Go

I've been playing for 40 years. Welcome to the community!

As you pointed out, go seems to be a richer game than chess. The structure is much simpler but it leads to far more complex situations. Because of this, computer programs don't play very well. They've been working on go software for as long as they've been working on chess software. Forty years ago there was a million dollar prize waiting for the first program that could simply understand the rules--make only legal moves, not kill its own stones, and stop playing when the game was obviously finished. It took about twenty years before somebody collected, at which time toy chess computers could beat the average social player. Now that a chess program has won at least one game against a world champion-caliber player, the best go programs are barely strong enough to give me a good game.

My point is that playing against software is not a good way to learn more than the basics. They're good at tactics and will demolish your groups if you're not good at making two eyes. They're very good at deciding which of two decent-looking moves will have a greater impact on the score. But they have no sense of strategy. When you get past the point of wondering which move will keep a totally surrounded group from dying, you'll find that they lead you off into quixotic battles that no human player would pursue, for the very good reason that you can make more territory more safely by doing something else.

One of the most important skills in go is pattern recognition. Suddenly you don't see individual stones and sit there going, "This one is connected to this one but these two could be cut apart if that one goes that way," and instead you see a group that has the shape to make one eye, is connected to another group, and will stay connected as long as enemy stones don't get any closer than there. This kicks in about the time you reach the level of 10-kyu (Japan/U.S.) = 16-kyu (China/everywhere else). I presume you know the ranking system. You start out at around 30-kyu (Japan/U.S.), the lowest rank at which you play consistently enough to be rated. Each rise in rank is equivalent to about ten points in score. Someone who consistently beats you by 30 points is 27-kyu.

We use the Japanese system because 40-50 years ago when go first hit America Japan was the only country that took the game seriously, had professional teachers and competitors, and big-money tournaments. I don't know if they're still hanging onto the championships, but Chinese and Korean masters are proliferating and quite a few Westerners have made it into the professional ranks. When you reach 1-kyu the next step up is 1-dan and the numbers go up as you get stronger. 6-dan is the highest amateur rank, if you're stronger than that you've achieved 1-dan professional status and the highest professional rank is 9-dan. (The Chinese scale doesn't have amateur ranks, so an American or Japanese 6-dan amateur is 1-kyu to them.)

The beauty of the game is in the Oriental culture that comes with it. We don't say one player is "better" than another, we say "stronger." We don't say "You messed up there," we say "Black played a very weak move." It is considered the duty and honor of the stronger players to teach the weaker, otherwise where will the next generation of stronger players come from?

The handicap system is arguably the nicest and most elegant part of the game. It makes it possible for two players of vastly different skill levels to play a game that is fun and challenging for both of them, and which gives them both an even chance of winning. White gives one handicap stone to Black for every difference in rank. The nine dots on the board are where the first nine handicap stones go, in a pattern which your software probably shows you. I know where the next 24 handicap stones go, although very few people like to play against more than 9.

So if you want to pursue this game seriously, you need to do two things.

One is to continue playing against anyone you can find, including--for now at least--the software. You'll get an understanding of how small a group can be and still make two eyes if attacked, how far apart two stones can be and still make a connection in an emergency, where the important places on the board are to make big territories, what kind of moves are bonehead errors, etc.

The other thing is to get some "teaching" games with stronger players. The one thing you cannot learn by playing against opponents of similar rank is how to build a really strong, solid defense. They can't find the weaknesses in a merely good defense. They can help you learn how to fight and make land and all that stuff, but this one thing will hold you back. You'll approach the end of the game looking like you're fifty points ahead, and a stronger opponent will simply start killing all of the groups you thought were safe, until he's got his fifty points back plus one. A stronger opponent will also teach you about strategy rather than tactics. You'll spend the game building a bunch of safe groups that can't be killed, and when you look up the enemy will somehow have surrounded three quarters of the board.

So play some games against some much stronger players who give you a six-stone handicap, a nine-stone handicap, or even more. It's fair if they shave your handicap by 3-4 stones to make it a true teaching game. White wants to be able to make moves that are instructive rather than purely competitive. He does not want to have to make moves that are so complicated that you can't understand them.

Some simple pointers to get you moving in the right direction:

The corners are the most important places on the board. Try to claim them for yourself, or at least prevent the opponent from getting them or from getting too much land in the ones he has. The most oft-quoted Go Proverb is, "If you lose all four corners, resign." Except when two professionals are feeling frisky and try an experimental strategy, play always starts in the corners. When those are taken it moves out onto the sides. If there is any land left in the center--and there usually is very little--it's left for last.

If you're wondering about that, notice that it takes only two stones to surround one point of land on the corner of the board. It takes three to do that on the edge, and four to do it in the center. It's simply a more efficient use of your stones to make land first in the corners and then on the sides.

When in doubt of what to do, pretend that you're the other player and it's your move. Find the move you would want to make. Your own best move is probably in the exact same spot, or every close to it.

Don't get nearsighted and just look at the area you're playing in. Step back and look at the entire board. Fights in the different corners affect each other.

Every move must have at least two purposes.

Sente and gote are important concepts. A "sente" move is one that absolutely requires the opponent to respond to it, or else suffer a ruinous loss like a dead group or destruction of a large territory. A "gote" move is one that does not require a response. Whoever has sente has the initiative, the power to direct the development of the game until he can't hang onto it any more. Early in the game, having sente is equivalent to a ten point advantage.

When you play a high-handicap game, bear in mind that a different strategy is called for than in an even game. You want to play 100 percent defense. Look at it this way: If you have a 9-stone handicap it's like starting the game with 90 prisoners in your bowl. The only sensible strategy is: I can play really chicken and lose 89 of them. White's only advantage is superior strategic and tactical skills. So don't let him lure you into any fights, which will enable him to use those skills. Just make land, and when in doubt let your motto be, "Better safe than sorry."

As a beginner, remember that like all Oriental pastimes, go teaches you humility. You have to make the move that works, not the one that looks like fun. You've got shapes that were discovered by the Ancient Masters back in the mists of history, and they still work. You will not discover a better one after you've been playing for four years.

I play on KGS.Kiseido.com . The community there is a little lacking in the Oriental etiquette that's supposed to come with the game. They're a bit too competitive and it's not easy to find a stronger player willing to teach. They love to play with timers, often as ridiculously fast as 15 minutes per side. At your level a game should take two hours. If you can find a better bunch, go for it, but this one seems to be pretty popular. There are people all over the world, so any time of night or day you can get a game. There's a chat window with each game, you can talk about the game or anything you want. There are "rooms" where they speak everything from Hebrew to Catalan, but English is most prevalent and nobody will discourage you from playing with somebody you can't talk to.

If you're looking for a teaching game on KGS, send me a private e-mail and I might accommodate you. However I would appreciate it if you wait until you have a solid rating of at least 27-kyu (Chinese system, all the internet rooms use it). You can get there by just bashing stones around with your peers.

Please Register or Log in to view the hidden image!

One last note. You should start out on a smaller-than-regulation board. 13x13 is pretty good. That way you get to see the results of your strategies while you can still remember why you chose them. And that way you can spend the time you need thinking about each move without the game lasting all day.

9x9 boards are good for learning the rules, but don't spend too much time there. Some people make a career of becoming the best 9x9 players and they're really good, they beat me every time. It's a different game. The edges are so close together that the geometry doesn't work the same as on a real board. If you start getting too good on a 9x9 it's very likely that you'll have to unlearn a lot of stuff to play well on a full size board, so don't fall into that trap.

Enjoy, and good luck!