Opposing komoku example
and form a opposing komoku position. (JP. Kenka Komoku)
It might be that many positions are also covered by the other fusekis, where they could occur.
tderz: In first instance, one would conclude that the position is less advantageous for White, because is as well a pincer as an extension from .
On the other hand ....
what to write here? Please comment!
- Perhaps one could play Wa+b?
- Another option is, e.g. to split up the left by offering to sacrifice .
Alex: Facing komoku is generally regarded as a bad idea for White, because Black will be able to approach first and thereby gain advantage, as here. His pincer supports his approach stone. For Black, playing a facing komoku is viable, because White must choose between approaching and allowing Black to take the last corner, or vice versa.
Playing a and b is thinkable for White, but regardless of how she plays, Black will always be a move ahead in the fighting on this side of the board, so there is no easy way out.
tderz: Alex, I agree with you about this fuseki being more stressfull and complicated for white up to disadvantageousness. As said initially, I played it experimentally on purpose. It was a serious club ladder game 45min/15s byo.
Alex: Sorry, I know I'm telling you things you know already - you're stronger than me, after all. That comment was directed more at a general audience, since facing komoku is a general strategic principle, not something specific to your game.
tderz: Thanks for the link, Alex, I will check it out! I was only looking for facing komoku, but did not imagine the search term opposing. I could not find anything in my fuseki books (except Kajiwara's generalizing "Move lost the game"
BTW, how do we now who is stronger - after all we never played, and my ideas might be deadwrong anyway.
Alex: Well, your homepage says you're 3d on KGS, whereas I'm only 3d CGA, which is a bit weaker than KGS ranks. But you're right, we don't really know.
tderz: I didn't show you my 1k?-account on KGS yet! OK, I got told and believe that I am not so bad in middle game, but perhaps my fuseki strength is only 5k.
Alex: Ah, okay. I often feel like I'm a fair bit stronger than my rank at fuseki and the early middlegame. Back when I was a 1d, I played a 5d in an even game, who told me afterwards that he felt I'd been beating him for the first 50 moves or so. Of course, I messed up horribly later. For me, the hardest part of the game is around the late middlegame / early yose, when I tend to get distracted by all the clutter.
I expected +(or b).
The idea was, that is further away (than white a) from the black top left corner and therefore cannot be pincered by Black c.
Alex: I prefer one line lower, since it's more severe against Black's hoshi in the lower left. Then if Black plays , White gets to make a double kakari.
tderz: I tried to find something in Joseki or Fuseki books, but unfortunately to no avail.
As I understand now, Black does not want to play +, as is too far away.
That's why he did only . (I guess)
Alex: Interesting. I've always thought of the - combination as a desirable shape in most circumstances and wouldn't have hesitated to play it here. However, I see the point that is far enough away to allow White to make use of her stone, while Black's thickness is already erased by the stable White group below.
tderz: one line lower did not occur to me, it is an interesting idea. Of course "there is no free lunch" (it's not better without disadvantages), because a 3-line extension is open to an invasion, hence actually need reinforcement (honte) or a double-kakari is not possible anymore (esp. if Black invades first and settles both Black and White)
Alex: Of course it has its ups and downs. What made me think of it is the same move cropping up in the orthodox fuseki (see "three space extension from the wedge" on that page). If Black invades immediately, it will probably end with White taking gote to capture the invading stone, while Black extends as shown there. So, White fails to gain sente, but is actually stronger than the two-space extension (at the expense of letting Black reinforce his corner). This strength will help White in whatever happens in the top left.
tderz: nice idea - next black a or b?
Alex: I think a. Black's pincer on the top side already supports his kakari, so I don't think any further play there is urgent. None of the common joseki that I can think of give White a result so good that Black would want to avoid it.
Alex: If Black just extends, White could slide at a and play the honte move at b, but since here - unlike in the orthodox fuseki - there was no checking extension, I would prefer to make an extension in the other direction from the wedge stone. That makes White safe enough above that she can probably just answer the invasion by taking the corner. also makes it a bit harder for B to decide how to handle . Disclaimer: This paragraph and associated diagram are straight out of my head, not based on any real games or theory I've seen, so they may be utter nonsense.
tderz: + look nice (for white, to me).
