Working through the kyus! (Easier said than done)

moyoaji · **#61**

Played some KGS games today. The first one I'm pretty sure I was losing and my opponent forfeited to keep his rank lower as his match history has almost a dozen forfeits and just 1 win by time in all those games. Maybe I'm wrong and he just has been in a bad spot and is giving up early. I was thinking I might be able to kill the white group in the center, but it looked really hard and my reading said white could at the very least get out. Maybe my opponent just thought he misread and I was going to kill his group? (Another undeserved win to get you back to 3k. Congrats. )

The other game I thought I was losing when my opponent resigned, but then I counted and realized I was up by 20-30 points. I almost resigned myself even after seeing that my groups were alive when he lived in my corner. (You almost resigned a won game?! That would have been hilarious! )

I think I played decently in this game. Not my best, but my reading was pretty good if somewhat fuzzy in areas of my own living/connecting back.

So yeah, I'm back to 3k and I beat a 2k for the first time. Still not gonna call myself a 3k, but I think my studies are still continuing to improve my play. I'm particularly happy with my life and death skills. They have gotten much stronger since my 9k days. ("Much stronger" as in "I don't die everywhere all the time" but sure, "stronger" )

mitsun · **#62**

A few notes on the second game:

:b23:

starts the first interesting fight of the game. I have no idea what the best reducing move is around here, but this probe looks like a reasonable try. It is okay if this stone dies, as long as you get your full money's worth from it. To that end, why not atari at D4 before B6? W cannot capture at A4, as that would permit you push through to D3 for a very good result. So you could get a free and possibly useful forcing exchange before coming back to the game move. Every little bit of aji helps.

:b51:

should be at K13 for the center fight. It looks like W could play at that point in the game to capture the B cutting stones, which would be a disaster. This is the sort of game-ending fight you have to take time to read out. Fortunately W let you play there a move later. Now I think B has a won game.

:b55:

is risky, giving W chances for counter-play against the center stones. Of course, if you read out that this kills the W group, then go for it. But if you are at all unsure, why risk a won game? The simple and safe response at K15 strengthens the center stones and keeps W strongly enclosed. Sure, W will probably live along the side, but B would still be far ahead in territory. This would be a good time to count the score, to see if a kill is really needed.

:b63:

is a mistake, losing the important cutting stones in the center. Switching to L12 was still possible, if you saw the danger. W would get a good result above, but I think B would still be in the lead.

:b67:

is not necessary -- B already has a connection here. Jumping into the W moyo at F9 is huge, and might be considered even if B was not safely connected. After :w68:

it looks like W has taken the lead.

:b85:

and

were game winning responses to W overplays. Good job staying focused and finding these moves!

The final fight in the upper left corner was another case of taking needless risks with a won game. :b129: at C16 would give a safe and easy win. :b135: at B16 would stay connected, killing W without a ko fight. :w156: at G19 would kill the B group, although B would still win the game.

moyoaji · **#63**

Thank you for the comments, mitsun. I hadn't considered most of those moves. In the game, for example, I hadn't even noticed that jumping out at :b67:

was a good idea until my opponent prevented it. And you are definitely right about my play at the end of the game. I was trying to keep too much when I didn't need to, but I was too wrapped up in the game to step back and check the score. Again, I thought I was losing up until white resigned and I counted after the game. Had white killed my group I may have actually resigned even though I was ahead. (Which would have been hilarious, no? )

It seems my primary problem this game was not so much reading as counting. This is a new problem for me and one I have not been looking forward to working on. I've often fantasized about being that player who can win by playing solidly after he is ahead, but I've never taken the effort to develop solid counting skills to make this dream a reality. Daydreams aside, though, I almost lost this game because I didn't know the score. That has to change. (Ah man, I wanted to see him give up a won game. )

So I think I need to work on counting next. I need to make a conscious effort to count every single game at least a few times to know what I have to do to win. I've heard learning to count improves your rank by at least 2 stones, so that will be an important step toward reaching dan-level. (A 4k player and he still doesn't count. Now you see why I give him such a hard time... )

Ember · **#64**

Good luck with working on counting! It's one of my own (many) weaknesses and I never seem to get the hang out of it. So i think that it will be interesting to read about how you're working on it and how you improve at it. I hope you'll share some insights with us. :-)

SoDesuNe · **#65**

I made some comments for your last lost game. Overall I would recommend to see the bigger picture (i.e. not immediately defending a small area at the top) and don't randomly attach to stones or cut them ; )

moyoaji · **#66**

SoDesuNe wrote:

I made some comments for your last lost game. Overall I would recommend to see the bigger picture (i.e. not immediately defending a small area at the top) and don't randomly attach to stones or cut them ; )

Thanks SoDesuNe. Looking back on that game I realize I was being far too aggressive. Cutting and attaching everywhere trying to get my opponent to slip up. I attribute it to playing some games on FlyOrDie earlier that night and taking advantage of far too many mistakes those players make. The dual 5-4 also was problematic for me as I almost never play against it. I sometimes have my opponent play 1, but never both. It confused me in how I would approach his side of the board. (I thought you said in an earlier post you were comfortable with 5-4 joseki? Or was that only when you were the one playing for crazy influence?)

