Fuseki Experiments

    Keywords: Opening

It would be nice to discuss some unusual fuseki patterns on this page. Who's come up with a weird idea they'd like to share?

An example  

Black is experimenting. White doesn't have to answer this way of course, but it's one of the more common responses. The first time I had a dan player as opponent at a tournament, I played this way (I was 3 kyu at the time). After playing a weaker yose I lost by 3 1/2 points. So I haven't lost confidence in this pattern.

A second example  

The 7-7 point. I've tried this one too, but don't like it. Why did I even try it? I heard the rumour of there being a German 5 dan who claims that this is his favourite first move with Black. Does anyone know more about this? Can anyone come up with a reason why this might be a good move? (Mark)

Not 7-7, but I don't know another eccentric German 5 dan, so: Robert Jasiek: "Since autumn 2004 known for starting with the 9-7 as Black."

Dieter: I know for fact that Pierre Colmez 5D played this opening against Guo Juan 7D, in the Brussels Grand Prix some years ago. After B7 both players were laughing, though I'm convinced that Guo didn't think at all that victory was assured yet.

Guo (W) - Colmez (B)  

IaGo: I recently saw this center fuseki on KGS, and I would like to make a few experiment with it, so if you want to give it a try, maybe we could make an open game here or on KGS. Just let me know :)

Very interesting. It would be nice to see the game record. I played that fuseki several times on KGS a year or so ago, with reasonable results. But I'm much weaker than 5D, and my opponents were not professionals either. I really didn't expect to see it has been played at that level.

More recently, I have sometimes played the following pattern. White can of course play differently, but quite often White takes corners and waits to see what is going to happen.

I think I prefer Black in this diagram. Black stones just seem to be working better together. But it's hardly a won game yet. Any opinions? Ideas for good continuations?

-- Miz

Schmorp: I played this pattern quite a number of times against Gnu-Go 3.2 on 13x13 and 19x19, and, depending on how Gnu-Go answers (sometimes it takes the points itself), this seems to be a successful opening. I guess Gnu-Go is only prepared for more standard ones. All in all, if I am able to also get two of the 4-4 points (Gnu-Go often attacks one of my early moves), invasions by Gnu-Go are not successful and I win the game.

Corners first?  

French Championship 2002, second game  

In the second game of the French Championship, relayed live on KGS 2002-11-10, Pierre Colmez opened in the center against Farid ben Malek. For much of the game it seemed as if Black was heading for a win but Malek won the game and went on to win the third game and the championship. To retrieve the game look on KGS under the User Information of ChampFr for the first game on 2002-11-10.

You can see this game (SGF or mori applet) on [ext] the Paris Go Club Forum (game2).

-- Jean Pierre Tavan?

I found an interesting fuseki by Takemiya, and was wondering if anyone would care to comment.

Takemiya fuseki  

This is a very intriguing pattern to me, with both sides playing Yonrensei and then with Takemiya switching to a very unique pattern. It obviously emphasizes the center, a trademark of Takemiya's, while his opponent (Cho Chikun 9 dan) focuses on the left side before invading with 6. Any thoughts?

Lynx: The thoughts of mine, a lunatic who is responsible for such wonderful bastions of wisdom such as The Power of the B2 Bomber are the following: All this only goes to show that we really have no idea whatsoever of how the game truly works.

"What?" some of you may react automatically. "What has this crazy person been drinking? Of course we know that corners are the uber good!!!!"

Do we? Prove it.

There lies the main problem. There is no way to "prove it." We can only judge by what is effective.

See also Great Wall and Orion's Belt.

19x19 diagram  

jwaytogo: Here is a fighting style fuseki that I used as black for a while. Not bad, got pretty good results with it.

Tamsin: I believe it's important to ask yourself why perform experiments at all? If you're simply trying to trick the opponent, then that seems counter-productive. I have started making some experiments, but with the intention of understanding more about go. If an experiment is conceived to put a hypothesis or creative idea to the test, then I'm sure it can be very helpful. But if you're just playing unconventional moves for the sake of being unconventional, then it's a fast-track to a heavy defeat.