Alex: Me too. I'd be happy with this result, as White. However, it does seem to me that if this is the plan, White might want to have made his original wedge move on the fourth line. Or not. Maybe in this situation it's okay to be flat, in return for the added solidity.
Black Kosumi in the corner
Bill: Interesting discussion. :-)
It may be a little old-fashioned, but is worth considering. (I would play the kosumi-tsuke, myself, though.)
Then looks good.
tderz: What would/could happen afterwards?
Bill: Maybe Bp - Wq, Bm.
Are you (for White satisfied with Black a to e?
Yes (given the previous play). Then Wh. Black will make thickness in the top left, but Be feels a little close, and is working.
Or would White react differently?
How important is White f? Or should one rely on white g (or h?) after a black kake k?
What, if Black plays kosumi-tsuke m?
Bm is the reason for , in my view. I think that's a better formation than the hane, because of . Black is facing in the right direction to build a moyo.
How important is the n area?
I cannot help it, fuseki is to me a bunch of unanswered question.
Bill: After posting this diagram, I liked so much that I now think that maybe instead of a play in the top left is best. I showed this position last night to a US 5-dan friend. He immediately said .
tderz: You mean this keima-press-, right?
The is quite far away, White might dare to attach and accept the inside hane .
Bill: I don't think that works so well. It's hard to find a good play for White in the top right after . White wishes he had started on the 3-3 instead of the 3-4.
On the other hand, I find also the 's on the left very tempting (= tenuki on the right).
Bill: Very complicated. But is a little too far away, no?
tderz: If, after the right tenuki, Black approaches at , + look nice
(of course not all 3 are present :-) )
This -again - will depend much how the top left result will look like.
tderz: This is possible too.
Should White worry about how to handle it?
Excurs low vs. high approach
tderz: Alex, what you say seems to be the professional consensus. I checked a bit some books ("Fuseki", Vol. 2, Y27) and found indeed much more on the high than my low a.
Even the two-space-high kakari b appears.
Please notice that on the right is a low kakari in the references (vs. a high kakari in the game).
Commentary: "...white's pincer looks promising..." Will try this next time !
Alex: Yes, I like this. If White finds herself stuck in an opposing komoku situation (e.g. if Black made the opposing komoku on move 3 and White decided to take the last corner instead of approaching), a distant approach is a good strategy. This is because the advantage in an opposing komoku situation goes not to the player who plays first, but to the player who pincers first. Distant approaches discourage pincers, but if the opponent plays something like here, then it is the second player who ends up pincering first, for a good result.
The empty triangle
 What actually happened was, that Black kosumi-tsuked-ed ,
White didn't want to be heavy with tachi , hence played instead,
Alex: Here's another thought. Since you said that B doesn't want to play and (which is why he didn't start with ), couldn't you tenuki? That is, the usual reason you don't ignore a kick like is that it allows the opponent to play at for good shape. So if Black doesn't want to play there... what's he going to play instead? Certainly not a... there's no reason for Black to crawl on the second line...
tderz: Black did not like to crawl with = and give White a small wall (with later chance for some white invasion i), so he played for the direct cut
and ready is the mess!
wants to exploit the liberty problems of the empty triangle.
Alex: I don't think this is an acceptable result for White, as she is split into two weak groups and has essentially nothing to show for it. You wanted to avoid playing at , but ended up playing anyway. If light play was the goal, at would have been the normal idea, I believe. Is there a viable sabaki sequence starting with something like a? I don't know.
tderz: Light play was the goal if Black subdues with as reply (but he refused). Now White played anyway AND Black has an empty triangle.
Alex: I still don't believe that Black's empty triangle makes up for White's weak groups, since his corner is alive after .
tderz: (Alex: "white split in 2 groups is bad") If it had not been my own game, I had agreed with you immedeately.
So I was looking for higher authorities and could not even find one entry for my sequence - so it must be pretty disadvantageous.
tderz: By allowing the empty triangle out - white is not split.
or c and after follow Black a and White b (hence my is misplaced - too far).
