Hi everyone,

This is my first post here! I've been playing for a couple of months or so (have played my first 100 games), but the game below was one game I played today where I lost by komi. There were a few basic mistakes from me (like my "throw-in" in the lower-right), but if anyone has any recommendations/suggestions on my play I would be very glad to hear. (Go easy on me!)

I was black.

Best wishes,
Jim.

Hi Jim, Congrats and welcome.

This is a big move yes; however, if you look at the whole board,
do you see you have any weak groups ? W can play something like E8 -- your C8 group is in big trouble.

Yes, big move.

- Bad habit, bad exchange, if you don't plan to cut at J9.
The end result here is your just bump yourself, bump your head, into White's wall.
You reduce your liberties. You are behind White.
If the J9 cut is bad for Black, or, you don't want to cut at J9,
then don't do this - exchange at all.
Imagine if W simply connects at J9 afterwards -- do you see your stone is (almost) completely useless ?

Now you can see is at a wrong place.

Please see Toothpaste. Do you see you've made a toothpaste shape for yourself here ? ( W H9 is toothpaste. )

Very slow. H16 much bigger for White.

White let you off too easily here; W should continue to push through at L3, destroying your bottom here.

- Both of you don't understand the broken shape at H4. Please see again Toothpaste.

Please see again Toothpaste.

103 Correct local shape is P17 instead. (Equivalent of E3. )

Welcome to Go! Given the short time since you started playing it was a decent game. It was finally given away due to small and overcautious moves in the endgame - especially empty triangles (http://senseis.xmp.net/?EmptyTriangle).

A few comments from my side

- I do not like this move. After white's reply the local shape of your upper left group is not yet settled. With the - exchange you let white split you and are working against your own group. In the subsequent phase this group has to crawl a lot to get life with only few points while white gets a solidified upper side during that process. Also due to the splitting of w your c10 / c8 stones look awkward for a couple of moves, a stronger w player would probably have killed them. Instead a typical joseki would be as follows:

Click Here To Show Diagram Code

[go]$$Bm18
$$+------------
$$|............
$$|............
$$|..d.........
$$|...O.O......
$$|............
$$|..XcX.......
$$|...ab.......
$$|..O67.......
$$|..21........
$$|..43........
$$|...5........
$$|..8.........
$$|....9.......
$$|..O.........[/go]

Since w has helping stones at lower left it is probably not the best choice for b (that also means that even your should be discussed, maybe a double corner approach instead?), but at least your left side stays connected. White can later peep at a and - when you block at b - cut at c. However as long as the upper left corner is open, it is no problem for the c14 stone: You jump in at d (3/3 point) and either live in the corner or connect with c14.

There may be more to say about the opening but that's an area where I do not have many clues either.

Now about the small empty triangles in the endgame (only a selection, they are certainly more aspects which stronger players could explain better than me):

- Move 154: (two empty triangles simultaneously) -> should be at e.g. c16. White will have to protect inside and there is no danger if he plays e15 later.
- Move 158: If you just played e7, could white really cut you at d8?
- Move 160: What was your plan if white now cuts at f8?
- Move 162: Since the nearby w stones are still weakish you should proceed with your plan rather than fixing the cut...
- Move 166: White just played an empty triangle on his own, can you punish it now?
- Move 172: What are you protecting?
- Move 174: Instead of the empty triangle, hane at L10 looks a bit more interesting. If white defends the cut L11 (I don't know if he needs, at least it could cause some confusion), you can atari at n10 and capture n9 afterwards.
- Move 188: One point in gote. Instead you could have 4 points in sente at c1. At least you get c1 on move 198, but the other sente endgame A18 was never played.
- Move 232: There is still A18! IMHO this was the last losing move.

Edit: After the - exchange, black can still play a move which connects the upper group with c10. This seems much better than the joseki variation which I gave above. Can you see it?

It's late where I am and I have to get into bed soon, so no review from me for the time being. You did not play too badly for someone who has been playing Go for only a few weeks.

At your level, the most important thing is to weed out bad habits through constant practice and refinement. First thing is to play as many games as possible and have each one reviewed by players stronger than yourself. Whenever you cannot find human opponents, you can play against computer ones. For beginners computer opponents offer a good challenge. Online Go Server has computer opponents you can play against. If you have an iPhone or Android device, a go-playing program such as Go GridMaster will give you a good game.

