Josekis? Nosekis?

[Akura]

Hey guys,

can you look at the first 50 moves of the following game and tell me whether the top and bottom right corner sequences were josekis and where the mistakes were? Of course you can look at the rest, too, but it is nothing to speak of.

I was Black.

[skydyr]

Looks like black got swindled in both cases. In the top right, I don't think black should push in... it just removes a liberty from black. Maybe hane on top of the two white stones and let them run if white wants? I'm not entirely sure about how to handle the bottom right, but it looks like black should be able to cut white and fight.

[mitsun]

In the upper right, B did fine through move 27. However, once B is secure on the outside, W must come back to add another defensive move. See if you can find a good attack for B in the game position. (B should be able to capture four stones in sente or play ko to kill the entire corner, depending on how W responds.)

In the lower right, I would not worry about joseki, but about direction of play, which is more important. At move 38, W invaded deep into a very strong B position. B should be able to profit by attacking this stone.

In the game, B used thickness to make territory directly, but then W was able to make compensating territory in the other direction, and overall B gain relatively little. Try counting the W and B positions after move 50, then look back to the position at move 38. It feels like B is entitled to more than he actually got. Another way to see this is to note that the B territory near Q12 is over-concentrated.

The right strategy for B is to attack from the left, pushing W toward the B thickness above. B will then profit by making new territory along the bottom side. As W runs, he will make little or no territory of his own. True, some of the potential territory near the B thickness will disappear, but the gains on the other side will more than compensate for this. Try playing out some simple sequences starting with move 38 at O4 or P5. Count the resulting territories for both sides and see how well B does, compared to the previous result.

[Magicwand]

175 Ko threat was terrorable.
you should almost never play such ko threat.

[Akura]

Thank you all for your replies.

@skydyr: In the top right corner I think I confused the situation with a joseki I once saw. Therefore the push-in.

@mitsun: During the game I thought that in the following diagram is definitly sente:

$$Bc
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . X c . b . |
$$ | . . . . . . . . . . . . . O X O O a . |
$$ | . . . . . . . . . . . . . O X X O . O |
$$ | . . . O . . . . . O . . . O X X X W . |
$$ | . . . . . . . . . . . O O X X W W W X |
$$ | . . . . . . . . . . . . X X O X X X 1 |
$$ | . . . . . . . . . . . . . . O . . . . |
$$ | . . . . . . . . . . . . . X . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+

Click Here To Show Diagram Code

[go]$$Bc
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . X c . b . |
$$ | . . . . . . . . . . . . . O X O O a . |
$$ | . . . . . . . . . . . . . O X X O . O |
$$ | . . . O . . . . . O . . . O X X X W . |
$$ | . . . . . . . . . . . O O X X W W W X |
$$ | . . . . . . . . . . . . X X O X X X 1 |
$$ | . . . . . . . . . . . . . . O . . . . |
$$ | . . . . . . . . . . . . . X . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . X . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . O . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]

But now I think that if White answers at 'a' or 'b', Black can catch the marked stones in sente. And if White answers at 'c', we get a ko for the whole corner.

At the bottom right you are right. I tried too fast to convert my influence to territory.

@Magicwand: Why shouldn't I play such ko threats? Because it eliminates further threats to the corner?

[Magicwand]

@Magicwand: Why shouldn't I play such ko threats? Because it eliminates further threats to the corner?

if you see professionals using ko threat you will see that they alomst will never play ko threat that they will lose points..
you used ko threat to give his group life that he didnt have before.

learn basic life and death in the corner.
memorize them and you will be surprised how often they show up.

[Akura]

This makes sense. Thank you!

[lightvector]

If you're not careful in choosing your threats, it's actually surprisingly easy to lose enough points to make winning the ko not worth it overall.

Say you have a ko to kill a medium sized group where the difference between winning and losing is a whole 40 points (you gain 40 if you win, you gain 0 if the opponent wins). Say that a regular move gains each player about 10 points, which is actually probably a low estimate for many midgames. And say that you have have several ko threats that lose 3-4 points each, and not enough non-losing ko threats. Note that even something as small as making a future first-line endgame hane no longer sente for yourself is already a loss of 3 points.

