Relying On Trick Plays

    Keywords: Strategy

Bangneki gambling strategy aside (as detailed in the wonderful novel "First Kyu"), I think that hoping a "trick" play will fool your opponent is generally a subtle Bad Habit that players fall into when playing against weaker opponents frequently. (I do it myself, but I'm trying to stop.)

I am not referring to such tricks (hamete?) as described in the KillWithABorrowedKnife page; nor do I mean such ploys as not killing a killable group right away (in the assumption that the weaker player doesn't see the danger, so you can kill it at leisure). Instead, I mean such things as making a cut you know will not work because you're pretty sure your opponent can't read that it doesn't work. See Playing with the hope that your opponent doesn't see what you're doing for a better expression of what I meant. (On further thought, maybe it's ok in an informal game, if she shows Black how she swindled him later when they analyse the game.)

The "proper" way to catch up in a high handicap game as White is to lightly make sabaki, and wait for Black's slack moves; not resort to dubious trickery which can blow up in your face if Black stumbles into the correct refutation. If Black does not make slack moves, then the handicap should be lower -- good! Black is making progress.

I think this "bad habit" hurts the one who does it more than the opponent (unlike SocksAside or ContinueWhileDozensOfPointsBehind), so perhaps it's not an ethical question. But ultimately, the player you disrespect by relying on swindles is yourself: you are saying that you don't think you're really that much stronger than the weaker player.

Yi Ch'ang-ho learned this in a game with Rui Naiwei, as described in Janice Kim's American Go Journal (Summer 2000, page 10) review of A Stone Fell from Heaven (Yi's autobiography):

...Chang-ho mentions that he considered giving up playing Go because of the disgust and regret he felt after a game he lost many years ago against Rui Naiwei. After reviewing the game over and over... Chang-ho realized that he should have lost the game, and, through a long inner voyage, admitted that he was playing as if his opponent would not play very well instead of hoping that she would play the best moves possible.
SAS: Does anyone know which game this was between Yi and Rui? My guess would be their game in the 2nd Ing Cup on 15 July 1992, as this was presumably the first time they played each other.

The corollary Good Habit must be to respect your opponent's ability.


I had hoped to spark a little debate with this page. Perhaps I should have put it in the Controversial Statements, like "Relying on Tricks is Bad." But since nobody has debated it, I've been debating with myself :-) and I have to admit, I'm not as sure of myself as I was when I first wrote this page. This is because, in a tournament, I won't actually break the rules, but I'm surely not above accepting a win by swindle; and, if (for the sake of argument) I'm way behind in a handicap game and I think my opponent can't read that the ladder doesn't work, I'm not above playing the ladder. (Often to my regret.)

-- TakeNGive (11k)


Jenny Radcliffe I agree with you entirely that it's a bad habit and was actually going to put something up myself about it. A player at my club, who I have been playing for months now at a 5-stone handicap, does this a lot - plays overplays, which I know are overplays, but I don't know and can't work out how to punish them properly. I don't learn from it, he doesn't learn from it - who gains?

Jenny Radcliffe Oh yes. And it's particularly bad, in my view, when you're miles ahead! I don't mind so much when my opponent does it to get ahead. But when he's already ahead by 60+ points, I just get disheartened by it.


If there is no debate, that's maybe because everybody agrees. Nice anecdote on Yi. --DieterVerhofstadt


I have to admit that I don't really understand the topic of the discussion. I don't know any trick moves. Do you? Can you give an example? What do you mean by a trick move?

I play in earnest. Sometimes I don't know what the honte move is in a given situation. Sometimes I get punished, sometimes I don't. Quite often I play in the opponent's territory if that does not hurt the outside, but that happens if I don't see the result clearly and think that I can get some advange of it. Sometimes I can make an invasion in the corner to set up a semeai which I would lose but which would help me to get surrounding moves in sente. Actually, it is a technique (See Squeeze, Semedori.)

But the main principle is the same as for other games. You are more likely to find a winning combination when your position is better.

What's more. My opponents make mistakes. A lot of them. That's why I win some games. I like to analyse the games I have won. It is less painful. If I see that the opponent's move was not optimal and it was me really who made a mistake I have a consolation. It might seem that I made a trick move but actually I did not.

