Life In 19x19

macelee wrote:The original question was not just about opening. There are indeed 9-dan players playing quite unique mid-game as well. One name coming to my mind is Miyazawa Goro 9-dan who plays a lot unconventional moves in mid-game. For example, have a look at this game (which I randomly selected from his game collection)

http://www.go4go.net/go/games/sgfview/31938

See his move W14, W34, W42, W52 etc. Quite a lot surprises in his game.

The black move 23 in the following game is also highly unusual. We amateur will be laughed at if playing that way.

http://www.go4go.net/go/games/sgfview/28349

Boidhre wrote:I've read some pro opinion, I think it was Yilun Yang but I may be wrong, that 4-4 and 3-4 are favoured not so much because they are best (comparing to 5-4 or 5-3) but because the flow of the following game is somewhat more predictable and he said that pros by their nature are very adverse to such unpredictability at the start of the game when there isn't a need to complicate yet.

Which I would say the predictability means most pros think it's the best. Predictability is a positive factor for them.

oren wrote:

Boidhre wrote:I've read some pro opinion, I think it was Yilun Yang but I may be wrong, that 4-4 and 3-4 are favoured not so much because they are best (comparing to 5-4 or 5-3) but because the flow of the following game is somewhat more predictable and he said that pros by their nature are very adverse to such unpredictability at the start of the game when there isn't a need to complicate yet.

Which I would say the predictability means most pros think it's the best. Predictability is a positive factor for them.

This is why I made sure to use "objectively optimal" XD I knew this sort of comment would pop up :p It's true and indeed one interpretation of "best", but I suspect not the meaning intended by the initial comment as it then doesn't mean very much.

oren wrote:

Boidhre wrote:I've read some pro opinion, I think it was Yilun Yang but I may be wrong, that 4-4 and 3-4 are favoured not so much because they are best (comparing to 5-4 or 5-3) but because the flow of the following game is somewhat more predictable and he said that pros by their nature are very adverse to such unpredictability at the start of the game when there isn't a need to complicate yet.

Which I would say the predictability means most pros think it's the best. Predictability is a positive factor for them.

Sure. What the author was trying to get across I think was just because pros forgo a certain move doesn't mean it's not perfectly playable (or that it's playable of course). It just might not be optimal when you make your living from the game.

Someday tengen will be the concensus optimal first move, regardless of komi.

Prove that wrong.

beerandchesstroll wrote:Prove that wrong.

Take a 19x19 board with 1000 points komi for white. Optimal play for black is to start crying and slap the opponent in the face with a used (for various purposes) sock.
Q.E.D.

illluck wrote:This is why I made sure to use "objectively optimal" XD I knew this sort of comment would pop up :p It's true and indeed one interpretation of "best", but I suspect not the meaning intended by the initial comment as it then doesn't mean very much.

My comment was actually based on a half-joke, one with a lot of truth behind it. It goes that Korean pros say they have found the best first move. Would you like to see it?



If you look at a database of pro games you will see that the incredibly vast majority open on either the 3-4 or 4-4 to the point of almost no other opening moves are being played by most professionals nowadays. I was able to find exceptions, but those are just that - exceptions.

I say the corners are most urgent for the reason pak0 stated - the reason every beginner is taught the axiom of corners-sides-center: it takes fewer stones to gain territory in the corner. That makes the corner move urgent, because if you don't play there, your opponent will and you will be behind. A proverb says "lose 4 corners, resign" - that isn't always necessary, but it does mean you'll have to fight hard for your victory.

You are right that there is more to consider than just territory on your first move, which is why the 4-4 is becoming the most common first move even though technically it doesn't gain you any territory. But even the less standard moves (6-4, 6-3, and 5-5) are still "near" the corner and the idea is that, when your opponent tries to take it, you already have an approach stone ready to go. Even if they don't take the corner, their location is based on the idea that the corner is the most valuable and so your opponent will have to play there or else you get a giant corner. The 5-5 is arguably more center oriented, but it is still just as close to the corner as it is to tengen meaning it still places influence toward a corner. Tengen and the 8-8 opening are the only two professional first moves I have ever seen that do not attempt to influence a corner, that doesn't mean there aren't more (I would be surprised if there aren't). Such moves are not about playing for points, they are about trying to claim the whole board, they are about outfighting your opponent, they are about complicating things, they are big and bold, not calm and logical and I doubt any professional would say they are optimal.

If a better strategy emerges, I would be surprised. Maybe some of us will live long enough for go to be solved and we'll know for sure, but if I had to put money on it, I'd bet the 4-4 or 3-4 is the "correct" first move. For now, pros like Gan Siyan can play interesting games that defy commonly held beliefs, but that doesn't mean that the 8-8 opening is going to become standard.

I've always assumed that the "it's easier to make territory in the corners" line was just a half truth that pros tell us because it's easily understandable by anyone and naturally leads us to play moves that actually achieve a purpose. (Something that's harder than it sounds )

Who knows what goes on in pro heads, but I see corners as key points from which to affect the rest of the board, and I find it hard to believe that any pro has ever found themselves thinking "I'll play a 3-4 because it makes 6 points."

Splatted wrote:I've always assumed that the "it's easier to make territory in the corners" line was just a half truth that pros tell us because it's easily understandable by anyone and naturally leads us to play moves that actually achieve a purpose. (Something that's harder than it sounds )

Who knows what goes on in pro heads, but I see corners as key points from which to affect the rest of the board, and I find it hard to believe that any pro has ever found themselves thinking "I'll play a 3-4 because it makes 6 points."