Scenario 1:
You fight the ko and in total play 4 ko threats each worth -3 points after the opponent answers. (-12)
Your opponent finally runs out of ko threats and plays a regular move. (-10)
You win the ko and kill the opponent's group. (+40)
And it's your opponent's turn next.
Result: +18

Scenario 2:
You stop fighting the ko when your non-losing threats run out and play a regular move. (+10)
Your opponent wins the ko and his group lives. (+0)
You play another regular move. (+10)
And it's your opponent's turn next.
Result: +20

[Magicwand]

If you're not careful in choosing your threats, it's actually surprisingly easy to lose enough points to make winning the ko not worth it overall.

Say you have a ko to kill a medium sized group where the difference between winning and losing is a whole 40 points (you gain 40 if you win, you gain 0 if the opponent wins). Say that a regular move gains each player about 10 points, which is actually probably a low estimate for many midgames. And say that you have have several ko threats that lose 3-4 points each, and not enough non-losing ko threats. Note that even something as small as making a future first-line endgame hane no longer sente for yourself is already a loss of 3 points.

Scenario 1:
You fight the ko and in total play 4 ko threats each worth -3 points after the opponent answers. (-12)
Your opponent finally runs out of ko threats and plays a regular move. (-10)
You win the ko and kill the opponent's group. (+40)
And it's your opponent's turn next.
Result: +18

Scenario 2:
You stop fighting the ko when your non-losing threats run out and play a regular move. (+10)
Your opponent wins the ko and his group lives. (+0)
You play another regular move. (+10)
And it's your opponent's turn next.
Result: +20

NO NO NO and NO
if you use -3 point ko threat..your opponent only need 37 after ko threat and 34 point ko threat 31 and 28..
which means his ko threat will increase... after few ko fight you alread lose that ko without fighting.
that is why you have to calculate all ko threat before starting it.

[illluck]

If you're not careful in choosing your threats, it's actually surprisingly easy to lose enough points to make winning the ko not worth it overall.

Say you have a ko to kill a medium sized group where the difference between winning and losing is a whole 40 points (you gain 40 if you win, you gain 0 if the opponent wins). Say that a regular move gains each player about 10 points, which is actually probably a low estimate for many midgames. And say that you have have several ko threats that lose 3-4 points each, and not enough non-losing ko threats. Note that even something as small as making a future first-line endgame hane no longer sente for yourself is already a loss of 3 points.

Scenario 1:
You fight the ko and in total play 4 ko threats each worth -3 points after the opponent answers. (-12)
Your opponent finally runs out of ko threats and plays a regular move. (-10)
You win the ko and kill the opponent's group. (+40)
And it's your opponent's turn next.
Result: +18

Scenario 2:
You stop fighting the ko when your non-losing threats run out and play a regular move. (+10)
Your opponent wins the ko and his group lives. (+0)
You play another regular move. (+10)
And it's your opponent's turn next.
Result: +20

NO NO NO and NO
if you use -3 point ko threat..your opponent only need 37 after ko threat and 34 point ko threat 31 and 28..
which means his ko threat will increase... after few ko fight you alread lose that ko without fighting.
that is why you have to calculate all ko threat before starting it.

I didn't know either reason before and just knew that you should avoid point-losing threats unless you must, so thanks a lot for the explanations! I think, though, that it's probably the combination of the two reasons (or it's really the same one, except expressed in different ways).

[Akura]

I obviously have no idea how to fight a ko. Thanks you, Magicwand and lightvector, but I think I need some time to digest what you have said.

@lightvector: Do you say that this loses 3 points when played as a ko threat?

$$B
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . . 1 . .
$$ -----------

Click Here To Show Diagram Code

[go]$$B
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . . 1 . .
$$ -----------[/go]

[skydyr]

I obviously have no idea how to fight a ko. Thanks you, Magicwand and lightvector, but I think I need some time to digest what you have said.

@lightvector: Do you say that this loses 3 points when played as a ko threat?

$$B
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . . 1 . .
$$ -----------

Click Here To Show Diagram Code

[go]$$B
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . X O . .
$$ . . . 1 . .
$$ -----------[/go]

He's saying that if you play a ko threat that turns into a gote move, it loses points.