trick moves

[hyperpape]

If I can join in the analogy game:

Belief aims at truth. But there's still something going wrong when you make a lucky guess that turns out to be true.

[jts]

That's really interesting. One aspect my conception of hamete is that at my level you know hamete if you memorize hamete, and if you don't memorize them you can't see the trap in advance;

As I said earlier I don't really believe this. Most hamete rely on the opponent seeing the trick play as overplay and trying to punish it. After they aggressively cut/captured/whatever and got what they wanted they look around and realize that the game has just radically shifted in their opponents favor.

Um. Maybe. I don't try to learn hamete, so I guess I'm totally unqualified to talk about "most". But what about the 4-4, low approach, second line tsuke. Taking the eye of an elephant jump doesn't strike me as trying to "punish" your opponent, it's a good instinct. The correct reply strikes me as utterly counterintuitive. Do you think this fits into your theory of hamete or mine? Maybe you could provide some example of hamete that fit your understanding?

If you think your opponent is just playing a new move you'll try to whether the storm and play defensively. You might still lose some points but you won't have been tricked.

If this were generally true of hamete moves - that the best response was to be submissive and lose a few points - then they wouldn't be hamete, they would be tesuji.

but as players' reading abilities wax they can simply read out the whole sequence so, at higher levels, the whole concept of hamete evaporates.

That's where the definition of trick play comes in again that was raised earlier in this thread. Does a complicated taisha joseki, that you've studied earlier and your opponent not count as a trick play?

Well, the key difference is that (modulo the whole board situation) playing a joseki move is still good for you even if your opponent plays the best response. What's special about the taisha is that it's both very long, and contains a lot of junctures where a "mistake" (i.e., a deviation from joseki) is much more costly than the average joseki deviation. But from the limited amount of time I've spent looking at the taisha and the avalanche josekis, each play is independently a sound play with good principles backing it up.

Again, I'm not insisting that learning/memorizing hamete is so different from learning/memorizing joseki. After all, Kageyama agrees with you (ish). I'm just asking for evidence of professional-level hamete. That would disprove one implication of my interpretation of hamete: that hamete are an epiphenomena of weak reading.

[Bartleby]

Do you know which Dosaku game inspired Kageyama?

Well, Kageyama immediately follows his introduction with a trick play in a Dosaku game, so perhaps that's the game he is referring to. However, Kageyama's description is rather vague:

"In the early period of the Tokugawa Era, Honinbo Dosaku faced an opponent who had come to Edo from Okinawa and was playing with a four stone handicap."

He later describes Dosaku's opponent as "Okinawa's best."

In a later chapter (by a different author), another Dosaku trick play is given from a game played in 1668 in which Dosaku had White against Chitetsu.

[Bartleby]

I think you can view a game from many different perspectives and play it for many different reasons, all legitimate.

From the perspective of a serious competitor, winning (within the rules, of course) is the primary goal and a trick move that works serves this goal and therefore is a useful tactic, albeit one that must be employed with care.

From the perspective of a player who is trying to play optimal moves every move (and I personally feel that this is very optimistic; unless you are a pro, a large percentage of your moves are going to be anything from subtle inaccuracies to gross mistakes), playing a trick move may not serve your goal and therefore is to be avoided.

These two perspectives are both legitimate but I think the first one is perhaps more realistic and more likely to lead to the development of a really strong player. Although I can't speak about Go, I can say that every very strong chessplayer I have known was highly motivated to win. It's one of the reasons they are very strong players. They will play on in positions where objectively they are losing, they will set traps for their opponents, they will fight to the end for a point or half a point.

As for trick plays in particular, in chess young players are often encouraged to play gambits of various degress of soundness and unsoundness because it helps develop a feel for the initiative, a sharp eye for tactics, an understanding of the principals of attack and defense. It is not that these gambits are considered entirely correct play; rather, it is that they have a lot to teach the player who employs them.

I see no reason why trick plays should not serve a similar purpose in Go. Those who avoid trick plays because they are not "best" may be missing out on a chance to learn more about the imbalances that can be created on the Go board, about techniques for sharpening play, and about the pitfalls that lurk in ordinary positions for the unwary.

Even if you are of the "play the best move" viewpoint, I think it is wise not be offended or bothered when trick plays are played against you. It is better to play your own moves and let your opponent play his. Trying to control your opponent's style to suit your own preferences is both wrong and ultimately self-defeating.

Just my two bits.

[Shaddy]

That's really interesting. One aspect my conception of hamete is that at my level you know hamete if you memorize hamete, and if you don't memorize them you can't see the trap in advance;

As I said earlier I don't really believe this. Most hamete rely on the opponent seeing the trick play as overplay and trying to punish it. After they aggressively cut/captured/whatever and got what they wanted they look around and realize that the game has just radically shifted in their opponents favor.

Um. Maybe. I don't try to learn hamete, so I guess I'm totally unqualified to talk about "most". But what about the 4-4, low approach, second line tsuke. Taking the eye of an elephant jump doesn't strike me as trying to "punish" your opponent, it's a good instinct. The correct reply strikes me as utterly counterintuitive. Do you think this fits into your theory of hamete or mine? Maybe you could provide some example of hamete that fit your understanding?