Furthermore, Rui Naiwai comments "Black's corner territory is too big. This is White's loss... should be reserved for special situations"
Imagine a later black magari (bend) at d. This is powerful!
Alex: Yeah. Worse, there's very little aji in the marked stone. I don't think it would have to be very special situations indeed for me to want to play this.
Until now I have not found any sequence where books/professionals give their approval. ---> Conclusion, the low kakari ( in the game) leads to problems and should better be avoided in the future.
The stand & other alternatives
tderz: Almost in all diagram & references they stand.
White can continue with a, b or c.
tderz: Rui Naiwei states in "Essential Joseki" that the consideration of the relationship + vs. the lower left corner is very important
(I guessed so too :-) ).
tderz: Also this to is offered.
tderz: I quite like this:
is local tenuki, resp. elsewhere (.e.g. my C10)
Alex: Yeah, this is cute. Good example of light play. Though again, best saved for special circumstances, as Black is taking a fairly large amount of profit.
tderz: This seems to be an "old Joseki".
There is , however a tight-looking sequence after the black hane a (instead of ), which seems to be too good for the given circumstances here.
Similar(?) games in history?
A SCENE ON GAME DAY AT THE SHISHOKAI
tderz: Honinbo Shuei - Tanaka Masaki 3-dan with 2 stones in the year 1899
is here 2 lines tighter to .
tderz: please notice here, it's marvellous!
As usual in Go, the order of moves , and is important, Black is forced to play .
tderz: is more than a just a geta-type tesuji, it is also a strategic move which allows Black to take some stones.
In exchange, White gets a wall in sente.
tderz: White gets kakari with a backing wall.
tderz: Kawakita Konosuke (W) - Kishimoto Saichiro (B), Tenpo 11 (1840)
tderz: here is spectacular too, IMO.
Life of the black corner
White continued (in our amateur club game) with the hane before jumping to .
If black should take it with b (unlikely), white can sacrifice it for c.
On the other hand, if White gets a chance to connect at both b and d, the black corner can be killed by white e.
Black knows this and will attack the .
tderz: I would not been bothered by a - here (is honte? and better than a further stretch?) which is too easy for white, because she getas with .
Furthermore, Black does not need to capture now, because he can always capture at b, and even if both are saved and white b is connected make life then at f.
tderz: Black plays the very, very nice !
tderz: White cannot and should not move out with an empty triangle at g, because that one would simply be captured by Black at (black has always the push p in reserve).
Connecting at d only costs liberties and is thus bad style too.
tderz: The ugly is forced. If at g, Black t captures these stones.
What should Black play next? Your comments please!
tderz: Me too, yet unfortunately in this position I had the white stones
tderz: This is the position after the shibori.
Black has (at least) two choices, which are mutually exclusive:
- Black a
- if then White moves out with b, which induces Black c, then White d, each side has done bad and good moves; the evaluation must be: "Will the center stones be captured or the upper white stones?". Black has only one weak group.
tderz: After the shibori I prefer this over the suggested hane at a (= above).
White could defend now the immediate capture of her cutting stones by
- a <-> slow -> leads to capture of upper white group by black i,
- g is bad style and leads to fast defeat of the upper group: ....
tderz: Black g leads to variations like this.
White seems dead above.
Even if not dead, White is totally enclosed and stands very unfortunate.
tderz: ladders are ok for White. Still ...
tderz: ... this standard technique by Black is annoying:
etc. kikashis first, now Black must answer, if wants to hold up the squeeze.
tderz: By playing white h, thereby inviting black to squeeze, playing intermediate kikashis, the situation has somewhat changed though: the corner became vulnerable and the white top group is not as weak as it looks: ....
tderz: After -, Black must decide whether he wants to (try to) capture by Black k or make life with l.
Bill: How about ? tderz: keima , two diagrams above, is indeed not necessary for White.
tderz: indeed seems to capture by just 1 liberty.
at does not seem to work ( at ).
tderz: usually White should play at (direct atari), I guess.
