How to win against 9 stone handicap
Some among you might think me setting up this page a waste of time, but for the terminally curious among us, I'd like to have at least some idea of how this sort of game goes about.
Now, my (amateur) analysis. A kakari at a seems like a decent start... but what if your opponent accepts the invasion and decides to tenuki? Or, if he decides to give you the corner, in exchange for a solid wall? ... hmmm... I guess if he ignores it, you could go ahead and build a moyo, but how are you going to win in a moyo building contest?
Playing at b has the same disadvantages I guess, and worse, attacking that corner that way seems like awfully heavy play to me...
How about c or d? I'd say that move has to get a response from Black, since giving up the center like that isn't such a hot idea. On the other hand, can the center be invaded like that, with any chance of success at all?
For Black, I assume the best the best moves might be e or f for some quick territory, or maybe g for aggressive moyo building?
Any feedback at all is encouraged, perhaps an example of what the first couple of moves should be like, or better, maybe an example game?
cazort: Whenever I invade and allow black to build a wall (which is a common occurrence in 9-stone handicap games), I try to play stones to make black's wall ineffective. There are different ways to do it--placing your stones near to the wall, if you can get away with it, is great. More often I settle for cramping black into an overconcentrated shape, which can still be highly effective.
adamzero: This is the basis for any strategy I have in a nine-stone game:
As White: Cut black groups apart, stay connected, play lightly, never force him to strengthen his stones, take sente and tenuki constantly, nearly always take influence over territory, because White cannot win in a territory race but only by using his strength in the middle game battles.
As Black: Never ever let White connect his groups, play thickly, use kikashi but never ever if it means aji keshi, destroy white's shape.
I think of the game as an invasion into a moyo from the first move, and play accordingly, as either color.
However, I don't feel that I'm too great at 9 stone games, so I eagerly await what the strong deshi have to say.
Also, this would be a lot more usual game, at least in my experience:
White plays 5 as a sort of probe, to see how Black intends to react, and to get something to chase as a way of settling his two initial kakari. Later he aims for moves like a, to start more trouble, and b to get a base. Black must keep all of White's stones separated, while making sure to stay connected to avoid dying to White's superior fighting strength.
Stefan: You also often see 3 played one space to the left. In general adamzero's advice captures it very well. I would only add "as White, throw away whole bunches of unimportant stones. It is good advice for any game, but in 9 stone games it is particularly effective in giving Black the impression he's doing supremely well". By the way, in handicap games White's first move is usually marked as '2' (not that it matters much).
Floris: A small variation to this is also to randomly slam down a few stones in the opening and prove them useful later on. This strategy is still work giving 9 stones to 9 and weaker. I am expecting this to get more difficult as i become stronger and giving 9 stones to 3-5k range.
Pashley: As Black, I'm inclined to play 4 at 5. It is a bit risky, but more aggressive. Of course, White is likely to answer somewhere near 4 and try to give me grief in that corner, but c may be an adequate answer.
Dieter: I get to play nine stone handicap games fairly often against fellow club members. I do not care at all if I win or lose. 9H's are teaching games. What I do is play on a large scale (the above joseki are a good example), play flexibly, invite Black to secure small territory while getting the advantage in a large central zone. Killing a big group is often the decisive blow. I do not try to psychologize my opponent too much. If they have learnt to cut, to connect and to throw away unimportant stones, they will win and shift to a lower handicap.
By the way, the letters in the diagram at the top are intended one line lower, no ?
Jasonred : One line lower? Erm, sure, I knew that... (furious scribbling in personal book of joseki, ala Tsutsui from Hikaru no Go...)
Charles Matthews A typical start to a nine-stone game in which I'm White (I probably play 50 of these nine- or eight-stone games a year against players mostly 9 to 7 kyu).
As White I aim to develop quickly over the whole board. The idea is to stay out of trouble.
Comments: Black has already made some mistakes. Black 2 at a is somewhat better (larger-scale). Black 4 at 9 is preferable, though this can't be called a serious mistake. Black 8 at 9 is another joseki, again preferable because on a larger scale. Black 10 is wrong (play at b).
A continuation without Black making serious mistakes, but White able to move around the board. Next White at a can start a fight in which the marked white stone can prove useful.
Clearly White has only caught up a small distance, so far. Currently Black might take the whole right side (but that would require three more plays).
Jasonred: Am I right to assume that the trade was the b area for the c area? Which probably is a good idea, but being me, I don't know how to go from there. Make a moyo on the top, and try to connect it to the lower one for White, or expand from there, or just leave it as is?
Charles Matthews White gets a strong group on the left side (well, strong in nine-stone game terms). White can indeed expect about ten points of territory here. But you should look at other effects. For example White 7 and 9 are weak(ish) but their presence makes the marked black stone on the side weak(ish). It might be a good target a little later. And if 7 and 9 get into trouble in fighting, White can play at the circled point and reach out towards them: Black wouldn't yet be able to capture those stones on a large scale. This matters because White will have to invade several times before she gets into the game, and doesn't want to worry too much about stones on the outside.