If you think your opponent is just playing a new move you'll try to whether the storm and play defensively. You might still lose some points but you won't have been tricked.

If this were generally true of hamete moves - that the best response was to be submissive and lose a few points - then they wouldn't be hamete, they would be tesuji.

but as players' reading abilities wax they can simply read out the whole sequence so, at higher levels, the whole concept of hamete evaporates.

That's where the definition of trick play comes in again that was raised earlier in this thread. Does a complicated taisha joseki, that you've studied earlier and your opponent not count as a trick play?

Well, the key difference is that (modulo the whole board situation) playing a joseki move is still good for you even if your opponent plays the best response. What's special about the taisha is that it's both very long, and contains a lot of junctures where a "mistake" (i.e., a deviation from joseki) is much more costly than the average joseki deviation. But from the limited amount of time I've spent looking at the taisha and the avalanche josekis, each play is independently a sound play with good principles backing it up.

Again, I'm not insisting that learning/memorizing hamete is so different from learning/memorizing joseki. After all, Kageyama agrees with you (ish). I'm just asking for evidence of professional-level hamete. That would disprove one implication of my interpretation of hamete: that hamete are an epiphenomena of weak reading.

I give here a trick play which fits Mnemonic's description. By the way, the eye of the elephant's jump is not always a good move- it just so happens that most times when the elephant jump appears in kyu games, it is an unreasonable shape and playing at the eye punishes it. But sometimes the elephant's jump is the correct shape, and those times occur when your opponent can't play the eye.
[sgf-full](;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2]
RU[Japanese]SZ[19]KM[0.00]
PW[White]PB[Black]
;B[pc]
;W[qe]
;B[qg]
;W[od]
;B[mc]C[A classic trick play. Black's move looks like overplay, doesn't it? So white's instinct is of course to punish by cutting. But..]
;W[oc]
;B[ob]
;W[nb]
;B[pb]
;W[nc]
;B[pd]
;W[pe]
;B[oe]
;W[ne]
;B[of]
;W[rc]C[The center of three stones, the vital point. All of this is a one-way street]
;B[qd]
;W[rd]
;B[re]
;W[rf]
;B[qf]
;W[se]
;B[pf]
;W[re]
;B[md]
;W[nd]
;B[mb]
;W[na]
;B[mf]
;W[qb]
;B[nf]
;W[pa]
;B[me]
;W[oa]
;B[ma]
;W[qc])[/sgf-full]

[Redundant]

[sgf-full](;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2]
RU[Japanese]SZ[19]KM[0.00]
PW[White]PB[Black]
;B[pc]
;W[qe]
;B[qg]
;W[od]
;B[mc]C[A classic trick play. Black's move looks like overplay, doesn't it? So white's instinct is of course to punish by cutting. But..]
;W[oc]
;B[ob]
;W[nb]
;B[pb]
;W[nc]
;B[pd]
;W[pe]
;B[oe]
;W[ne]
;B[of]
;W[rc]C[The center of three stones, the vital point. All of this is a one-way street]
;B[qd]
;W[rd]
;B[re]
;W[rf]
;B[qf]
;W[se]
;B[pf]
;W[re]
;B[md]
;W[nd]
;B[mb]
;W[na]
;B[mf]
;W[qb]
;B[nf]
;W[pa]
;B[me]
;W[oa]
;B[ma]
;W[qc])[/sgf-full]

Fun fact: I've seen Shaddy let this wall die ...

[mongol]

"Go is not about winning, it is about total humiliation of your opponent." (from KGS kibitz)

Somebody's been watching too much poker on tv. Sounds like the Jack Strauss philosophy of busting your grandmother at the table. Usually total humiliation will only occur if you're sandbagging, or your opponent loses a large group.

I don't think there's really such a thing as a trick move. If you don't respond correctly, the move will work. If you do, the opponent suffers a loss. It's more of a deliberate mistake, but then if we're behind and take chances, we're kind of doing the same thing, hoping the opponent responds incorrectly.

In live play, any of you guys ever play a move and then pretend to notice something in another part of the board, and stare intently at that, hoping to distract your opponent from where you just played?

[shapenaji]

The odd thing about hamete, is that with a slight change in the surrounding positions, it can actually be good, even assuming they know the "refutation"

That being said, I usually avoid it in even games (mostly because I think it's a waste of time to learn joseki patterns by rote), I think it's fine to leave traps on the board, that's part of the strategy. The trick, I think, is leaving your traps light, so you can get aji out of them. As soon as you commit to a hamete sequence, you usually become a good deal heavier.

I probe with hamete, if they respond correctly, I play somewhere else and leave the aji for when the fight gets there.

[shapenaji]

I thought, and still think, that the priorities listed above are in the wrong order, certainly winning must come first.

This is a game, with a goal, and any judgment of beauty or interest can only be made based on a moves ability to attain that goal. We can make good shape to our heart's content, but if the move does not serve to lead us down a winning path, there is no beauty, or interest, there is only misplaced artifice.

