Please share your wisdom.
I don’t see things I could have made differently tbh, unless AI review gives me huge spikes I don’t know what to make of it.
1 Like
Four things that can easily be improved:
a) Direction of play
b) Do not create too many groups
c) Have a plan for your stones.
d) Good shapes
Examples:
a+b) Direction of play and too many groups:
In the span of five moves, you’ve played in three different directions and created three new groups, which, of course cannot all be defended.
c) Have a plan for your stones:
Ask yourself: What did I want these stones to originally when I played them and what did they ended up doing?
d) Good shapes
Here is how you can get small advantages. In this case you both played moves that were not optimal and then went to play elsewhere. However this is an easy way to create sub-par shapes.
Move 1 is sente and move 2 is also the typical response. Now Black is overconcentrated (C12 should have been at C11), your corner is secured and you still have sente.
Similarly, this should have been at j3:
If it was, it would have made your later fight much easier and it would really support your n3 stone. E.g. if the opponent chose similar moves:
Now it was Black that had to fight to live.
3 Likes
I feel that last diagram is a bit wishful thinking to white’s advantage (katteyomi).
To me that feels as if black cooperates to fall into white’s trap.
A black player that is more aware of white’s intentions would not cooperate like that, and instead resist by something like this to make white’s life a lot harder to settle that N3 group:
I think the root cause for white getting in some trouble here is move 6 at N3:
This move is too close to black’s strong group in the lower right, giving black a target to attack. So it’s an immediate liability for white.
A beter idea would have been for white to stay further away from black’s strong group, for example L3, J3 or even just F3.
4 Likes
I play it with the mind that I shouldn’t let the opponent get both sides of the corner. So the proper response is to play on that side, but further away?
3 Likes
Yeah K3 or J3 are better, or even just enclosing the lower left makes the whole bottom less valuable for black too…
3 Likes
I totally agree with you, which is why I had added the caveat “if the opponent chose similar moves” to keep the moves on what could be expected at those ranks. 
If we increase the level of play, a lot of those stones shouldn’t be there in the first place (n3 - which is one of the points I made about re-thinking the purpose of those moves) or they should have been re-inforced way earlier, by some actual base-making extention (e.g. k3 instead of e3) or jump before things got critical.
If I had to play a few moves after move 9, I’d probably do this:
C is bigger in terms of direction of play, but I think that re-inforcing that stone is a bit urgent. So, I’d probably play A.
I assume that you, at 3 dan, will have better ideas, in terms of evaluating this board and picking those next moves, however is we could realistically expect people of lower ranks to make those better evaluations, then they wouldn’t be in those ranks in the first place, therefore we have to adjust our feedback in what they and their opponents would probably play and what could be improved from there.
Improving the direction of play, the shape and having a plan for your stones, are all reasonable “next steps” which can get someone relatively fast in the next level (SDK in this case), from which they could then move to building/improving other skills.
If someone asked us for feedback on their driving we wouldn’t go “well, I’ll tell you what. A nice scandinavian flick would have gotten you much faster through that turn”. It would, but it is not realistic. Any feedback should be adjusted on the current improvable skills that the recipient has at the moment.
Incidentally that’s why I do not “train with AI”… I do not understand how the AI plays, so there is no point in me getting feedback from it, since the choices the AI makes require of you to have the strength and reasoning of the AI, which I do not.
1 Like
This is not a principle you should be following in the opening, in my opinion.
Remember: Corner > Sides > Center (as a general guideline)
So your opponent just played a corner enclosure and secures the corner territory with it. The move N3 aims to prevent Black from expanding to the side, but it has little effect on the corner.
So instead it would be good to play a corner enclosure for yourself, or approach / invade the 4-4 in the upper right.
Basically any move in one of the open corners is good.
4 Likes
OK I’m very embarrassed right now, because I always thought the point of taking “the other side” of an enclosure was to prevent the opponent from taking that corner.
I hadn’t thought that when an opponent goes for a corner, I should go for the corner that has no moves to it.
3 Likes
OK, but while you didn’t expect black to spot your 17 move sequence, you did expect white to spot it. Was that realistic at the level of these players?
I just offered an alternative that is more impartial, without being unrealistic as to what players of this level would be able to come up with in their own games. Also I tried to use that to point out the actual cause of this situation and how to avoid it. Not trying to deny that white is in some trouble here.
1 Like
But they are the same moves, I got the variation from the game itself. I didn’t think of it, at all.
Here are those “17 moves” from the actual game (some of them where played earlier, in the real game, but hey):
Comparatively the only thing that changed in my variation is that White’s stone was one move closer.
…and that just brought in a small improvement in the end result if the opponent chose the same response (trying to atari the stone at L2).
As I said, I totally agree with that
This technique of taking a sequence and then moving 1 stone to a different spot to discover some potential optimalisation is used in tewari analysis, but you have to be careful with ignoring the possibility that player B might also play differently when player A plays differently.