I have to agree that if you know some moves that work against weaker opponents only it is better to forget about them. They are not fun and what is more they prevent you from looking for far better moves at this particular moment. And, yes, of course, I like aji in opponent's constructions. I don't like to play ajikeshi. --HolIgor


I have had the unpleasant experience of playing an ordinary move and having my opponent "know" it was a sleazy trick play, probably because Chinese Go books differed from Japanese Go books on the merits of the move. I was visiting the Canadian office of a group of programmers our group was collaborating with, and their boss was a Go player so we played a game. I played the joseki used in move 8 of game 3 of the 1971 Honinbo Tournament. He got angry, saying I was trying to trick him.:-( The variation shown in Dia. 2 of Iwamoto's book is somewhat tricky, but I wasn't intending to play it, just the very simple one Ishida followed in that game. I just tend to like low, solid variations, so that e.g. I often open on the 3-3 point. I still play that joseki sometimes, and wonder whether other Chinese players are offended too. Probably I should've photocopied two pages from Iwamoto afterwards and mailed them to him, "the kind of trick play that's suitable for use against Rin Kaiho," but unfortunately that didn't occur to me at the time. -- WilliamNewman


TJ: At 15k KGS, I've started noticing that "trick plays" sometimes aren't. When I'm really playing well (you know, those games where everything just seems more clear, that you have on occasion?), my opponent will sometimes say he made mistakes. On analysis, he feels that this "mistake" is falling for something tricky. However, my point is that sometimes these "trick plays" are actually NOT relying on my opponent doing something silly. A snap-back that one falls for is sometimes set up with another advantage/line of play if the snap-back is found and the group endangered by the snap-back is made to live. So, it was not a trick play after all. Those rare games where I feel like I over-reach myself, I have started to see myself in my opponents. I am recently resolved to try not to make the mistake of claiming that a few mistakes on my part were responsible for losses too soon; perhaps this way I will learn to recognize what is, instead of "tricky" stuff I fell for, actually coming up against someone with over-all superior ability, and this way I can learn from it.

Hikaru79: Sorry, but I'm going to have to disagree with a lot of what's being said here. ^^; Hey, at least it'll cause debate, right?

Quote JennyRadcliffe

plays overplays, which I know are overplays, but I don't know and can't work out how to punish them properly. I don't learn from it, he doesn't learn from it - who gains?

What good is knowing that a move is an overplay if you don't know why it's an overplay, or how to stop it? Personally, I think this is a great thing for a stronger player to be doing--setting up positions that it's quite possible for the weaker player to get ahead in, if they solve it correctly. Especially if there's a review afterward, this is, IMHO, a great way to learn.

Quote Original Author

nor do I mean such ploys as not killing a killable group right away (in the assumption that the weaker player doesn't see the danger, so you can kill it at leisure).

Now that is a bad habit. Why? Because it's misinforming beginners. They think they have a live group, and seeing a stronger player completely abandon that part of the board is certainly not going to shake them of that illusion. Certianly when you come back and kill it later, they'll realize it, but if they're new enough to not see dead groups that can be saved with one move, they are more than likely going to think it was due to some fault of their own. Another reason why this is poor practice against a beginner is that your are wrongfully taking sente away from them by playing elsewhere--if they earned it by making a group that is at least dead in sente, then just give it to them. Capture the group, show them why and how it's dead, but don't leave it there half-alive half-dead on the assumption that they'll never know.

Basicpract: Actually, until the large endgame stage of the game, it is silly to kill many "killable" groups. Killing a group requires making a gote play in a usually relatively settled area, just as saving a "killable" group does. That makes it a largely undesirable gote play (unless there are real implications to living and dying, at which point it is much better to kill it when you can, rather than risk losing bog later). Until the Early endgame, it is much better to get two moves in a row than it is to take gote and give the opponent two moves in a row....

Instead, I mean such things as making a cut you know will not work because you're pretty sure your opponent can't read that it doesn't work.

