safety play
An underplay that is
- deliberate, and
- intended to help win the game by removing one's own bad aji, or in some other way simplifying the position.
but see also [1]
For example, in a competition game in a situation of time pressure, one may play in a way that probably loses points but seems on the basis of positional judgement still to leave one clearly ahead.
Such plays are always deprecated, from a coaching point of view. There is no way that underplays are safe, as a habit. It may be that one cannot read everything; but getting into positions that are well beyond what one can read is deprecated, too. For example, a safety play may be used definitely to prevent an invasion of one's territory. The need for one might have been caused by an earlier mistake, such as trying to take a big corner that is really too big, and will always be subject to bad aji.
On the other hand choosing a variation that simplifies into a clearly won endgame position, over another one that is complex but might lead to a larger win, cannot always be called a safety play. For example an exchange that brings the middle game to an end may theoretically lose points: but really there is no obligation on players to try to win by the largest possible margin in positions that are genuinely too complex to read out or count. There may be many ways to end a ko fight: you aren't always required to choose one that allows your opponent to fight on effectively.
Therefore there is perhaps a psychological element here: not all inartistic practical plays based on trying to win the game with low risk are safety plays that should carry a stigma. Maybe there could be a point that amateurishness has to be involved, too: the low-grade safety play falls short of being honte.
Interesting...
But does not the proverb state: "Rich men should not pick fights."?
Contrast with Contract Bridge where frequent safety plays are a hallmark of expert players.
The Bridge definition being: "A play that guarantees the selected result by sacrificing the possibility of obtaining a better result.".
Charles Bridge is a game of incomplete information. That's a big difference: there is skill in making the contract however badly the cards you cannot place break against you. Possible overtricks that put the contract at risk may be given up; but the scoring system must favour safety. I don't know much about bridge, but I wonder whether the comment applies equally to duplicate.
tb Quite right that it depends on the scoring system. In rubber bridge, or IMPs, or systems where your score is roughly proportional to the number of points you win on a hand, safety plays that guarantee your contract but abandon the hope of an overtrick are very wise. In matchpoints, however, where the size of a score doesn't matter, but only whether it is bigger or smaller than others, you should play for every overtrick that has a better than even chance of making, even at the risk of your contract. (The only exceptions are where you think it is very unlikely that many others will be in the same contract as you.)
Alex Weldon: Like the anonymous writer above, I'd also say that playing these "safety plays" is good, at least if your goal is to win that game, rather than to gain experience (if it's just a game for fun, perhaps seeing what your opponent can do to you will be more instructive.) Especially on IGS, where an opponent who is losing badly will often make a bunch of unreasonable invasions in the endgame, hoping I'll make a mistake, I will occasionally play a move in the endgame that is probably -1 point gote, if I'm clearly winning, and doing so will remove any chance of me screwing up and losing a group and thus the game.
To quote Winston Churchill, "Victory. Victory at all costs."
Charles Oh, players can always justify safety plays to themselves. Some amateur opponents are a nuisance, and just fool around when they have clearly lost: it is tempting to close down the game against such people with a succession of safety plays. Lacks fighting spirit, though.
Alex Weldon: I guess it comes down to what you think the important thing is; winning, or artistry. Certainly, safety plays are ugly and inartistic... but if they guarantee a win where it would otherwise be possible to make a mistake, I don't see the problem, being a pragmatist myself. I guess you're a purist, which is fine too. :)
Charles Writing on this page has made me think further. I suppose that if fear of the opponent enters as a reason, then the criticism is more just. To improve one must certainly learn to trust one's own reading, rather than back down for no real reason. See also fear.
DaveSigaty: I think one issue that we all have to face in practical play is whether our "safety first" moves actually meet the criteria above for a true safety play. Too often our plays fail to help win the game because they are actually wasted moves that we would realize are unnecessary if we understood the game better :-)
Timing seems critical
- If a player makes a supposed safety play at the correct time (aka when its value suggests the safety play is the largest play on the board), then it is not a safety play.
Context seems critical
- It may not be proper or professional, but seems common (at my level) to balance out previous overplays, with later safety plays. And if the overplays work out well enough, who needs the safety plays?
This is a terrible way to play. After your opponent's correct defense exposes the thinness of your previous overplays you need to waste a move to fix the problem you made for yourself. Yuk.
[1]
Alternate "Safety Play" definition:
James Davies calls a "Safety Play" something entirely different. In his usage, a safety play is a tesuji that eliminates all possibility of resistance, where lesser moves either lead to ko or fail outright. An example from Tesuji (from the chapter "When Liberties Count":
is the "safety play". Now white wins without ko. In contrast, if white plays at a instead, then after black throws in at 1, white has to win an approach ko.