Playing regularly and getting reviews is only the beginning, though. Developing skill at reading out life & death and tesuji situations on the board during play is highly important. To do so, get a collection of life & death and tesuji exercises; these 2 categories are the bread & butter of many a Go player and so deserve constant attention. Be sure to do lots and lots of easy exercises to reinforce the ability to read out board positions quickly and accurately; if an exercise takes you more than 60 seconds to solve, then it is too difficult for you. In life & death, there are some proverbs and tips to remember, such as reducing internal eye space and playing on vital points, and "there is death in the hane" and "On the 2nd line, 6 die and 8 live". In tesuji, there are themes on techniques such as connect-and-die, snapbacks, throw-ins, jumps, ladders, nets, peeps, cuts, etc. As well, you should do exercises in the fuseki, joseki, middle game, and endgame categories, but L&D and tesuji are the most important categories. Out of the 4 aforementioned, endgame is the most important, since many close games are won or lost in the endgame phase and strong endgame skills can give you the advantage if you are falling behind by a few points in a game.

You can go here for online exercises: http://goproblems.com

If you want a printable collection, go here: http://tsumego.tasuki.org

There are many printed exercise collections out there. Kiseido Publishing (http://www.kiseido.com/go_books.htm), Slate & Shell (http://www.slateandshell.com), and http://shop.gogameguru.com have books you can purchase.

If you have an iPhone or Android device, you can download apps from App Store (on iPhone) or from Google Play (on Android). Two good programs to consider are MagicBaduk and TsumegoPro.

The more exercises you complete, the better. Be sure to do as many easy ones as possible. After a while, try new ones and work on them till they become easy, too.

Also, if you play against the same kind of opponent over and over, you pick up their playing habits, win or lose, and not all amateurs have good playing habits. Pros, on the other hand, go through hard training to weed out as many bad habits as possible, so they almost always play in a way that makes sense, even though it often takes a dan-level player to understand what is happening in pro games. That being said, getting a pro teacher would solve the problem, but not everyone has access to one. Thus, I recommend replaying pro games from start to finish. You will benefit in several ways. You will learn how to think whole-board and get to observe how a game of Go is played between two experienced players. You will get your brain accustomed to reading out long lines of play; many of the hardest Go exercises have answers a few dozen moves long. You will be able to think up new lines of play and see how all that Go knowledge is put into action, not to mention getting fresh new ideas that you can apply during play in your own games. Of course replay of pro games is not a magic bullet, but it will help you to go in the right direction as far as your future development in Go is concerned. In the end, you will be able to play confidently in a style that suits you best.

For some downloadable game collections in PDF format, go here: viewtopic.php?f=17&t=10643
All of the games are in diagram form, so these may be a bit hard to play through with board and stones at first as it takes some getting used to. That is, you will be able to replay games faster and faster with practice, since you will be able to guess where on the board each move is likely to be played.. For wet-behind-the-ears beginners a game rendered as a move list may be easier to replay. (See below for attachment.) These are best replayed on a board that has the lines marked as you would see them in an SGF editor.

If you prefer digital replay here are two places you can go online.

For modern games: http://gokifu.com/ (uses Adobe Flash) or http://www.go4go.net/go/ (uses EidoGo SGF editor)

For historical Japanese games: http://jakkoji.blogspot.com/ (uses EidoGO SGF editor)

To get you started, here are 6 game records as examples. Just click through them. The 2nd one is the "blood vomiting game" of 1835. The 3rd one is the all-famous "ear-reddening game" of 1846. The 4th game is a game between Ho'inbo Shuei - considered by many to be the strongest of the old Japanese masters - and Tamura Yasuhisa, later to become Hon'inbo Shusai, the 21st and last Hon'inbo before the title was given to the Japan Go Association to be decided in an annual tournament. The 5th one is a game that was being played in Hiroshima between Kaoru Iwamoto and Utaro Hashimoto - 2 of the strongest professional Go players of the 20th century - when the atomic bomb was dropped.













And below are the game records in move list format.

Thank you so much for the helpful feedback! It seems I have much to learn (but it's very fun doing so!)!

EdLee: Thanks for the analysis! I saw the C10 group needed helping, but I played almost as a sacrifice. I play chess to almost expert level, and one thing I am finding difficult in go is the idea you can give stones up to gain elsewhere. So, on this move, I thought I could allow to lose the C10 group if I totally gain the lower and right sides. I see now that I could have perhaps had both Your comments have helped a lot, so thanks for your time.

schawipp: Thank you for the comments! It seems everyone agrees my handling of the C10 group was very bad. You've given some excellent recommendations for how I could have better handled it, so thanks! Thanks also for the move-by-move comments. Yes, I'm trying to weed out bad shape where possible, and endgame is still almost alien to me (but will work on it once I stop dying in other games!). I saw the way to connect the C10 group to the upper left during the game, but again because it involves the loss of a stone my chess-brain almost immediately discounted it. I really need to work on this! Thank you for taking the time to comment!

tekesta: Wow! Lots of material to work through here! Thank you very much

Jim.