So, White ought to be playing a 'long game' plan - win in the endgame; and spends the early moves thinking how to deny Black good prospects (attacks, frameworks).
Jasonred: I think I understand some basic principles about what you guys are saying, I learnt something today! Now, I can sort of see why adamzero's sequence could be practically joseki, but how about if you attempt a shoulder hit, and your opponent plays tenuki: how to capitalise? Take the corner? Eg, my opponent the other day made a sort of weird play:
Maybe he was trying manego on me... anyhow, once another opponent played at a instead of 4. What now? Also, I suspect my attack on the corner with 3 was in the wrong place, corrections please?
adamzero: It seems to me that your opponent was betting that he could make territory more easily than you could gain profit by attacking. In a nine-stone game its a reasonable strategy if White is not that skillful. The biggest flaw with this plan is one built into the structure of a handicap game: all of Black's stones are on the fourth line, meaning they give influence, not territory. The best way to make territory from them is to use them to attack, or to force invasions which you can attack.
If they were on the third line then simpler profit taking would work better. As the game stands Black has made almost no territory because of the invasion point at b. If White can reduce the area in front of the corner before the invasion, then Black has gotten nothing.
Could you demonstrate White making a live group here, please? It's beyond my meager abilities to successfully invade here...
Charles Matthews White has two eyes in this diagram, doesn't she? But Black 8 in the previous diagram is a weak play.
I've added a standard invasion sequence on the butterfly page.
adamzero Also, in reply to your other question: First rule of shoulder-hitting: don't play tenuki, the follow-up should be (if the shoulder hit was a worthwhile one) too big. Also, a shoulder hit should only be used for keshi or to build strength when making shinogi or escaping. When you shoulder hit you strengthen your opponent's stones, which you never want to do in a corner. See more info at katatsuki. Last note: in a nine-stone game, there is not, or at least ought not to be, any moyo for White.
Jasonred : I think I like the joseki from the first diagram best... how about other people? For me, I tried it, and if my opponent follows it, well and good, if not, and I get the top left corner, with the other two stones, I've basically stolen the entire top of the board, no?
Charles Matthews This kind of capping play is used even by pros. It can create early confusion and fighting. My philosophy is to be less confused than Black; and I don't use it much. But that's personal taste. If you cap as White, you can learn how expanding and running fights have an impact on surrounding areas (which is interesting). My way relies more on neutralising Black's influence before fighting.
By the way, this formation is called the butterfly (Korean name). Traditionally it has a bad reputation - as was said above, it doesn't prevent invasion in the way that a play at a would if you moved the marked stone there.
Like so many old ideas, though, that is questioned in current Korean pro play.
See also: How to win with a 9 stone handicap.
Cazort: I'm not the best player but I sometimes play 9-stone handicap games as white, especially against newish players. I find that the hardest thing for me is to ensure that I have enough solid groups reaching into the center so that I can effectively reduce black's center territory. When I lose such games it is always because of a large black center territory. Anyone have advice on how to do this? It's hard because starting plays on the fourth line and above can easily create weak groups for black to chase.
togo: Being chased around is actually a good thing with 9 stone handicap. You will reach every region on board. But you have to play lightly of course. Further reason for this working so well is, that black will give you eyes by error. So you are losing points by working for them by yourself.
tapir: I believe much of this page is misguided. The main thought when playing 9 stone handicap games should be, what will the player receiving handicap will learn from it. I tend to emphasise: it is not easy (and not necessarily good) to aim for territory too early (to get this message across I work hard against players who simply try to take territory and win with 5 points after receiving 9 stones), strength is for fighting, but fighting needs patience - keep strong in the meanwhile, sacrificing is good, throwing stones away without profit isn't, white stones are mortal, you can't get everything...
(dent)-- In the book BasicTechniquesOfGo Chapter 2 deals with playing a 9 stone handicap game:
There is some pretty extensive explanation for why 7 and 8 are Joseki. There is also some variation. But then the author explains how, when white plays elsewhere, 'a' becomes a big point for black. The 'iron pillar', he calls it, saying that this forces white to jump out at 'b'.
So my question is, why doesn't white play 9 at 'a' instead of making another approach elsewhere? What is the inherent weakness of this play?
It's mostly just unpalatable to be left with the chance for Black to play tenuki. White doesn't have time to spend sente twice in this narrow area, and you can see how 'b' will work with something like 'a' if Black gets both.
But suppose Black just responds with normal shape in gote, as White might want. White secures the group in sente, but allows Black to strengthen without giving the opportunity to make mistakes. This is unpalatable in a 9 stone game. White must rush to other areas and hope for the best on some points. Black is likely to be much worse at attacking on a global scale than at making locally solid shape, otherwise Black wouldn't be taking 9 stones against scum such as yourself. Being pushed out if White fails to reinforce the group's eyespace will surely be unpleasant, but there would still be some chance that it wouldn't happen due to Black's weakness in taking sente, and even if it did, the running battle is likely to feel more comfortable to White than it would in an even game. That said, there is some reason for White to try to tenuki in a way that will give some peripheral advantages to this group, should it be forced into the center as you have described. It's not a very light group now, so is becoming hard to discard.