Just for the ease of editing I play like this here.
tderz: How to continue for White?
tderz: Black captures white, therewith save his group, White only gets an edge ponnuki.
tderz: White got 6 liberties vs. Black's 4 in the corner.
...to be contnd.
tderz: threatens to capture Wc with Bg, hence White had to play there herself (gote). Then black could hane at h or play somewhere else.
tderz:  Perhaps this ? It has the advantage that it does not help the upper white group. Black could attack it later wit k or better (?) enclose it with m.
White can only connect to , not achieve much more.
On the other hand, once white has connected to with Wh, she could play d, if then Black connects underneath with u, killing chances vs. the corner reappear (however that depends on Black replying to Wy with Bz and on Bv and whether an exchange Bw-Wt has taken place or Bv-Ww (then Wx is possible)).
In any case, Black could always give up +, play Bf, then he has gained one move, because he did not answer the slow Wd-connection.
Black would not worry, White had then played there 8 stones only to capture 2 black stones.
So, up to now, I quite like this with the Bm follow-up.
White has not much to chose now, there is no time for Wp, Wh is urgent (Wd takes own liberties and leads to capture).
Game continuation: moves 27-
tderz: I wonder whether cutting now at is the best move.
fills at .
Bill: The shibori is still possible. In fact, it is better for Black after the - exchange.
After , maybe is best.
tderz: excactly, I thought the same.
Black always has the push p in reserve,
White could only save a dango.
The ugly above was forced. If at here, Black captures pivotal stones.
(White m is followed by Black n)
tderz: is forced, yet looks like a thank you move, helping White to get control of the stones.
Only playing at p or q (with r as follow-up) are surely better.
tderz: White gets the chance to throw in a cut at .
Giving another atari at s is unnecessary as Black does not want to play there, because that would strengthen White. Hence, White can always find the time to play there if needed and is not losing options now.
tderz: Thereafter it's time to extend to .
Around move , White was in a very arkward position having two weak, split groups with funny shapes.
Now come Black's moves which bring White back in the game.
tderz:  I find all moves to extremely bad.
They are saving cutting stones which split two weak white groups, which speaks for them, but ...
... this is no compensation for their undisputable, detrimental effect on the single .
Only would be better (which corresponds to a after move , earlier above ).
It seems, Black wants to attack the white -group.
However, there are now 4 groups involved: two white ones, two black ones.
White has got already compensation on the top, hence can fight freely and exchange easily if necessary. What should be White's next play?
Your comments please!
tderz: I thought deals best with the weak white group and tries to exploit 's shape problems (magari at the head of two stones).
At the same time the get a little bit weaker.
What should be Black's next play? Your comments please!
tderz: (=41) is understandable, but I wonder whether there are better moves.
Black hane h might look like a move having bad aji around black i or r in connection with the possibility of s, but first of all, White had to jump, perhaps to .
attaches, makes shape for White and further is looking forward to exploit the shape problems of the .
tderz: - is an often occuring shape - ending in gote for Black.
Should White attack with or bend at b?
tderz: makes a bamboo joint, extends.
- is a very bad exchange for Black, weakening the bamboo further, yet not contributing to black's strength, nor weakening white.
tderz: Not seeing anything particuliar, this allows to neutralize the left around n and jumping to somewhere in the j-area.
tderz: prevents the jump to the right, stabilizes Black above and attacks White somewhat.
tderz: Not seeing a direct capture, White squeezes.
Ed?: Black has to play 'b' or similar in response to because white has a ladder starting at 'c'. at .
tderz: Yes, the hoshi is not a ladder breaker
(careful though, should not be played at )
tderz: to followed.
Is there something more effective than ?
tderz: White gives atari once, in order to stabilize that group,
then takes care of the top group with =74.
What will happen after a black cut at x?
tderz: What a pity . I had hoped to find out, get told or have a discussion how to deal with the cut aji at x.
Moves, options: using the other cut at b, attacking at c, the bad push d and simply n sprang to mind.
tderz: Anyway, Black cut at (=).
