J@jk_182 said in #1:
> Comments on https://lichess.org/@/jk_182/blog/do-attacking-players-play-more-forcing-moves/1maN1mgD
In my opinion, forcing moves aren't about what style player use, it's about :
The positions and what player chose.
When there is a position to check, and the opponent have no other choice but to use a piece to block the check, that's a forced move, and it depends on how the board is, and position of your pieces. if your peices aren't in position to check from the angle that will force your opponent to block by a piece, and if you check, they will have more than one choice, that's not a forced move, and it don't depends on how a player play chess, it depends on how the board is, and position of your peices!
But style can be effective, as a player in attacking style might move a piece to another square to make a forced move, but it's about what player decides to do, and maybe an attacking style player don't move his/her piece to make a forced move, so it depends on what player do.
(This is a personal opinion!)
@JudahH said in #10:
> "The standard definition of a forcing move is a check, a capture or a threat."
>
> How is this a good definition of a forcing move at all, let alone the "standard" one? Would anyone call it a forcing move if the opponent has the same set of good moves to choose from before and after? At best this is a proxy for forcing moves, and not even an especially good one.
Standard definition may be the wrong term, but it's the only general definition I've seen. And these are exactly the moves that (in general) reduce the number of good options for the opponent so I think it's a good general way to describe forcing moves, especially when using a more inclusive definition of threat.
I don't know if there is any better general definition, as "reducing the number of good options" is also difficult to pin down exactly. For example, one player can make a mistake in an equal position and the opponent has one winning line and several other moves that are equal. Would you then count the mistake as a forcing move because the opponent has only one winning move?
@FlameBreaker said in #11:
> J@jk_182 said in #1:
> In my opinion, forcing moves aren't about what style player use, it's about :
> The positions and what player chose.
> When there is a position to check, and the opponent have no other choice but to use a piece to block the check, that's a forced move, and it depends on how the board is, and position of your pieces. if your peices aren't in position to check from the angle that will force your opponent to block by a piece, and if you check, they will have more than one choice, that's not a forced move, and it don't depends on how a player play chess, it depends on how the board is, and position of your peices!
> But style can be effective, as a player in attacking style might move a piece to another square to make a forced move, but it's about what player decides to do, and maybe an attacking style player don't move his/her piece to make a forced move, so it depends on what player do.
> (This is a personal opinion!)
I agree that the choice of move depends a lot on the position, but there often situations where players can choose between different ways of playing a given position.
I wanted to see if players that have a more attacking style prefer to play forcing moves compared to more positional players, which seems to be the case. But it's difficult to determine if the actual forcing moves are the difference or if attacking players end up in positions where everyone would play forcing moves.
@jk_182 said in #12:
> Standard definition may be the wrong term, but it's the only general definition I've seen. And these are exactly the moves that (in general) reduce the number of good options for the opponent so I think it's a good general way to describe forcing moves, especially when using a more inclusive definition of threat.
> I don't know if there is any better general definition, as "reducing the number of good options" is also difficult to pin down exactly. For example, one player can make a mistake in an equal position and the opponent has one winning line and several other moves that are equal. Would you then count the mistake as a forcing move because the opponent has only one winning move?
Let's grant that it's difficult to pin down the meaning of "forcing move" in a way that's conducive to compiling statistics and that your "definition" is somewhat helpful to that end. I'd still like to call it a "proxy" instead of a "definition", for the reason that it's easy to come up with cases of "a check, a capture or a threat" that I think most of us would agree are definitely *not* forcing moves. A particular common case like that is the last capture in a sequence of captures.
I agree that the basic meaning of "forcing move" is both too subjective and too imprecise enough to serve as an objective classifier: one player might class a given move as "forcing" while another classes it as "non-forcing" and a third equivocates. I have no issue with referring to any objective classifier of forcing moves that essentially accords with this basic meaning as a "definition". (By "essentially accords", I mean doesn't classify anything in a way that most players would agree is CLEARLY wrong.) The disagreement between such an objective definition and players' classifications would then be no more than the disagreement among players' classifications—i.e. it would simply reflect the imprecision inherent in the term.
The reason I insist on calling your particular formulation a "proxy" instead is that I believe there are many cases (such as the kind I described above) where almost anyone would agree that it misclassifies moves, effectively adding noise to the statistics you use it to compile.
Even as it stands, I agree that your heuristic is better than nothing, but—even given the practical limitation of being simple and precise enough to use for compiling statistics en masse—I think it stands to be improved (e.g. by replacing "capture" with "capture followed immediately by another capture").
How is the rate of forcing moves calculated, exactly? I guess it must be something like "forcing moves divided by the total number of moves played", but there are certain subtleties that I would like to know if were taken into account. For instance, are you starting the count after the opening is finished? (If not, would "e4-e5, Nf3" be considered a forcing move?) Are replies to forcing moves played by these players discounted from the total number of moves?
This is not an easy analysis to make, but, to make it clearer, maybe the rate should be calculated by first identifying key situations (positions) where different plausible moves (forcing and not forcing) are available (by plausible, I mean the objective evaluation of the position wouldn't be significantly different by choosing one path or the other). Since you have already figured out how to identify forcing moves, now it would be a matter of identifying those positions where forcing and not forcing moves are available, leading to roughly the same objective evaluation.
What is the sample size? How were the games chosen? Which differences between the players are statistically significant?