. Here we can see these sequences would have made chunkier instead of potentially ovetconcentated centre of you already had M11 to stop black m10 reducing in sente hane.

--> I was afraid of the h7/g6 cut actually. I could not read if the center would be dead (and thus my stones alive). But you are right, I don't need to defend that cut.

- oh you are Tygem 5d now, that'd be why
--> Yeah but still only egf 2k (my last tournaments are a while ago and were not good) and trouble keeping at 5d, might fall back to 4d. With these reviews I hope to be able to keep tygem 5d soon Thank you

Uberdude · Post by **Uberdude** » Wed May 06, 2020 3:13 am

Knotwilg wrote:
Uberdude wrote: - M4 then R7 combo showed an admirable advanced idea, a pleasant surprise clicking through. Funny that the other reviewers didn't like it
Well, I'm happy that my weak understanding of Go at least serves your amusement.

In my no-LZ review, I thought White's play was inconsistent and used a kind of tewari to show that. LZ did approve and so do you. Is R7 then induced by M4? What would happen if R7 were played directly, omitting the M4-K4 exchange? Would Black take M4 for himself? That looks too close.

I hate to reduce the smirking potential here, but would love to understand the idea.

Locally speaking with the corner R7 is a very valuable reverse sente move, that prevents black from playing the r6 attachment as a way to build some solid shape for his wall (e.g. if white extends out at s7 then you can make r2 p2 in sente, creating emergency connection around m3). So it's a thick move that prepares to attack the wall, though black still has the q7 attachment as a resource (often black will make the q7 q8 r6 s8 p8 q9 exchanges early, because later in the fighting, e.g when white is running out the o7 area q7 might not get an answer). White could have played r7 directly instead of m4, indeed whether to play this solid preparatory move, and give opponent a chance to play defensive extension on lower side, or pincer directly and allow black the choice of counter pincer on lower side or r6 is a tricky question, I'd say generally both are decent, only need to worry about picking the wrong one at high levels. r7 directly could be considered more honte slow-and-steady style, pincer more fighty. If white does r7, then black has a choice of what extension, and whilst he might pick k4, and then it reverts if white invades at m4, maybe he would:
- tenuki to some other big place (e.g. upper right). Then if white is first to play on lower side, black might admit that the 3 stones are weak so might not play the k4 counter pincer which says "m4 is weak and I want to attack it (but as you run out my q4 wall gets weaker too, so maybe it's a bit overplay)" but from q7 direction or even tenuki to another big place. k4 commits to making the lower side fighting more urgent and un-tenukiable
- play a closer, more defensive extension like l4
- q7 directly: if black can make the standard sente kikashis, then making the further extension to k4 is safer than directly playing k4. The q7s become a probe of "do you want to let me get some moves in sente (which do help white build right) to help my wall so I can make a big extension with better synergy with lower left moyo, or will you tenuki so I get nice followup locally but you could dive into lower side first?". But the key point here is black follow up after white r7, black q7, white tenuki is not as good for black as no white r7 and black first (at r6 or r7 himself), the classic principle of as more moves are invested in an area the value decreases, first to get a bite of the pie gains most.

I'm not saying L4 IS the perfect extension for black after white r7 directly, but if it is, the idea is similar to the split a side and then tenuki the checking extension idea (Bill I think mentioned some old pro like Takagawa wrote about it?):

Playing the m4 pincer, then taking r7, does mean you have to endure the cap and handle the subsequent fight well, and it's easy to make bigger mistakes in that fight than the benefit of m4 r7 combo, hence being an advanced idea. It's easy to imitate a bot in doing that move combo, playing the fight as well as a bot is harder. The key idea is white only has one weak group to handle and is planning to counterattack the black wall if black overplays, the connection between m6 and q5 is thin. It also depends how much real vs loose profit black can build in the attack, the left side is actually pretty nice for black here so I personally would be nervious about using this technique and might have just jumped out for white, if d9 area was a white stone I would feel more comfortable doing it.

SoDesuNe · Post by **SoDesuNe** » Thu May 07, 2020 8:31 am

Thanks Uberdude for the insights!

otenki

I looked into D4-D5... gets a little bit crazy, I can understand ^^ But generally I think White is fine here.

Hm, good question whether the low chinese corner is a better place. I'm from ye olde times where Black really doesn't want to enter there anyway ^^ It will be a good guess that the bots evaluate things differently now.

+

I share Uberdude's concerns here. Your tenuki might well be playable in bot terms but you bank a lot on this fight turning okay for you. Meaning Black does not end up consolidating his framework in the bottom left (which he could have, if he had cut your keima), while accessing the center or even profitably sacrificing his three-stone-wall. It's too do or die for my taste ; )

Bill Spight · Post by **Bill Spight** » Thu May 07, 2020 10:59 am

The more I study bots, the more I am impressed with Shusai's play.

Click Here To Show Diagram Code: [go]$$Wcm36 I am not impressed. $$ --------------------------------------- $$ | . . . . . X . . . . . . . . . . . . . | $$ | . X X X X . X O . . . . . . . . . . . | $$ | X . X O X X O O . . O . . . . . . . . | $$ | . X O O O O . . X , . . . . . , O . . | $$ | . O . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . a 1 . . | $$ | . . . . . . . . . . . . . . . . . O . | $$ | . . . . . . . . . . . . . . . X . . . | $$ | . . X , . . . . . , . . . . . X O . . | $$ | . . . . . X . . . . . . . . . X O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

I think the bots are not impressed by Black's large sphere of influence, and Shusai was not, either. Shusai's problem was that Go Seigen was a genius.

I think that Shusai might well have played

, or maybe the keima at a.

Click Here To Show Diagram Code: [go]$$Wcm36 I am impressed. $$ --------------------------------------- $$ | . . . . . X . . . . . . . . . . . . . | $$ | . X X X X . X O . . . . . . . . . . . | $$ | X . X O X X O O . . O . . . . . . . . | $$ | . X O O O O . . X , . . . . . , O . . | $$ | . O . . . . . . . . . . . . . . . . . | $$ | . . . O . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . X . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . O . | $$ | . . . . . . . . . . 3 . . . . X . . . | $$ | . . X , . . a . . , . . . . . X O . . | $$ | . . . . . X . . 2 . 1 . . b . X O . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]

Well, I am impressed, myself.

My first thought was the shoulder blow at a. But now I like the 3d line wedge. I think that White has the lead, and I think that the wedge is easy for White to handle. I anticipate

and then the jump to the center for

. But maybe the extension to b is OK, too.

Harleqin · Post by **Harleqin** » Fri May 08, 2020 2:32 am

If you really want to keep black's sphere of influence in check, then b certainly seems the wrong direction to me.

Life In 19x19

Why did my game become so close?

Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?

Re: Why did my game become so close?