White invented a special strategy with ,
namely attacking Black's center group and thereby strengthening the white group.
Most probably it's objectively more ambitious & wishful thinking than realizable.
On the other hand - subjectively - I did not really see a very strong counter by Black, so I just tried my plan - for fun.
Black connects with empty triangle. White creeps around with .
tderz: For defending with , I think that cutting at p is more effective.
On the other hand it seems that Black is planning to invade at q or r with .
White helps connecting a bit more with , wherafter Black feels obliged to wariuchi at . is for safety of the (lower) white group in connection with a destabilization of the black corner group.
tderz: Black attacks-extends with (to me =q= looks more overall connected & balanced).
White gets her way out with the very , which also separates the black center group. After there seems time to exchange with .
tderz: tries to make use of the .
After , 's achievement is only the firm protection of the cut x in sente.
With that sente tries to make some territory at last.
tderz: wants to make life for the white corner stone.
Because Black reacted (with , perhaps not necessary now),
that was in sente.
While seems justified in first instance, with the reason to make some white life in that corner,
the addition of Black 9 is not as important.
Deducing from there, this means that + were White's mistakes too (cf. dia. .
The difference between a black extension e and a white pincer h is so crucial, that most probably White 6 is the first mistake.
tderz: is the pincer, using White's thickness.
Imagine Black's last move () used for making a safe extension , which is also a kakari and neutralizing White's influence! (cf. dia.  That would be very tough on White)
does not want to help Black fixing up his shape.
Since around , White is in byoyomi (15s/move).
tderz: I give this an extra diagram :
This makes a safe extension, neutralizing White's influence!
tderz: secures and wants to create a framework.
Black invades with .
could (should?) have been tried at .
What should have been the next play? Your comments please!
tderz: - look blunt and are often not good style.
If White connects at b, than Black is left with something like only 1 eye around e.
Instead of , what could have been a more flexible approach?
What should have been the next play? Your comments please!
Use of influence
tderz: Being in byoyomi (15s/move) for about 20 moves now,
White must decide how to put the influence into something useful.
Connecting at b?
Connecting at b is just a second line move and only would create low territory in the fff-area, yet only if another move is added there.
If Black uses the influence of his upper wall and
invades before that around z, then White just created another weak group.
Reasons for White :
Hence, is used to enlarge the center area and close-off Black's right group's escape route.
Last, but certainly not least, it connects the left white groups which might be troubled if Black jumps into this area (who knows what would eventually work around the cut with x; e.g. starting with Black y) and protects the cutting point j.
What should have been the next play? Your comments please!
tderz: Black chose to play . This was most probably careless.
Why? Where was a need and
where was a place for active reinforcement?
How? (I am really asking for comments and finding good counter-play!)
What should have been the next play? Your comments please!
Trying to kill
unkx80: is painful. =)
tderz: Exactly - that's what happened (and was White's plan).
Do you think Black must die? (I mean, in the objective sense :-) )
Second question: If you had had Black with move ,
where would you have put it?
It should be a move which does not harm too much the right black eee group,
yet sufficiently defends the lower black group (now endangered by ).
Ed?: I think here is good for black to live, aiming at making a second eye by playing at (by threatening to break out) or a second eye inside. I don't think black can afford to play at as although he has access to the centre he will have no eyes on the side and white is pretty strong.
The Seki (I)
Ed?: White gets a sente seki and takes territory on the edge.
Ed? If white cuts at instead with from the previous diagram black can live but white gets very strong on the outside and takes some territory with . Black needs to gain an extra liberty.
Ed?: Black lives along the edge
tderz Yes, I think that is likely.
Ed?: Trying to kill black by playing here seems very bad since black is alive on the right, so the thickness built is useless, and this will destroy white's chances of centre territory if black lives.
tderz: Put it another way, that's White's only (?) chance to win? (I underlined if Black lives) This centre influence should be used for killing I think. I am too lazy to count territories, but after 2 alive black groups, he could nibble all center awayto s.th. meaningless.