If you follow this line of argument, you must also claim that a beautiful or interesting game does not exist, because someone lost it.

I don't know that that's what you get when you follow this line of argument. Losing a game does not mean that the person who lost did not play skillfully in pursuit of a win.

It would be hard to play skillfully if you were not trying to gain points. Any move might be considered beautiful.

Also, perhaps the beauty that we find in localized sequences (those that do not advance the goal of winning because they are played at the wrong time, and hence are not large enough) has more to do with the typical Amateur weakness when it comes to timing.

We have difficulty noticing the right time to start sequences, we miss out on this dimension of play, and as a result we miss the great glaring flaw in the jewel of a sequence.

Each sequence lives in:
2 dimensions of space, 1 dimension of length, 1 dimension of timing

[Aeneas]

...
I distinguish three aims:

1) to create a beautiful game
2) to create an interesting game
3) to win
...

I thought, and still think, that the priorities listed above are in the wrong order, certainly winning must come first.

You are right that winning in some sense has priority: Without the aim of winning there would be no game at all. And moves which do not aim at winning but even lose points are not really interesting moves. Also, a game without any interesting moves can hardly be called a beautiful game.

To me, however, a beautiful game is number one in the sense that it is the ultimate/highest/noblest goal. Some games are only about winning (e. g. a simple game of dice), some are also to various degrees intellectually stimulating. I like go especially because there is also the added dimension of beauty.

His philosophy may be flawed (in my view it is)

Yes, you are probably right (In particular I think I should work more on my concept of beauty). In a way, I would like to be a sage like Topaz, but it is just not very easy. And sometimes perhaps not fun enough.

but his actions speak the truth - I want to win pretty, I want to win fun, but ultimately - I want to win.

I am not sure my actions would necessarily speak your truth. For example, I might prefer to lose a game by some small mistake of mine rather than win the game because of a gross mistake of my opponent and I would rather lose a complicated game, than win a game where where the board has simply been divided into two big territories. However, in a game full of big mistakes and without any room for complex tactics or strategies, I would naturally prefer to win.

[daal]

I thought, and still think, that the priorities listed above are in the wrong order, certainly winning must come first.

This is a game, with a goal, and any judgment of beauty or interest can only be made based on a moves ability to attain that goal. We can make good shape to our heart's content, but if the move does not serve to lead us down a winning path, there is no beauty, or interest, there is only misplaced artifice.

If you follow this line of argument, you must also claim that a beautiful or interesting game does not exist, because someone lost it.

I don't know that that's what you get when you follow this line of argument. Losing a game does not mean that the person who lost did not play skillfully in pursuit of a win.

I must have expressed myself poorly again, because I agree with you. I meant that winning cannot be the defining factor in determining whether a game was good beautiful or interesting. It takes two to tango, and a beautiful game or even a beautiful move is not possible without someone being on the losing end of the stick. It is perfectly possible to play a beautiful game and lose it.

[jts]

I give here a trick play which fits Mnemonic's description.

Well, I think this one fits my description. First, maybe this is just my own stupidity, but I don't see anything arrogant or punishing about trying to separate the two groups with P17. Moves like P17 are common in a lot of josekis. After move 10, nothing could convince me that black would end up with influence on the top side. At my level, reading out twenty moves is impossible. So either you've memorized the 19-point trick play in advance, or it's not a hamete - it's a reasonable reading problem for you and your opponent.

I should say that the 19-point trick play, in particular, was useful to me because it's doubly tricky. I could never read out that sequence, but also at a certain point I would have been very happy with such a large corner... if I'd fallen for it when I was 10 kyu, I would have thought I had tricked my opponent. Being told that this is a disaster for W did sharpen the way I look at joseki. But that has to do with the specific advantage that B gains through the trick. If I memorized the actual sequence of the trick, I don't think it would make me a better player... perhaps I would win a few more games for as long as I'm consigned to play other people who can only read out five moves at a time, but that's all.

[Shaddy]

It's not really arrogant, but it's very inviting to black to punish the two-space extension by cutting it. If you asked a random sdk or 1 dan who didn't know the trick what the right move would be there, I'm sure they'd cut, and after they got what they wanted then they'd fall right in the hole.

[jts]

It's not really arrogant, but it's very inviting to black to punish the two-space extension by cutting it. If you asked a random sdk or 1 dan who didn't know the trick what the right move would be there, I'm sure they'd cut, and after they got what they wanted then they'd fall right in the hole.

... right! I think you agree with me, not Mnemonic.

[entropi]

Nice discussion, interesting ideas. But I noticed that there is some divergence in the interpretation of "trick move".

I think it is worth giving my own interpretation of it when creating the poll:

A trick move is a move that you know beforehand it can be punished, but still play it with the hope that your opponent will not be able to find the punishment.

It is not risky play, it is not a probe, it is not desparete tries when there is little hope.

So again, the only characterizing property of a trick move is that you know it can be punished.

It can be "if it is punished you lose 2 points, if not, you win 50 points". To my understanding it still is a trick move if you know this when playing it.