They would if the 3-4 was open in yose . (And, actually, that still might be there logic on move 1)

You are right, you don't just play the corner for points (otherwise the 3-3 would be the most popular opening move) but if you are going to make a first move on a go board, the corner will get you more potential for points than any other area. This is why shoulder hitting a 4-4 stone is so bad (again, there are always exceptions, but in general). A shoulder-hit to a 4-4 grants your opponent fourth line corner profit. In 5 moves they can have 15 points of solid territory. You can't get that anywhere else. I would gladly play a game where I get every corner on the 4th line and my opponent gets outside thickness toward each side.


I say the corner is the urgent point because nowhere on the board will you get so much for so little. Yes, other points on the board are big, but I don't feel they are urgent. Now, in the heat of a fight, even a corner doesn't feel like an urgent point (Lee Sedol's ladder game had a corner open for half of it) but it is always nagging at you and when your opponent gets a free moment they will play it.

Go is a game of territory, but it is also a game of efficiency. Yes, outside influence has value - but only because it becomes territory. You don't invade a 4-4 right away because black's wall will get him more profit than the 10 points in the corner. However, they are also not efficient points. You need to play 6 moves to get those 10 points.

Think of it like this: if you get 6 points on your first move: six real, permanent points, you have just made a great move. In a game of go where 200 moves are played (100 by each side) will either side have 600 points? What about just 300 points? 150 points? If you can make 1.5 points of territory per move you will win a game that lasts 200 moves. So if you can make 6 on move 1 that sounds pretty good to me.

Points aren't important in go.

Fighting and thickness..

NoSkill wrote:Points aren't important in go.

Fighting and thickness..

Only if that can be converted to points later.

My understanding is that the corners are important because they are the easiest place to make a stable group (which is different from territory), and stable groups are important as one prepares for midgame fighting.

@moyoaji & Pak0: I'm afraid I remain completely unconvinced. It's true that 6 points is a lot for a single move, but fighting is inevibitable and if you find yourself at a disadvantage you're likely to lose more than 6 points.

NoSkill wrote:Points aren't important in go.

Fighting and thickness..

Splatted wrote:@moyoaji & Pak0: I'm afraid I remain completely unconvinced. It's true that 6 points is a lot for a single move, but fighting is inevibitable and if you find yourself at a disadvantage you're likely to lose more than 6 points.

I am very confused as to why you aren't understanding this. In go, you win by scoring more points than your opponent. Fighting is only valuable because it earns you points - as pak0 already stated.

Basically, this is what you are saying: "Scoring an early touchdown in football doesn't matter because there is a lot of football left to be played in the game." I don't care how important controlling the line is, or how valuable having a good kick unit is, or whether or not our running back is better than the other team's. If my team gets a touchdown in the first minute, we are winning. We may or may not win by the end of the game, but being ahead is the first step in winning.

And this is completely missing the point of what I said anyway. I am not advocating that playing purely for points wins go (although if you don't know how to play a territorial game you are missing a big part of go strategy), what I am saying is that because points are earned more easily in the corners it is better to play there first. And, if you want to play for thickness and fighting, you should play a 4-4, 5-4, 5-3, 6-4, 6-3, or 5-5 stone instead of a 3-3 or 3-4 stone.

However, don't be upset if your opponent wants to let you have a giant corner when you play one of these moves - be thrilled. A 4-4 is not a magic move that automatically gives you outside influence. I can take it away in a heartbeat by playing the 5-5. Now your 4-4 is only facing the corner and will have no bearing on any fights that occur later in the game. So why don't people play this way all the time? Because the one who played the 4-4 should say: "Oh, you want to give me a 20 point corner? Well, I wanted outside influence, but I guess having a massive corner is okay too." And because fighting is a part of go and because your opponent doesn't do what you want them to when facing thickness they will simply reduce anything your outside wall tries to create and make sure you get less than 20 points on the outside instead of playing super aggressively and trying to fight you when you have a wall. Then they win because they have more points and you can't reduce their corner because, after all the fighting is over, go is a game of territory - a game of points.

skydyr wrote:My understanding is that the corners are important because they are the easiest place to make a stable group (which is different from territory), and stable groups are important as one prepares for midgame fighting.

This is definitely a good way to think about corners. Having safe groups is key in the mid-game and any group that isn't safe becomes the source of an urgent move (for either side). I would say this would go along with my thought that the corner is urgent. However, the reason being because it is the fastest way to get a stable group on the board. Yet there are those who play a territorial strategy in go and win doing it so I don't think it's wrong to value the corners for the points they can give you.

Go strategy is broader than anything we've brought up here. I think some could argue that the corners are valuable because the moves there are flexible. If you play on a side you typically need to make a two-space extension or run for the center, but in a corner you can choose to go for a side, or stay in the corner, or head toward the center meaning any stone played there has 3 options and your opponent can only eliminate one so you'll still have a distinct choice afterwards - or if they don't eliminate one you have 3 ways to build from your corner.

One side is talking about "points at the end of the game" (A). This is the only aspect of the game that matters according to the rules and is a trivial statement. The other side is talking about "solid points" (B) at specific locations that cannot be taken away by the opponent. Is this not what is happening here?