Hi Jim, You're welcome.I didn't say anything about , because IMO it's not a problem at these levels.
You decided to start a fight when you jumped out at --
you could have instead simply taken 3-3 (C16), a peaceful result.This... seems a bit strange, since chess has analogous features:
in chess, everything is also an exchange;
you have no problem giving up pieces as long as you get an equal or better
result:
All else being equal, you're happy to give up:

• a pawn for a pawn;
• 2 pawns for 2 pawns;
• a pawn for a knight/bishop/rook;
• 2 pawns for (pawn + knight);
• 2 pawns for (pawn + bishop);
• a pawn for a significant gain in tempo/position;
• etc.

It's the same in Go: you're happy to give up any groups
so long as your global result is equal or better than before the trade.
Like in chess, all else being equal,
you're happy to give up 50 stones if you kill 51 enemy stones.
The tricky part is the evaluations.

Hi EdLee,

Oops, I meant , not ! Sorry about that!

The relation to chess I guess will become clearer as I improve at Go. What I meant basically is that at the level I play at, the loss of a pawn can be the game lost. Such a small amount of material at the upper levels can really be a decider. Yes, you can give material for compensation (tempo, initiative, attack, etc.), but I think at my current level of understanding with Go I don't yet see the compensation and just feel the loss of a stone; therefore I often look to save material almost automatically, and I am having to fight that urge until I have a better understanding of the compensation in Go. (Beginners in chess often don't see the importance of development & initiative, for example.)

Thanks again!

Jim.

At the highest levels, Chess and Go do not differ by much. Just as the loss of a pawn in Chess can bring the risk of falling behind for much of the game, in Go trailing behind by just 2 or 5 points is a big deal. For beginners 5 points may not seem like much, but among pros, 5 points is a huge margin.

There is a common proverb in Go that says, "A rich man never picks quarrels." What this means is that if you are ahead in territorial gain as the game progresses, your priority should be just to maintain the lead. Fighting is what your opponent will want if he is behind; if you mess up during the fight, he will take the lead and you will fall behind. So it is better just to develop your groups peacefully and stay out of fights whenever possible. In our time this may not be 100% true, though; ever since the advent of komi scoring in Go, Black no longer enjoys the natural lead it once did in the pre-komi era. This means that in the opening phase of the game Black side will want to play more developmentally efficient fuseki styles and, at least nowadays among Korean and Chinese pros, fighting tends to occur often during play. This is because territory gained through fighting is considered to be more secure than territory that is directly created.

The 4th Golden Rule of Go says, "Sacrifice stones to gain initiative!" There are several situations in which this rule applies, but the most important such situation is when you are willing to let go a few of your stones in order to gain initiative (sente, in Japanese) to expand your groups on a strategic basis. Which reminds us of the 5th Golden Rule: "(Put aside the small) and concentrate on the big."

Losing material is not alien to Chess. Ever heard of the gambit? Capturing that pawn might seem like a step ahead, but the opponent actually intends for you to damage your own position in the center of the board. There is also the "queen sacrifice". Not easy to pull off, but spectacular when it does happen.

In Go, throw-ins are a common technique to aid in the capture of a group of stones; the liberties available to a group under attack are reduced. The "Crane's Nest" problem is probably the most famous example of this.

When it comes to accurate assessment of material gained and lost, it all boils down to experience. The beginner will not have the background experience to appreciate the importance of a lost pawn or a couple of lost stones, but the experienced player knows better and so careful measures are taken at all times during play. The 7th Golden Rule of Go says, "Avoid impulsive moves!"

In Chess, one starts with 32 pieces - 16 per side - and the number of pieces tends to dwindle as the game progresses. This is why chess endgames tend to be simple. In Go, one starts with an empty board and the game develops with each stone that is played. One can say that Go is more like poker than like Chess.

As a small bit of consolation, here is a classic example of monkeys falling from trees:
https://www.youtube.com/watch?v=qt1FvPxmmfE

Hi Jim, yes; the evaluation -- as in chess, it's a never-ending learning curve.

Are chess endgames simple now? I found myself surprisingly good at chess and xiangqi for having never really played them. The "skill" I feel that transferred was checking I was not making a gross mistake somewhere on the board every move, and checking whether my opponent had.

.5 point end game means endgame is a good thing to practice, right. Make a one point better play somewhere in the endgame.

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