But E4 is the only fault in white’s formation and I really, really want it dead.
So, for most of my recent games I play F4 inviting a fight, because I currently don’t see any immediate problem with doing that.
This is my idea of studying joseki, if the move is wrong my opponent will show me how it’s wrong. But I’ve been doing it for quite a while and yet it always ended up pretty favorable for me.
Most of the time my opponents would tenuki - which is definitely not what I would do tbh, the stone is too big - and soon after I can fetch a sente to capture that stone and get a neat fortress facing the center. The game would then be lopsided in my favor imo.
If I were black I’ll probably do something like this:
Not knowing the joseki follow-ups, it looks playable for both to me to cut there.
It just seems that it will lead to complex fighting with multiple groups.
I would be comfortable playing this as either side against other 5-6d OGS players, personally, and just expect there to be fighting/running and potential complications from the fighting.
But perhaps someone else has a clearer idea of the aji involved.
Checking with the OGS site supporter AI, it evaluates the board within 0-1 or so points of score difference (after the cut and along various sequences afterwards) too.
If you’re more comfortable playing this sort of game + style, then it could be to your advantage too, to choose something which will develop that way. (especially within such a small margin of difference)
If there’s a problem it’s probably just that white is very heavy in variation shown.
It’s up to black when to start this fight also, same with when white plays g3 instead.
They’re not that dissimilar, white needs add another move to fully resolve, and it’s whether that move is worth it, and how natural it is to find a move that fixes the aji and gains some territory would probably be a factor.
You can forget about that. White needs two moves to capture the black cutting stone. It only dies if Black allows it. In that case, which two moves would you like?
Black does not have to, but may move out the cutting stone. If he does, you might be forced to play again so that F3 does not become another cutting stone.
1 and 3 are like playing out a ladder that doesn’t work. This is normally a blunder, but here I suppose it works because black becomes heavy.
In principle, you do not want to push against the cutting stones or middle group from both sides. For the same reason, you might keep your distance with A instead of B in the following position.
Just want to say that white 4 should take the stone instead of extending. Extending will just give black more sente like 5 and H5 later.
There are some situations where F4 is better so I wouldn’t really consider it a “deviation”. However, if your main reason for playing F4 is that you really want E4 dead, then probably you should think again. When you want a group dead, a better way would be to attack it indirectly rather than directly. In this case F4 is a helping move that helps E4 become alive rather than killing it.
It’s not an outright mistake, but it is unusual and leaves different and more severe aji if/when black runs out the stone. Tenuki is a fine answer (my default), because if black plays some move on the side and then white feels obliged to spend a move to capture the stone to suppress that aji black is happy to make that exchange because white’s a bit overconcentrated.
g3 is following the general principle of when you have 2 cut groups, you don’t want to play close to the cutting stone that separates them because when it runs (and doesn’t die, it’s different if you can capture/net it) it ends up hurting both sides, like squeezing out the toothpaste.
Thanks Chinitsu for starting an interesting conversation! My first instinct was that F4 (in the diagram at the top of this thread), encouraging black to pull out of atari at E5, will weaken the white stones on the left. But it also makes the bottom white group stronger. So maybe it’s a fair trade.
I got the toolbox out: Kombilo and KataGo. The database search gave me 687 hits for the usual joseki (G3 instead of F4), and 67 hits for this variation (with F4). So the pros have a clear preference, but are happy to try both variations. Maybe F4 shouldn’t be called a “deviation”, but just another joseki?
KataGo (with 4-4 stones in the other corners) has a mild preference for the usual joseki. The difference is about half a point. In the F4 variation, KataGo prefers to pull the black stone out of atari (and indeed that’s what the pros do more than half the time), but tenuki is only about 0.2 point worse. If black does tenuki, then white should capture E4 in a loose net rather than just taking it off the board, according to KataGo (I’m sure that Kageyama would disagree!) Of course, results may vary if the other corners look different.
Summary: yes, all of these variations are playable.
I think in this day and age there shouldn’t be anything that is considered a “deviation”. The term “deviation” sounds negative and people would tend to avoid it. Everything is playable!
Looking at the etymology of deviate, it stems from “off the road” and interpreting a joseki as the road seems pretty reasonable to me.
Not everything is joseki. Deviating from a joseki is possible.
So unless you are saying we should treat “deviation” like the n-word, because of how “deviant behavior” was (or is) seen as sinful, I dont understand where you are coming from. If you are, discussing that could be interesting too, but it should be done in a separate thread, I’d say.
I think that there is a risk in having a decent amount of joseki knowledge and then judging moves that are not in your personal joseki dictionary as bad moves that should be actively punished immediately [¹].
But in many cases the “deviant” move is just fine or only slightly suboptimal and there is no real punishment other than continueing to play normally. If instead you immediately try to prove that your opponent is a bad player because he doesn’t know the “proper” joseki moves, chances are that you will be the one actually overplaying, and you may very well get a bad result.
[¹] I think this tendency is fairly common for middle to strong SDK level players. It may be part of the reason why there is a proverb that “learning joseki loses two stones in strength”. I suppose another part is that relying too much on joseki knowledge may promote tunnel vision and inhibit flexibility and creativity.
P.S. A psychological tip for in-person games when you’ve tenukid from moves like f4 which leave some bad aji of pulling out the stone: keep glancing menacingly at the area to intimidate your opponent into defending there, to save you the trouble of actually deciding when to pull out the stone and start a fight.
The funny thing is I was doing a quiz to decide whether the word is positive or negative and “Deviation” was negative. Granted there wasn’t a “neutral” option.
I think the whole concept of “joseki” needs to be revisited in this day and age. Last time we didn’t have AI to evaluate the move for us, so we followed whatever the pros played as “joseki”, which is basically a set of “correct moves”. This brings us to this point mentioned by gennan:
And I did this a lot last time.
Nowadays even if the move is different from the joseki, we can see that it doesn’t really matter too much and the game is playable. Hence, the idea that the joseki is a set of correct moves no longer stands. And if there’s no joseki, there’s no deviation either.
This is also why I don’t think the joseki and opening problems that we used to do last time are valid anymore. We shouldn’t be limited by joseki or opening theory. Everything is playable
I can’t really blame the player putting in practice his knowledge somewhere. So for me at the moment someone departed from the joseki or tenukied a bit early, i tried to get advantage from it. Naturally it was exagerating. I didn’t understand that the disavantage could be very small, like just one forcing move more or a ko threat less. Not something to put in action at this stage of the game anyway.
They are. I vastly enjoy the patience of strong players to wait and use the aji resulting from a not so joseki sequence with a good timing.
Timing is one of the most complex concept so if by my very limited knowledge i have, i can still appreciate it (even just watching) the josekis are still very valid.