- Take the Miyamoto Fuseki for example. This is worth trying because there are some definite ideas behind it. It is an experiment with a purpose. - - Likewise, consider the situation when White chooses to make a kakari on her opening moves instead of taking an empty corner. She ends up appearing to be behind, but by playing tenuki at the right time she can start to take back the lead in other corners. The idea is not that taking empty corners is small (which is definitely isn't!), but rather that one set up ways to make favourable trades in big places. The lead is only relinquished temporarily, because when White plays tenuki in one corner, she makes creates as situation in another where she has effectively played first. - - For example. Assume Black has taken four corners in succession, and in each case White has played an early kakari. Play might proceed like this: - - Corner 1 Black plays his first follow-up; White plays tenuki and plays in Corner 2. - - Corner 2 White now has the first play in Corner 2. Black plays tenuki, and plays a second follow-up in Corner 1. - - Corner 3 White now plays another tenuki in Corner 1. She uses it to take the lead in Corner 3. - - Corner 4 Black decides to reply to White's lead in Corner 2. White chooses a variation which ends in sente. She uses sente to make another leading move in Corner 4. - - End Result: Black has played 3 times to 1 in Corner 1, but White has taken the initiative in 3 other corners. If she still has aji in Corner 1, then it is quite possible that she is now in overall lead of the game.

Let's take a more concrete example:
19x19 diagram  

White appears to be behind, doesn't she?

However, what if play continues according to my scheme:

19x19 diagram  

Black plays the large-looking attachment at 1. White decides to play tenuki.

Suppose Black is determined that 1 was sente, and decides to follow it up.

- >Bill: Now, why would Black do a dumb fool thing like that????? Besides, B1 looks better at a.

- >Tamsin: Steady on Bill! I agree playing B1 at 'a' is better, but B1 is often seen in isolation. Further, I don't think 3 is at all a foolish move, although I doubt many people would play yet again if White continued to ignore.

- >>Alex: I think Bill's point is that you're criticising the original contributors to this page for playing unorthodox moves simply to confuse their opponents (which I agree is a bad habit), but you're committing a sin similar to self-indulgent reading, which is arguably even worse. That is, you claim that your fuseki experiments have thinking behind them, but what you've shown here only goes as far as: "If my opponent plays as I hope he does, then I get a good result," which is true of pretty much any plan, be it sound or not-so-sound.

- >>>Tamsin: Why are you being so harsh? I think often people DO play unorthodox moves without any specific plan, and this IS a way to get into trouble. I did NOWHERE criticise the original contributors to this page. The Great Wall, for example, obviously has a serious plan behind it. Also, I have investigated a number of ways in which my plan can go, and I stated that I am STILL investigating them. I only had time to put together one example of how things might go, and that example was intended to show how White could get an advantage if Black played reasonable-looking, but not necessarily good moves. Pardon me for not having six or seven hours to spend just to write maybe 10 examples!

- >Bill:If you are attempting to assess the worth of a position (and the strategy that produces it) by looking at further play, having one player continue with inferior moves does not help. In the following diagram, if B1 is joseki, I missed that class. B3 is OK locally, but the bottom left corner is more urgent. B5 is silly this early in the game, but that may be offset by the fact that Black gets to play two stones. ;-)

- > Tamsin: I'll try to make a better example in a few days' time. I am glad that you point things out, but I took initially took a little offence at the phrase "damn fool".

- >>Bill: Actually, I said 'dumb fool'. This is a family site. ;-) And I don't get the impression that you thought that Black's play was OK.

- >> Tamsin: No problem. Indeed, I did not think Black's play was okay. I was trying to show what would happen if Black played what seemed to be "natural" moves. I have seen the contentious B1 played on many occasions, and I agree it's second-rate. But it looks like an easy way to take a fair-sized chunk of territory (but that's only its superficial appearance). Anyway, I think in the best lines, playing four kakari should be even for White or at least no worse than any standard fuseki. Of course, achieving parity is the reasonable objective.

If Black plays an ikken tobi instead of B1, then White can just ignore it and take the lead in one of the other three corners. Another way it might develop would be like this:

Corner 1 Black plays ikken tobi, White plays tenuki

Corner 2 White plays a second approach, Black answers locally

Corner 3 White plays a second approach, Black replies locally

Corner 4 White plays a second approach, Black replies locally

Corner 1 White goes back and continues with her own ikken tobi.

Alternatively, suppose Black decides not to answer White's move locally.

Corner 1 Black plays ikken tobi, White plays tenuki

Corner 2 White plays a second approach, Black plays a second move in Corner 3

Corner 3 White ignores Black's move in Corner 3 and plays a second move at Corner 4

Corner 4 Black goes back to Corner 1.

If play went like that last, then Black would be leading in Corner 1 and 3, but White will have effectively played first in Corners 2 and 4. The point is simply that by playing tenuki at the appropriate time, White need not be behind even though she originally played second in all four corners.

