My games to analyse

[goame]

Please help me to analyse my games.
I haven't played since some years.
In this game I had black. Looks like I have done some suboptimal moves in the opening and overlooked some things.
What can be improved?

[jlt]

1) I think you should play against a stronger AI (possibly taking handicap), this one makes too many stupid moves.

2) Move 7: it is almost always better to respond to the 3-3 invasion than to the corner approach. So play either Q17 or R16.

3) Move 11, etc.: same remark.

4) at R2 is the appropriate shape (big corner + remove your opponent's base).

5) see remark 2.

6) not necessary, you have enough liberties. But even if you are afraid, jumping to D2 would be better (increase your liberties and threaten the corner).

7) would be better at N17: separate two groups so that you can attack L17.

8) useless. You can take the stones at any time.

[goame]

@jlt
Yes you are right.
Thank you very much.

[goame]

Game 2
What are my mistakes?
Maybe I should have focused more on defending the right corner?

[jlt]

If you want to defend K3, is better at O4, your three bottom stones would form a good shape.

is small. A corner enclosure at S6 (edit: I meant R6) looks bigger.

at R7 is the shape.

at R18 is the shape, if you want to defend your corner.

is not needed.

doesn't accomplish much.

doesn't accomplish much. Are you trying to make territory around F13? The area is very open, your opponent can play around F15 for instance to reduce it.

Move 121: looks unnecessary, I think White is already dead.

[EricBackus]

is small. A corner enclosure at S6 looks bigger.

Of course you probably meant "R6 looks bigger".

[skydyr]

I'd recommend that you play a stronger bot again. Keep playing a stronger one until you start losing more than 50% of the time at least. You want an opponent strong enough that you can be challenged and learn from their play, while the ones you are playing seem to make a lot of nonsensical moves. If your opponent isn't pointing out your weaknesses through play, it's harder to improve.

[goame]

@jlt
Thank you very much.

@EricBackus
Yes I also thought R6 is the move he meant.

@skydyr
That's what I'm doing, every time a little stronger.
Maybe 13 games left until 1 dan.
But I have fixed now some bad moves and ideas.

[goame]

Game 3
It was a little funny.
But somehow I have done some bad blunders or am I wrong?

[jlt]

The AI is still too weak, in particular too passive, so doesn't help much to detect your weaknesses, but here are a few ideas.

is off-topic. M15 is a weak group, you have to continue attacking it, for instance hane at M14, or protect the side at Q14.

generally you want to avoid empty triangles. If your goal was to reduce White's territory, then jumping at O8 looks better.

bad shape, already seen in previous games. J17 is the shape.

[Mike Novack]

I'd recommend that you play a stronger bot again. Keep playing a stronger one until you start losing more than 50% of the time at least. You want an opponent strong enough that you can be challenged and learn from their play, while the ones you are playing seem to make a lot of nonsensical moves. If your opponent isn't pointing out your weaknesses through play, it's harder to improve.

ABSOLUTELY. If trying to learn from playing a bot, you do NOT want to be playing a bot at your level. You want to be playing a bot significantly stronger than you are so that it will not be making the same level istakes that you are and will be able to punish the mistakes you make.

I would change the recommendation somewhat. If you insist on playing even games against the bot, you want the bot enough stronger that you lose say 85-90% of the time. If that would be too discouraging to you, want to be winning 50% of the time, take say three handicap stones and set the bot strong enough for that result. Three stones will not unduly distort the game, especially if you immediately challenge in the corner that the bot plays. This would probably give you the fastest improvement.

[goame]

@jlt

I thought about move 35 to be played on M14 or Q14 but I thought G13 would be better to build a big mojo.
I thought Q14 isn't a big benefit for me or for white and M14 I wasn't sure if white can life and if I still can build the mojo with G13.

65 Why I should generaly avoid empty triangles?
Yes the goal was to reduce white's territory.

95 Yes J17 is clearly better.

[goame]

Game 4




Game 5


Game 6

[skydyr]

ABSOLUTELY. If trying to learn from playing a bot, you do NOT want to be playing a bot at your level. You want to be playing a bot significantly stronger than you are so that it will not be making the same level istakes that you are and will be able to punish the mistakes you make.

I would change the recommendation somewhat. If you insist on playing even games against the bot, you want the bot enough stronger that you lose say 85-90% of the time. If that would be too discouraging to you, want to be winning 50% of the time, take say three handicap stones and set the bot strong enough for that result. Three stones will not unduly distort the game, especially if you immediately challenge in the corner that the bot plays. This would probably give you the fastest improvement.

I completely agree. Don't jump by just a one rank difference either. Try playing against an 8k bot, and if you win or get close, play against a 5k one, or stronger. The worst thing to do is to play against a weak one because you repeat what you see. If you see a bot playing weird sequences, or not invading where it clearly can and should, you will assume that that's how the game is to be played and it will hold you back.

[Bill Spight]

Let me second the notion of not playing against weak bots if your object is to improve.

Taking appropriate handicaps is a good idea, and find a bot that can give you 3 stones is also good. That way you can practice play in the open corner.

Another idea when playing a stronger bot is to play take-back go. After the bot has replied to your play, if it indicates that your play was a mistake, take it back and try again. That way you get immediate feedback about your play, and the bot's play indicates how to refute your mistake. You also get the chance to correct your mistake right away. That is important for getting better.

Good luck!