1 Like
Quite true, I get that wrong myself, quite often
Alternatively in your suggestion, if it was my own game I wouldn’t cooperate by playing n4 either and I’d just sacrifice that stone and look to these choices:
Assuming that the square marked stones were alive (which, sadly, they are not at that point), N3 retains some good aji, but it is probably not getting out of there alive, which is fine. It is probably a losing bet, but it is at least going to be a fun fight.
I am playing a game right now with someone at 14k, here was the board a few minutes ago:
Black tenukied early, so I got the corner. This time he chose to re-inforce his stones and now I got sente.
What is next now? Where is the biggest move? Taking an empty corner or fighting around A?
We do not have to prevent the opponent from taking any corners. If they have one and another one is open, then that is valid compensation. So, A, is smaller than all the choices in the empty board.
Now, B,C,D or E,F,G? 
Imagine you getting one and the opponent the other and decide which is best.
You play around B,C, D:
With some simple moves/joseki both players get a large framework, half and half the board seem divided and Black has sente.
You play around E, F, F:
Now things are not so simple and there are various choices on what White can do with sente, but the choice of large frameworks is not directly evident.
Both ideas are valid. Which are you more comfortable with, having a large framework or fighting? And then you choose.
This, however, is sub-optimal:
Black can ignore the 4-4 move (since by ignoring it, then it is like you played there first and he invaded the 3-3, for which there are easy known joseki) and then you started to “refute the opponent’s claim in one corner” and they got their stake on two other corners instead, as well as a joseki in the corner they had to defend. All you got back is sente and a wall facing Black’s stones on the top left. A sub-optimal result.
1 Like
Simple direction of play:
The opponent chose 3-3 this means that now both of his approachable corners have similar choices/joseki and the choice is totally based on direction and the rest of the board:
A or B?
If you play A, then both the square marked and the triangle marked stones are facing your potential wall and can provide places to extend/grow towards.
If you play B, then only the triangle marked stone provides such help and Black will probably get to expand towards his own stones at top left.
So, the choice is clear: A.
This is also the concept of creating a plan for your stones and what do you want them to do.
It doesn’t have to be complex, at all. There just has to be a plan that is based on some good principles (direction and shape).
1 Like
After black’s pincer at L3 on move 31, my feeling is that this could become a bit of a running battle between black’s L3 group and white’s N3 group, and direction of play becomes especially important.
So in my own game, I’d probably not even play white 1 from the direction where white is already strong (lower left) to provoke black’s L3 group to move out ahead of white’s N3 group.
I’d rather play from the direction where white is weak, like M4, trying to get into a situation where white’s N3 group moves out ahead of black’s L3 group.
KataGo is of course much more skilled than I am, so it comes up with much more elegant ways to help out white’s weak-ish N3 group, for example:
Still, that’s a similar direction of play as I’d have in mind.
3 Likes
No. The proper response is to not respond, and instead of thinking about destroying your opponent’s thing, build your own thing. They make an enclosure, you make an enclosure. This video of mine with early opening principles would have guided you to a better move here. https://youtu.be/vupa_IM1wWY?si=NUfErvxHGWg6s0j9
3 Likes
Duly noted and will dutifully watch
1 Like
Perhaps that is a slight misunderstanding of the admonition that a double-wing formation is too good to allow. This is an example of a double-wing formation (I marked the “wings”):
In the game, your opponent’s corner enclosure had 0 wings on move 6, so that admonition didn’t really apply yet.
Still, the K4 wing in that diagram is the superior wing of the two. Even without them having the Q10 wing in place, it’s good to deny the K4 wing when you have the opportunity.
But the general way to deny a wing is to play around that same location, so around K3.
This distance is good because even when they pincer you, there is enough space to make a base. That’s even the case when you approach their corner enclosure as closely as L3:
But when you approach as closely as M3 and they pincer, the available space is already a bit too cramped to make a nice base:
And when you approach as closely as N3 and they pincer you, there is too little space to make a base, so basically your only option is to start running:
And your opponent can look forward to harrasing that group which is not going to make many points. In fact your opponent will probably be making points while chasing that group.
2 Likes
As a follow-up from that, instead of playing move 6 around K3 as a hard denial of black’s best wing, I recommended F3.
The reason is that it makes a corner enclosure for white (which is basically always a good move) and it’s also a soft denial of black’s best wing, because if black plays at K3, F3 is just close enough to deny black an optimal base to settle K3 if white invades at some point:
So a more careful black player might hold back a bit by playing at L3 to ensure base potential with A, just in case (better safe than sorry):
But that means that this wing is already smaller and thus less valuable than black would like, so black might prefer to play some other opening move instead of L3. Either way, mission accomplished for white.
Edit:
A side note: I kept calling it a “wing” for consistency in my latest posts, but usually this is just called an “extension”. AFAIK the term “double-wing” is only used in the case of having wide extensions on both sides of a corner enclosure (or perhaps from a 4-4 corner). Double Wing Formation at Sensei's Library
3 Likes