I agree that "big picture" thinking is my current weakness. This is a big reason I want to work on counting now. If I can step back and count I'll hopefully better see what is big and what is small on the board.

moyoaji · **#67**

So I have another loss. This game I really tried to count, but I got greedy somewhat early on and lost because of it. (Aw, but greed is good. ) My counting was pretty accurate actually, but I lost it early.

Still, I could have made it all work if I had seen the move I need to make at :b53:

. I have a lot of trees and comments and stuff for that move because I think it was the key to my win. I considered the move I needed, but didn't play it. If I'd only seen it then I may just have had an easy win... (Well if "ifs" and "buts" were candied nuts then we'd all have a very merry Christmas...)

Other than that I have a few notes here and there. At 117 2 I have my counting analysis of the game and that is why I decided to go for the big kill. Evaluation on that would be nice. (Did you learn how to do that whole counting thing in kindergarten? )

Here is the game:

Shaddy · **#68**

Exercise: Black to play and win at move 120.

moyoaji · **#69**

Shaddy wrote:

Exercise: Black to play and win at move 120.

L7 then J6

I am a moron... (Finally he admits it. )

Bill Spight · **#70**

A few comments.

Capturing a few stones would not have given you the lead. Think big.

Kirby · **#71**

Thanks for posting this journal, moyoaji. And congrats on achieving your goal.

As a side note, are the parenthetical annotations in your posts also written by you? I feel like I missed something regarding the format of this journal.

moyoaji · **#72**

Thanks Bill. Perhaps greed and counting are not the main things I need to work on, but instead finding what is important in any situation. Studying counting should actually help with that. I'll try to observe beyond the local situation more in my games for longer than just to count the score. (And it took you 'til 4k to get that the whole board matters. )

Kirby wrote:

Thanks for posting this journal, moyoaji. And congrats on achieving your goal.

As a side note, are the parenthetical annotations in your posts also written by you? I feel like I missed something regarding the format of this journal.

Thanks Kirby.

I will break from the format of this journal for a bit to explain. (Hang on, I didn't agree to this...)

The comments in parenthesis are written by me, yes. I like how thirdfogie put it on the first reply to this thread - they are my "internal critic." Over the course of this journal the critic speaks to the reader as a third party observing what I do through a harsh lens - emphasizing my faults and being slow to praise any accomplishments. The critic is a stereotypical "wise-guy," full of snark and sometimes wit, that tries to belittle me.

It is intended to be a humorous personification of the part of myself that finds fault in what I do or tells me that I will not succeed. I, in this journal, do not respond to my critic. This is intentional as I am refusing to accept what he says about my mistakes. (So you admit you've been ignoring me... )

I wanted to make my journal more entertaining for those reading it - I wanted to give it something unique that set it apart from other study journals. I want the critic to have something of a personality, and so I use the smilies in a lot of his comments to give them more feeling.

There are some posts that do not have my critic in them toward the beginning, but more recently he has been a part of every post I make here. He does not, thankfully, follow me to my other threads. (What do you mean "thankfully?" I'm the best thing that ever happened to this thread. )

Kirby · **#73**

Thanks for explaining, moyoaji. :salute:

Ember · **#74**

moyoaji wrote:

I wanted to make my journal more entertaining for those reading it - I wanted to give it something unique that set it apart from other study journals. I want the critic to have something of a personality, and so I use the smilies in a lot of his comments to give them more feeling.

You definitely did/do a good job here, I really enjoy reading your journal!

moyoaji · **#75**

A hard loss and a hard-fought win

(Oh boy, a title... )

So I played 2 games today. One on KGS and one on IGS.

The first game on the KGS was a difficult one. I wasn't feeling great physically, but I wanted to play a game anyway. Go with a headache is a bit like go with an opponent's handicap. (Excuses, excuses. A pro could beat the folks you play with a headache while blindfolded and giving them a 13 stone handicap. )

The game went as many of my losses do. I mostly played for territory, my opponent got lots of influence. I saw a weak group. I tried to kill the weak group when I didn't have influence. I failed to kill the weak group. I messed around for 20 or so moves and then resigned.

Later I was feeling better, but while my head was stronger, my loss had weakened my pride quite a bit. I have been doubting my sudden climb to KGS 3-4k because of the games that got me there. (You mean the guy that let you kill everything and the one that played the bad ladder? You mean those games weren't testaments to your newfound skill? ) I thought it would be nice to play a non-KGS game and I thought IGS would be a good choice since I have nothing to lose there - I've never even won a game on that server anyway. (Yeah, why'd you have to go and ruin that fun stat?)