Why not? Why shouldn't you play moves because our opponents may not know how to respond to it? Like I said above, you're not doing your opponent a favor--s/he is going to have to learn how to counter moves like that sooner or later anyway. If you think you can risk it, play the move. In quite a few cases, I'm sure you'll find that you've underestimated your opponent and that they DO know how to respond to it, and then they've learned something.

Now, I am not saying that it's good to RELY on these. It's not, it will most likely lead to a sudden increase in rank, and then a sudden brick wall when people start seeing through your facade. But it's not a bad habit. If your opponent makes a move you think is a trick play, then just prove them wrong. If you can, that's punishment enough. If you can't, then there was nothing wrong with the move.

To tie this all up with an example, I was playing an even game with a player of my own rank a few months back. I had a corner group in the top right that had two eyes. However, as most people by now have probably figured out, two eyes can still die if one or both of the eyes are big enough to have a group in it. Not knowing this back then, we were in yose, and he started playing in my seemingly live group. Snickering to myself, I played some other yose move. He plays again, and I realize that the group is no longer unconditionally alive--it's now a ko for life. This could have been avoided if I'd responded to his move a turn ago, but I didn't. In the end, I won the ko, and the game, but the point is still there. Should I have sat and whined that the move was unfair because I should have known how to counter it? This was one of those games that you put under the "learning" category.

BobMcGuigan: Routinely playing "trick" moves or moves you know aren't good can subtly degrade your own play. Subconsciously you remember that such and such a move "works" and gradually you forget why it's bad. Furthermore they lead you into the "wishful thinking" trap of expecting your partner not to see what is wrong with your move. This pollutes your thinking for games with stronger players. So I would focus not on the problems for the weaker player but on the damage to your own game.

ferdi: I never use trick plays against weaker players. Against players of equal strengh, however, from time to time it's fun to challenge the opponent with trick moves in the opening!

[Diagram]
 

In my last tournament game, I played this well-known trick play against another 3d and a very entertaining game resulted.

[Diagram]
 

Or how about this one? It's not exactly a trick move, but it's rare in professional play, I think because it can be played only in special positions (see your database for details ;). But most of us are no professionals, so why not trying B6 instead of always playing at "a"?!


tderz: not relying on it, but often in handicap games I play 44PointVeryLowApproach in order to reduce frameworks. During that course one can play a hamete which is difficult to refute by a tesuji where I've forgotten the name (but the White would be punished). The advantage is small but significant, one more push in Black's area.

[Diagram]
W2 hamete  

W2 is hamete. If Black at a, one remains in calm waters, if at b it becomes interesting (first e-, then d19 become important points, not easy to spot the first time!). After the game I usually tell the other player what s/he could have done or what could have happened.

See 4463EnclosureSecondLineSideAttachment


chrise I have difficulty with the term "trick play". To me a move is a move, and depending on the level of skill of the player, will work or not.

IanDavis A trick play is one which you play intending to trick your opponent into making a mistake. The name comes from the intent.


Anonymous I don't understand what all the fuss is about with "playing trick plays against weaker players". When playing as White in a high handicap game, you pretty much rely on trick plays (at least as defined by IanDavis above). I don't see how you could possibly win a handicap game otherwise...

IanDavis I agree, learning to beat tricks is an essential part of Go, just like life and death really ;-) I think the page ought to be renamed a little. Instinctive Overplay for instance.

Bill: When I am playing a high handicap game I regard it as a teaching game. I do not rely upon trick plays. Pedagogically, I think it is much more instructive to learn how you shoot yourself in the foot than how you goof in more difficult situations.

crux: I'd define a trick play as something that loses points when the opponent correctly refutes it, and gains points over the proper play if the opponent falls for it. Like Bill, I do not rely on trick plays in handicap games, as I feel that doing so teaches both players the wrong thing. Handicap games are won because Black doesn't need inducement to make huge mistakes.

ferdi: The following position is taken from a DGS-game of mine, 2d against 8k, 2 stones, 2 komi.

[Diagram]
 

My first idea was to play W1, but after B2, if White plays W3 instead of passing, may be Black would realize that there is a seki in the corner after Wa ("Why does White play a dame point? Hm..."). And if Black defends, he will win by one point. So, after some thinking...

