Pros falling for a trick play?

judicata · #1

I was looking at the example below of what Guo Juan calls a "bad trick," which is what she calls trick plays that, if your opponent knows the trick, lead to a bad result for you. (For reference, "good tricks" are those that don't leave you in a terrible position even if your opponent responds appropriately, although it may not be optimal).

Here is the trick move:

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . 2 1 . . | $$ , . . . . O , X . . | $$ . . . . . . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

And the sequence black wants:

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . 7 4 9 . | $$ . . . . 0 8 X O 5 . | $$ , . . . b O 3 X 6 . | $$ . . . . . . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

Black can kill at 'a', so white takes, and black commences a painful attack at 'b'.

White's position is better after responding to :b6:

as follows:

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . . . . | $$ . . . . . 3 2 X 4 . | $$ . 5 . . . 1 X O O . | $$ , . . . . O O X X . | $$ . . . . . . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

I first saw this around 8k or so, and had trouble understanding how this is bad for black. I won't go into detail, but it was eye opening for me to compare it to the following joseki:

All of this makes sense to me. I'm confused because pros often play the "wrong" response. In fact, from the position in the second diagram, pros play the "wrong" move more often than the "correct" move (admittedly a small sample-- 20 times vs. 15 times in my database). I assume that, if black plays the hane (at Q17 in the diagrams above)--which is played very rarely--he has already determined that white's position after the appropriate sequence is acceptable. So I also infer that these are special cases.

But I can't see what's so special about most of these games that makes the "wrong" move okay. Of course, I very seriously doubt that this move is actually "wrong" in these games. But I'm not sure why.

Here are two samples in which the move was played by Yuki Satoshi and Kitani Minoru:

jts · #2

In the second game black's ideal extension was already taken, so I can see why Kitani might prefer to deny white the corner and try to handle the weak stones in the center.

In the first game, I have no idea what B's cunning plan was.

Interestingly, I had heard that :b6:

at q13 is the trick, and the way you show as a (situational) joseki.

judicata · #3

I don't know why I neglected to look this up in the latest joseki diction (the update to Ishida's). It indeed calls the diagram (what I call the correct response) as a joseki. The book also notes that the hane does not work when black responds with the knight's move instead of the one-space jump.

Joaz Banbeck · #4

If black has a winning ladder to the southwest, this might be fun too:

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . O X O . | $$ . . . . X X X O O . | $$ , . . . . O O X X . | $$ . . . . B . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

It continues like this:

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . O X O . | $$ . . . . X X X O O . | $$ , . . . 8 O O X X . | $$ . . . . B . 3 5 6 . | $$ . . . . . 2 1 X 7 . | $$ . . . . . . 4 9 . . | $$ . . . . . . . 0 . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

...then like this...

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . O X O . | $$ . . . . X X X O O . | $$ , . . . X O O X X . | $$ . . . . X . O O X 5 | $$ . . . . . X O . O . | $$ . . . . . 2 X O . . | $$ . . . . . 7 1 X 4 . | $$ . . . . . . 6 3 . . | $$ , . . . . . , 8 . . | $$ . . . . . . . . . . |[/go]

...or this...

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . O X O . | $$ . . . . X X X O O . | $$ , . . . X O O X X . | $$ . . . . X . O O X 3 | $$ . . . . . X O . O . | $$ . . . . . 2 X O . . | $$ . . . . . . 1 X . . | $$ . . . . . . 4 . . . | $$ , . . . . . , . . . |[/go]

....or this...

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . O X O . | $$ . . . . X X X O O . | $$ , . . . X O O X X . | $$ . . . . X . O O X 7 | $$ . . . . . X O . O . | $$ . . . . . 2 X O . . | $$ . . . . . . 1 X 3 . | $$ . . . . . 5 . 4 6 . | $$ , . . . . . , . . . |[/go]

daal · #5

Here's your second diagram. I don't see the killing sequence if white plays b.

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . a . . | $$ . . . . . . 7 4 9 . | $$ . . . . 0 8 X O 5 . | $$ , . . . b O 3 X 6 . | $$ . . . . . . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

This is what I see happening:

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . . 2 3 . | $$ . . . . . . O X O . | $$ . . . . X X X O O 4 | $$ , . . . 1 O O X X . | $$ . . . . . . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

Click Here To Show Diagram Code: [go]$$Wc $$---------------------+ $$ . . . . . . 5 . O . | $$ . . . . . . O . O 6 | $$ . . . 7 X X X O O X | $$ , . . . O O O X X . | $$ . . . . . . . . . . | $$ . . . . . . . X . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ . . . . . . . . . . | $$ , . . . . . , . . . |[/go]

What am I missing?

tapir · #6

daal wrote:

Here's your second diagram. I don't see the killing sequence if white plays b.

What am I missing?

Continue reading after one pushing move, just imagine Black answers at N17 for now. At one point White will have to settle the corner and then Black can turn up and the White center stones will get very weak with Black taking profit on both sides or else White has to accept that the corner dies. 2 weak groups = bad. In fact, it is probably the best that can happen for Black if White greedily lives in the corner soon (after only one or two pushes) and then tries to rescue the center stones as well, because they will run forever.

In the game of Yuki Satoshi Black was pushing many times first and then settling only by sacrificing a pincer stone on the left side. And to my positional judgement it looked like White with territory and solid groups with a weak black formation still on board was already pretty comfortable by turn 60 and indeed continued to win by a margin.

Pros falling for a trick play?

Who is online