The game started okay, but went south as soon as I got greedy - as usual. I foolishly cut at :w24:

trying to preserve my side points and keep black split. In response black was able to kill my group, but not without giving me a few free moves as ko threats. My choice - build outside influence and play the san-ren-sei. (But you hate playing for influence and you hate that fuseki!?! ) I hate playing for influence and I really hate the san-ren-sei, but I felt that was my best bet if I wanted to win after that. (I just told them that... you know, you really should listen to me from time to time. )

I started going for that influence and building up with pushes and hanes. At :w50:

I played a long-knight's enclosure to my 4-4 and waited for my opponent's reply. He gave me a chance by approaching on the inside and I attacked his group. I killed half of it and then tried to build up the upper right. Sadly, my opponent invaded the 3-3. However, I had enough strength that I played the super aggressive variation. My opponent lived in gote and then I extended along the right to undercut him. At this point I was behind. I was counting on and off trying to estimate the score and I could see I was losing. I had to make almost the whole center mine. Thankfully, my opponent extended on the left instead of going to reduce my influence. I capped him at :w90:

to build my center while threatening to attack his group again. (You know what a cap is? Well, I guess it gets played against you enough that it should be in your vocabulary...) At move 104 I offered up a sacrificial lamb to secure the left side. Peeped at 114 and finished my walling off at 116. I let black make reductions here and there, but with that center I was back in the game. (See, maybe playing for influence is your calling? Imagine: moyoaji - this generation's Takamiya Masaki! )

Even so, I counted that I was behind about about 20 points. So I played 126 to probe black and he let me make an amazing reduction of about 10 points that ended up being the difference in the game. An overplay at 185 ended up costing black his chance of reducing my corner and so I was able to win by 8.5 points. (And we all learned a valuable lesson about friendship. )

This game has again taught me the value of learning how to count. The only reason I played the way I did was because of counting. My securing of the center and probing were choices I made because I saw I was behind. Knowing the score is critical and I know I will get faster and better as I count more each game. I can easily see myself improving a few stones just by knowing what I have to do to win and when I can play more conservatively to avoid throws. (But watching you throw is so much fun! )

skydyr · **#76**

For the first game, I think at move 102, white could just defend his weakness on the top, and then use the central thickness he build to greatly reduce or invade black's bottom side moyo. If black extends with that lone stone, white should be able to sacrifice the 5 stones to get forcing moves pushing white further into black's moyo. Running with those stones leaves white with a heavy and weak group that doesn't have a lot of prospects for eyespace in black's area.

In the second game, a pincer is ideal for black, so you can consider approaching the corner with an ogeima or one-space jump instead of the wedge, to make that less helpful for black. The result was, as you noticed, a disaster. The pushing seemed good for white, but once black invades with the keima approach and plays K17, you can aim to keep white on the second line instead of splitting. As you noticed, this game was won on endgame, which white was much stronger at than black. I think black could have reduced the center much better by extending with L10. There is so much aji there that black should be able to capture something.

mitsun · **#77**

In the second game, the upper right corner group is a common living shape. The continuation after the W placement (T18) is worth studying. In the game, both sides made many mistakes, allowing W several opportunities to start ko for the life of the entire group.

moyoaji · **#78**

I don't know if any of you saw my Game Analysis post about trying to beat the Fuego AI, but I have still had that as a goal to beat Fuego straight up without undos or tricks. I wasn't really posting about it in this journal because I wanted to keep the two seperate, but I realize that this is actually a more accurate place to be posting an ongoing thing rather than in Game Analysis. (No kidding.)

I've beaten GnuGo straight up before, but even when I couldn't Fuego was always the program I was trying to beat. It's been my comptuer AI rival. It would best me at every turn and I would need to undo so many things to get the game close. I used to play it at 5 stones to get an even game! (He said Fuego was his rival. A computer program he's never beaten as a rival. Can you believe that? )

Yet tonight. For the first time ever. I beat Fuego in an even game. (What? Pics or it didn't happen! )

This is a big milestone for me. A proof positive that I have gotten stronger in my play over the past few months. To go from having trouble at 5 stones against this program to a solid win by resignation is unbelievable to me. (It's not even a strong program. Why are you even making a big deal out of this?)

What is even more: the game felt easy. Once I got past the fear of all the influence black had I was able to relax. It seemed black was ahead, sure, but his area was reducible. Another key thing was that I didn't make life and death mistakes. I didn't fail to win on bad reading. My shapes worked. I took advantage of mistakes. It was just a win, straight up. I am a bit concerned that Fuego made more mistakes than usual - or perhaps it has always made this many mistakes and I just never knew how to punish. (So it was a free win then? I guess Fuego was trying to make you feel better. )

The most important thing I learned from this game was to always keep my eye on the center, but to not rush to reduce it. That has been a problem in most of my games against people and so is something I should remember at all times. I hope I can remember this in future games and that I will continue to have wins against Fuego. It is still a challenge for me, but not an unbeatable one. Who knows? Maybe someday Fuego will need 5 stones against me. (You just went there? Seriously? )

RBerenguel · **#79**

Don't downplay it, Fuego is a decently strong AI. Kudos for sticking to it, and now look for a bigger target

Bill Spight · **#80**

moyoaji wrote:

The game started okay, but went south as soon as I got greedy - as usual. I foolishly cut at :w24:

trying to preserve my side points and keep black split.

The game headed south earlier, when White played the knight's move (H-05) for :w16:

instead of the diagonal play (H-04), which prevents the underneath attachment.

Working through the kyus! (Easier said than done)

Who is online