And why play four kakari instead of two empty corners? Well, one reason could be to hinder the formation of a San Ren Sei or Chinese Fuseki. Another reason could be to initiate early fighting, if that suits your taste.

19x19 diagram  

Admittedly, it is hard to see Black being as persistent as to play 5 in either of the indicated places, but here we have arrived at an interesting situation. As I see it, White is now ahead, because she has effectively played first in three corners, while Black has spent three extra moves to make certain of taking one corner.

- >Bill: And that shows something besides the inferiority of B1 - B5?????

If you perform a kind of tewari analysis, Black has effectively played something like this:

19x19 diagram  

Even though White's initial plays in each corner are unusual, they are still perfectly decent moves, and they cover a much large area than Black's moves. Therefore, I cannot help but prefer White here.

I am still investigating the four-kakari way of playing, but I strongly suspect that White can always achieve parity if both sides know what they are doing, and can often gain a lead if Black plays naively and conventionally.

A broad knowledge of tenuki joseki would be very helpful in this style of play.

Variation 2  

Bill: B3 might be better at 'a', but this is one way to take sente and play B5.

Variation 2 (continued)  

Bill: W6 looks urgent. It prevents a Black pincer and aims at the aji in Black's top left group. Now B7 looks big. White will seek a chance to play at or around 'a' and make use of the white+circle stones.

Tamsin: Many thanks for this example, Bill. How do you assess the situation? As far as I can see, it looks even, but difficult.

- >Bill: I like Black, but not by much. I think that generally Black should reply to the high approach. The double kakari is severe.

To be sure, playing four kakari is more a style of play than an actual fuseki pattern in itself, as there are so many ways in which it can turn out, and of course, the high one point approach is not the only kakari in the world. I originally used to play like this when I was feeling angry or frustrated, and of course I used to get beaten. Now, though, I have started to look for reasons to play in this style, and so far I've found it to be very rather effective.

Variation 2a  

Bill: White may also go for the moyo on the right. B4 is big, though.

Popular on the Internet  

Alex: For some reason, this 1-3-5 pattern for Black seems somewhat trendy among low kyus/low dans on IGS/KGS these days. I've had it played against me a number of times. Personally, I don't like it, because W6 seems like the obvious move to me, and my feeling is that any pattern that makes your opponent's moves easy to find is unlikely to be good. The obvious counter-argument is that the 1-3-5 pattern of the Orthodox fuseki also forces W6 to wedge, but still... I'm not a big fan of this.

The usual moves I see for B7 are something like a, b or c here. Is there anyone who's played this opening as Black and can comment on the thinking behind it? I suspect it's just intended to generate complicated, fighting-oriented games of the sort most Internet players prefer, since (in my experience) they're usually weak at fuseki and endgame and would like to start the middlegame fighting ASAP and keep it going as long as possible.

Bill: For B7 I kind of like 'd'. It seems balanced to me.

Alex: I think d is good too, treating the left and right sides sort of like miai. Of course, the people who play this fuseki as Black usually play it without regards to White's first couple of moves, so W2 and W4 won't always be as shown.

Tamsin: I'd try B7 at the point above 'd' (high kakari). If Black's going to play at tengen, then maybe he should continue in the high, influence-oriented style. I have also met this pattern several times. Usually, if White plays a wedge, then Black tries to attack White's group. An unreasonable-sounding plan, but not all that unreasonble at amateur level maybe...

The Tent  

Anonymous: I've had this played against me. Found it difficult to counter effectively. Any ideas ?

MrTenuki: The comments posted on the Stanley Standard Fuseki page would also apply here: just take the remaining corner, because B1+B3+B5 is overconcentrated.

What do you think about these 2 fusekis? [ext] http://gosensations.com/?id=2&server_id=5&new_id=226 It's a handmade ponnuki in the center

[ext] http://gosensations.com/?id=2&server_id=6&new_id=229 Handmade ponnuki is powerful, so white decided to destroy it :) Might the 3-3 invasion have any chance of living in that diagram? ~srn347


I played a game against another person who was about the same strength I am (20kyu) who made the foolish comment that he ha memorized all of the Fusekis so as white I played the following opening in the opening game which completely threw him off badly (I ended up beating him by 44.5 points) and I gave him the lecture afterward that he shouldn't spend so much time memorizing any positions as he should studying the game (shape; influence; territory etc.) and I did my best to explain the difference.

Making A Point  

The following two patterns were played, with some limited notoriety, by two individual amateur players.

Belgian pattern  
Romanian pattern  

Fuseki Experiments last edited by PJTraill on November 18, 2019 - 15:12
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Login / Prefs
Sensei's Library