Page 1 of 1

DDK Review Request!

Posted: Sun Jan 15, 2017 1:19 pm
by TofuPython
I just started playing on KGS, and I managed to beat an 11k! I think I played pretty well, except I REALLY dropped the ball w/ that monkey jump invasion. I played black. Could someone give me a few tips?

Posted: Sun Jan 15, 2017 1:47 pm
by EdLee

Posted: Sun Jan 15, 2017 1:49 pm
by EdLee
Hi Tofu,

:b19: Did you read if W pushes and cuts at R3 ?
What's your plan then ?

Just o4 seems OK.

:b21: Jealous move that doesn't really do anything:
W simply connects and keeps all his cash here.
Better to eye the left side: wide open.

:white: 116 Bad habit. Just connect is better.

Re: DDK Review Request!

Posted: Mon Jan 23, 2017 7:37 am
by Rémi

Re: DDK Review Request!

Posted: Tue Mar 07, 2017 5:34 am
by Pio2001
TofuPython wrote:I just started playing on KGS, and I managed to beat an 11k! I think I played pretty well, except I REALLY dropped the ball w/ that monkey jump invasion. I played black. Could someone give me a few tips?


Hi,
Here are some standard ways to stop a monkey jump according to the position of the stone that is jumped over : forum/viewtopic.php?f=11&t=13636&p=210928#p210928

Once your opponent is pushing on the first line, instead of following, try to stop him with a net (play one intersection ahead). Read the sequence, and if it doesn't work, try a net two intersections ahead. Being two intersections ahead should be enough to stop anything.

Re:

Posted: Tue Mar 07, 2017 12:38 pm
by BlindGroup
EdLee wrote: :white: 116 Bad habit. Just connect is better.


Ed, I've seen you make this comment before. Would you mind explaining why just connecting is the better move?

Re: DDK Review Request!

Posted: Tue Mar 07, 2017 2:13 pm
by jeromie
The descent - white takes gote and ends up with 5 points. Black has 7 points.

Click Here To Show Diagram Code
[go]$$W
$$ X X X X X X X
$$ O O O X . . .
$$ . . 3 O X . .
$$ . . . 1 2 . .
$$ +------------+[/go]


The connect. White still takes gote, but play stops here for a while:
Click Here To Show Diagram Code
[go]$$W
$$ X X X X X X X
$$ O O O X . . .
$$ . . 1 O X . .
$$ . . . . . . .
$$ +------------+[/go]


Later...

If black gets to follow up - white has sente and four points; black has 7 points. This is a one point loss for white from the initial diagram:
Click Here To Show Diagram Code
[go]$$B
$$ X X X X X X X
$$ O O O X . . .
$$ . . O O X . .
$$ . . 2 1 3 . .
$$ +------------+[/go]



If white gets to follow up - White has sente and 5 points, black has 5 points. This is two points better for white than the initial diagram:
Click Here To Show Diagram Code
[go]$$W
$$ X X X X X X X
$$ O O O X . . .
$$ . . O O X 4 .
$$ . . . 3 1 2 .
$$ +------------+[/go]


If there's a 50% chance white gets to make the follow up, it means that on average connecting is .5 point better than just descending. But since white comes away with sente in either continuation, it's really even better for white than that. If counting, you'd count the sente follow up as part of the original move, so you could say the connect is two points better than just descending. It doesn't sound like much, but if you make the correct endgame move consistently it adds up quickly.

(If the surrounding stones are a bit different, there may also be aji for white to clamp instead of just hane in the endgame, in which case the situation skews even further in white's favor.)

Posted: Tue Mar 07, 2017 2:45 pm
by EdLee
Hi BlindGroup,

Supplement to Jeromie's notes:

Variation 1 (real game): W has forced :b2: to be sente ;
W has forced gote on himself:
Click Here To Show Diagram Code
[go]$$W
$$ ? O . X X . . ?
$$ ? . O O X . . ?
$$ ? . . 3 O X . ?
$$ ? . . . 1 2 . ?
$$ ---------------[/go]

Variation 2: W connects solidly. If :b2: still drops like the real game,
now W has a choice whether to reply at (b) and
revert to the game shape, or tenuki:
in this game, :b2: was gote, so W would ignore it:
Click Here To Show Diagram Code
[go]$$W
$$ ? O . X X . . ?
$$ ? . O O X . . ?
$$ ? . . 1 O X . ?
$$ ? . . . b 2 . ?
$$ ---------------[/go]

Variation 3: Re: aji mentioned by Jeromie:
In some cases, B may not drop to (x) immediately
and help W fix his shape: instead, B may wait for a chance
later to exploit W's local weakness(es): example: (d).
Click Here To Show Diagram Code
[go]$$W
$$ ? O . X X . . ?
$$ ? . O O X . . ?
$$ ? . . . O X . ?
$$ ? . d . 1 x . ?
$$ ---------------[/go]

In reply to :bc: hane, W has these options:
Click Here To Show Diagram Code
[go]$$W
$$ ? O . X X . . ?
$$ ? . O O X . . ?
$$ ? . e a O B . ?
$$ ? . d c b x . ?
$$ ---------------[/go]

(a) solid connect
(b) drop
(c) 1st line tiger's mouth
(d) 1st line jump
(e) 2nd line tiger's mouth
(x) special cases: 1st line hane
(o) other local replies
(t) tenuki

The correct reply depends on the exact, global board situation ( which is a truism :) ).
However, if we ( not only beginners ) develop a habit to favor one of these local replies without a correct global assessment, then it can become a problem.
( Posts 7 & 8 got squeezed into a few words in post 3 and other game notes. :) )

Re: DDK Review Request!

Posted: Tue Mar 07, 2017 3:08 pm
by BlindGroup
Thanks for the clarification.

Re: DDK Review Request!

Posted: Tue Mar 07, 2017 3:17 pm
by Bill Spight
You might enjoy this problem that I composed back in '01. :)

Click Here To Show Diagram Code
[go]$$ White to play. No komi.
$$ -----------------------
$$ | . . . . . . . . . . . |
$$ | . . . . . . . . . . . |
$$ | . . O O O X X X X . . |
$$ | . . O . O . . . X . . |
$$ | . . O . O X X . X . . |
$$ | . . O . X X . X . X . |
$$ | . . O . . O O X . . . |
$$ | . . O O O O X X X X X |
$$ | . . O . . X X . X O X |
$$ | . . . O O O X X O . O |
$$ | . . . . . O O X X O . |
$$ -----------------------[/go]


Maybe not so easy, but not so hard, either.

(I think I have posted it here before, but I couldn't find it. :-| )

Posted: Tue Mar 07, 2017 3:19 pm
by EdLee
Hi Bill, so much work ! :study: Thanks. :)