[Diagram]
 

...I played this W1, taking away the black liberty in a more natural manner.

[Diagram]
 

Now, if Black defends with B4, after W5 and B6, it's a one point win for Black again.

[Diagram]
 

But Black didn't think about the corner, and after W9 White was 4 points ahead.
So, was this a "A trick play which you play intending to trick your opponent into making a mistake." (IanDavis)?

I don't think so; from two ways to reach the best possible result, White played the more complicated one, hoping for a mistake. I think this is the right way to play handicap go, and it's possible to win high handicap games this way. If the weaker player makes a mistake in a position like this, it is his own fault and not that of a "trick play".

Anonymous Actually, what you've done perfectly conforms to the definition of "trick play" given above, as far as I can tell. I'm personally inclined to accept the point of view of chrise and stop worrying about so-called "trick plays" altogether. There are good moves and bad moves and that's enough.

IanDavis This is a strategy of confusing your opponent with complications. I still don't think there is anything wrong with using the odd trickplay though.

chrise Yes certainly. As I mentioned, there are no trick plays, only moves, which depend upon the skill of the player to decipher.

mencial That is outside the point. That play counts as a "trick play". You chose a suboptimal play hoping that your oponent's error would make up for it. It is a bad habit because usually it will make you lose more games than if you always went for the best move (or at least, what you think is the best move!). When going for trick plays, many times you turn your hopes into certainties...

TAPJoshua @mencial Actually, that wasn't a suboptimal play. The correct response results in the same score.

tapir: Using outright trickplays in teaching games (high handicap games are by default teaching games) makes the whole exercise pointless unless the trick involved becomes very obvious soon after played and provides a good lesson. I would blame the stagnation of quite some players entirely on this bad habit of their teachers.

Student? I thought that deception was the nature of this game, you may only profit where your opponent has losses, nobody would knowingly sacrifice pieces nor give up territory. So surely you mean it is rude to play moves hoping for a blatant mistake as opposed to a subtle mistake.

Anonymous: The point is that it is not good to play moves you know are bad, hoping your opponent won't see how to refute your move. To play go well you have to play te best (most correct) moves you can. We all make mistakes and so, therefore, do our opponents. It is the deliberate mistake played in the hope that the opponent will make a worse one that is frowned on.

Anonymous2?: I think a well thought out and executed trick play is a beautiful thing, as opposed to your garden variety overplay, because we don't know the joseki. Not so bad when you have a clock, but trick plays can make for a long game otherwise.

Anonymous4?: The definition of a trick move is a move which you know that if your opponent could answer correctly you would suffer some kind of loss. So if you try such moves, you risk losing because of them. In casual games it might be fun to try trick moves just to see how they turn out, but in serious games, as in tournaments with long time limits, trick moves are usually bad strategy unless you know you are likely to lose the game anyhow.


Anonymous: I have to admit, as a beginner, I'm very confused by this and other pages that deride "tricks", speculative invasions etc. so strongly. Mostly I'm confused by how people complain about trick moves after they lose to them; if it's a legal move according to the rules of the game, and if I predict that it will help me win, and if it does in fact lead to me winning the game, in what possible sense is it a bad move to play? On the other hand, if I make the move and it leads to my defeat, it was a bad move, but in what sense is it any more "wrong" than any other non-optimal move?

It's all very well talking about how it's bad to make a move that you know will be a loss if your opponent answers correctly, but as I understand it, "Understand the position on the board and make the best move given that position" is the game of Go. If my opponent fails to spot my "trick", that is a hole in his skill at playing Go; in that instance, I played better than him. If he spots it but fails to counter it correctly, the same. If he spots it but doesn't have a clear counter for it, it's not in any sense a trick, just a tricky play. And if he spots it and counters it, he played better than me.

I hope I don't come across as trying to strongly assert an argument! I'm sure there are answers to my issues with the concept; I just don't know them, and I want to lay out my thoughts as clearly as possible to facilitate getting an answer.


Relying On Trick Plays last edited by 196.210.57.116 on April 2, 2018 - 11:10
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
RecentChanges
StartingPoints
About
RandomPage
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Goproblems.com
Login / Prefs
Tools
Sensei's Library