First game for a long time. I'd like some tips how to improve. I made some self-commentary on that SGF file.

Move 23 (s7) seems like a B overplay. The fight looks good for W after pushing and cutting. However, W messed up with the exchange moves 32-35. This sequence removes all of the aji around here and severely reduces the value of the W cutting stones. Were you expecting B to prevent the connection underneath and get into trouble? B played a good block at O4, correctly emphasizing the important center fight rather than a few points of side territory.

Move 48 should just jump out to N13. B will still split the W groups, so you will be in for a difficult fight, but attaching directly to the B stones makes the split more painful for W. The exchange 52-53 helps B more than W. Your suggestion (L18) would be better, but I think best would be to just jump out patiently to M12. Move 54 (n12) is too thin, leaving a disastrous cut, which more or less ends the game. M12 still seems necessary and would leave the game up for grabs.

Just in the fuseki, I don't really like O3; I like playing on top better, I think. Then I think black should answer O3 at Q3. When you invade that moyo, I think Q3 may be a good way.

On the topic of top, I'm not sold on N15 jump. Black chose to pincer instead of defend his corner, so my inclination is to go after/aim at the corner. I think Q17 is not normally the way to defend there, either.

_________________
Revisiting Go - Study Journal
My Programming Blog - About the evolution of my go bot.

Agreed, the 2 standard ways to invade that shape of hoshi plus large knight plus big extension are either 3-3, which lives in the corner in sente and gives black a strong wall, or q3 attach which gives black the choice of whether to let you go to the corner or side. But it's too early to go into that small space now, the top is indeed more important at that time. If black gets sente to play p3 that's a nice shape for him, but you getting 2 moves elsewhere at this stage should be fine, and you can still invade at h3.

Thanks for responses. 32-25 sequence was bad. I tried to reduce black group's liberties, but I didn't realize where the direction of play was. I should have played on up side and instead let black connect underneath.

On top side the major problem was that I had two weak groups and the family feud that was caused by trying to defend one of them with move 48.

A few comments.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

A minor life&death detail: In the lower left, from move 193 onward, I think B3 kills. It was later played as a ko threat, but it just works outright.

At move 210, the hanging connection lives unconditionally. But it gives up 3 stones, so maybe in a desperate situation, it's better to connect solidly and fight the ko.

I was also about to say that black 213 at S1 kills unconditionally, because it looked to me for a little while like white couldn't connect to R10. But actually white can make it ko again with a move I didn't see at first.

Being on the weak side, I didn't comment when I first looked at the game, but before reading other comments the two things that occurred to me were that o3 seemed premature (will black defend there yet? probably not, i.e., doesn't seem urgent, and the top is bigger), and move 54, which makes a shape that sends my alarm bells ringing, particularly when it's trying to pull two weak groups together. In other words, if I noticed it too, you might be overlooking some fundamentals. Consider the normal move before trying to do too much with one stone.

_________________
Patience, grasshopper.

Shape and non-wishful reading.

There are so many optimistic moves in this game that simply do not work, and you would be able to read it if you would sit down to do so.