At last there is some discussion coming up (again, 100 moves before Bill commented), thanks Ed?!
tderz: With my question above, I was pondering whether Black 135 (=) could be around here for active protection ending in Black's sente. After White replies with o or m, Black then has time to capture with l.
tderz: If White is greedy and tenukis for trying to capture the right group,
then - (or similar) is bad aji against White.
tderz: Connecting at d doesn't deliver eyes, hence it is useless to play there as a white ladder on seems broken by black B107.
Below  is a better for dealing with the black cut.
tderz: Perhaps best to take at now.
Black could chose among n or x and the big d, in addition to some y-z-combi.
White seems to have lost out.
tderz: The ladder x-y is good for Black.
tderz: Black must play better.
tderz: last resource?
takes ko, can not kill the right white corner? ( or = the t of threat?)
How come, I only see variations which are bad for White?
(That must be due to playing critical with hindsight against oneself)
The Seki (II)
Ed?: At this stage I think white has to play at and accept that black will get a forcing move in the centre. Connecting the circled black group at this stage to the centre with 'a' or 'b' looks gote so probably is left for later. I think black should try to connect his 2 groups with something like 'c'. White's group in the centre also looks very weak now, but if black attacks directly white can probably sacrifice while trying to kill the black group on the right on a large scale.
tderz great ! Thanks a lot. Who are you Ed?
Ed? I live in London and am about 1 dan, haven't had the chance to play much recently. How did the game continue?
tderz Whow! From your detailed L&D analysis of the group I thought you're stronger than me (perhaps you are). The group indeed died, the right side became a neutral zone and eventually I won the game by 30+ points (he didn't count, but shove the stones together). It was by far from flawless however:
- the strange opening  (ok, experimental)
- I only got back into the game by those early bad pushing moves -39
- Black missed one strategic move  when he had a Sente
- He played too softly in the middle game
- In 15s byoyomi, I almost lost one connection on the right side (not shown), when confronted with the question of how to connect after a peep. I was forced to live which still was possible without collateral dammage. If I had lost those, perhaps a swing of 30+ points would have been involved.
Ed?: Note if white had played the circled stone anywhere else,
would give a ko if ignored.
secures the corner.
tderz Great ! Same technique as above
(It takes me some time to realize that this seki is much better than the reflex a; must be a blind spot)
tderz: However, on the left does not yield much after the cut . I show some more lines of the board and it becomes clearer.
here just leads to a loss of stones (all!).
Ed?: White can take the corner with here
tderz: Nice, that was the communication I was hoping for!
Atari did not occur to me!
Ed?: White must be careful and may not be best. White doesn't want black to get an eye along the edge.
Ed?: White exchanges for to stop black from getting an eye along the edge (aji at 'a')
tderz: If (above in dias (3, 4) was meant as some kind of sente-measure for life, I guess the sente moves Wp-q against the seki and then the jump to would be better. Aiming for sente, White could rely on the box formation after hane , etc. and tenuki.
here is not sente, because the excess of outside liberties (OL) is now only = 1, whereas the number of inside liberties (IL) is = 3. OL - IL = -1 , hence white can't win now(only after Wq OL-IL=0 and the one who starts could win; please cf. the relevant pages on senseis). Black does not need to react on the seki side and can take the hane . Please notice also the danger of capturing only a san-moku nakade.
Could save white? What about the danger of Br?
Is White t better?
White will have no time to make a connection at c to save the upper bunch of stones.
tderz: a (usual?) jump to (instead of above) is dangerous.
(Perhaps) only at is safer and even better.
tderz: White seems alive in gote after s. (W hane x would be answered by Bs, because Wy-Bz does not work for White)
Pseudo Ko (I avoid Belgian) with at t, =, =throw-in at (iilegal move - there is a stone already), =t, =u is even better. (Sorry, realized too late that there is already a stone. That is the big problem with solving problems when typing. It's better only to look at an empty board. These numbers just confuse)
tderz: Black now could give atari at c just to end in with Wf, Bx, Ws, Bz kills,
no need assuring the seki with g.
Concluding, White could live (7) but should forget about saving her 3 stones (initiade by in seki diagram (4) above).