Because he may not have blundered his Queen, as it may have been a poisoned queen.
I think it’s a stretch to call it a ‘poisoned queen’ if it’s referring to a one in a billion chance that 50 moves later black might win.
And that’s why nobody’s saying that it is poisoned, only that it might be.
This is assuming he actually blundered his queen. If it’s a po6 queen, black will either get the queen back or force.a mate (or a draw) at some point…
I accidentally posted this twice.
I no longer understand the point of your question. The OP in its original form made some sense if interpreted as “is White in a very good position here?”. The answer is “yes”. Ample evidence has been provided. Unless White makes a major blunder it doesn’t appear that it can lose.
Yet you keep claiming that maybe the queen is “poisoned”. If so, no computer analysis has been able to show any evidence for that. This strikes me as much the same as many of the disingenuous “maybe the science is all wrong” arguments, like the arguments for fast-than-light travel. Is it possible? In theory, anything is possible, but if it’s true, then our understanding of the fundamental nature of spacetime is wrong – an understanding that is supported by a wealth of evidence, and which would have to replaced by something new for which there is no evidence whatsoever. Such ruminations are usually a waste of time.
You’ve got your answer to the OP based on several different computational analyses, yet you keep insisting that maybe there’s some hidden truth here that no one can know. What’s the point?
And that’s the difference between the two analogies.
If we could explore all chess permutations and find that actually giving up a queen in this way leads to a winning game, it would be a very surprising result but it wouldn’t invalidate anything we already know.
No-one’s claiming that.
I think this tangent just started because some people were using terms like “forced win” when it’s unusual to describe an opening in those terms. So some people, including me, were just saying it’s technically, hypothetically possible (but in real terms I’d bet my life there’s nothing there for black).
I think the argument of whether “incredibly, proposterously unlikely” should just round to “impossible” is not a very important one, and not specifically a chess question.
Who said that? ETA: That is, it’s in the OP, but that’s about it. I don’t think anyone here has said we definitely know 100% that white can force a win, but it’s exceedingly unlikely. I think everyone has said, no, we can’t prove it 100%, but it’s 99.999999999% likely that white wins with perfect play.
How do you come up with this 99.9999% guess? From what little we know of chess, whites position looks better, but if we don’t know for sure if it’s wonderful, lost, or drawn how can you find out what the chances are?
I’m not sure what your issue is here. If you’re asking how pulykamell calculated an exact figure then that’s faecetious as the number of 9s were arbitary. If you’re wondering how (s)he is so certain, then it’s just a matter of common sense. I’m 99.99999999999999% certain that the koh-in-noor diamond is not in my bedside cabinet as it wasn’t there last time I looked and it would bevery unlikely for someone to just put it there. Likewise with a chess position from a general position with white a queen up and better opening position there is a vanishingly small chance that black can salvage a draw with perfect white play, so small as to not even be worth considering.
Oh gods, now someone’s going to say 99.9999999…%
We could actually come up with an estimate of probability, and I suspect that pulykamell’s figure is actually low. We’ve run the simulation many, many moves into the future, comprising a combinatorically-large number of possible games from that point, and so far, white has maintained (and increased) its substantial lead in all of them. A reasonable estimator of the probability (or at least, of a bound of the probability) of black winning after all would then be 1 divided by the number of lines examined.
Here’s how I see it: the position after 4. Nxd5 really isn’t that much different from if Black played with a queen handicap from the get-go. I think we’d all accept that in the latter situation, there’s no win for Black against flawless play by White.
But in the position we’re discussing, Black only has one piece developed, has no pawns off his second rank, and is about to have to waste a move with Kd8 to avoid losing a pawn and a rook. It’s not a position where Black’s down a queen, but has some countervailing advantages that could leave the outcome unclear against topnotch play by White. It’s just a cruddy position with no possibilities beyond those inherent in the starting position, plus Black’s down a queen and won’t be able to castle.
So contra Mijin, yes it would invalidate everything we know about chess if there was a win for Black against perfect play by White. It would all but say that Black’s chances are so much better than White’s that there’s a sure win for Black against perfect play by White, even with a Queen handicap. But for some reason, nobody’s noticed Black’s awesome inherent advantage, in all these centuries that chess has been around.
The only alternative to that reasoning is that there’s something so special about this position that it grants Black awesome inherent advantages, awesome enough to make up for that queen deficit even against perfect play by White. But no: there’s no hidden wellspring of Black power in this position.
Just out of curiosity, I checked if this position exists in chess365’s database of 3.5 million games. Surprisingly, it does. Unsurprisingly, it’s happened only once, in an under 10 game, and white won by checkmate in 17 moves.
I’m not asking a question where the answer is governed by random chance. Black can either get out of this position, or scant. If black has a winning line, he may have more than one. How do you determine what the chances are?
If you’re asking is there exhaustive proof that black cannot win or force a draw, the answer is no, or at least I think it is. It doesn’t seem to me there is enough computing power combinatorially to show that with 100% certainty. The best we can do is pit the strongest chess engines against each other in this position, and see if white wins every time. But I suspect that will not answer the question to your satisfaction.
Is your question simply one of is it possible to definitively solve this problem with the computing power currently available? I’m not entirely sure what the point of this exercise is. The answer to the question of whether white is in a good position is a resounding yes, and I expect any grandmaster when pitted against another grandmaster in this position would simply resign as black and move on to the next game.
Heck, I’d have a pretty good chance of beating Kasparov from this position, though I’m sure I’d do it very clumsily and that he’d draw out the game a lot longer than he should be able to. And the only reason I’d have only a pretty good chance, and not a guarantee, is that I’m a sloppy enough player that I do occasionally (though rarely) make a blunder that costs me as much as a queen.
Actually, this could be done relatively easily (sort of).
We assume that “perfect play” is a standard that human players (and even modern computer players) cannot reach. However, one can assume that the abilities of increasingly better players (human or computer) approach the limit of “perfect play” somewhat asymptotically. Thus, to approximate the chances under perfect play, look at the chances in competitions between two players of equal level, at different levels, and take a limit.
So, two players with Elo ratings of 1100 would make blunders, etc., allowing for a fair amount of unexpected results. Two players of Elo 1900 much fewer such mistakes; two players of Elo 2600 vanishingly small number of mistakes, etc.
I am willing to bet very large sums of money that, given a queen advantage in this position, the asymptote of results is “white wins” (p = 1.00). Thus, claiming 99.99999% chance of winning does not seem at all unlikely.
I don’t know what the odds are of white winning or drawing, but I do know that the odds are not 99.99999%. I know that the odds are either 0% or 100%.
0% and 100% are not odds. Odds are what we make given incomplete knowledge. With all that we know about chess, it is rather unlikely there is a winning or drawing solution for black. Let us assume that eventually chess will be solved and an answer can exhaustively be found in response to this question. You can certainly put odds on that.