Given a game between 2 expert chess players of equal quality are there any statistics about which chessmen are most and least likely to still be standing after a fully played out game? I would guess the centre pawns (d2, e2, d7, e7) would be most unlikely to survive but everything else I’m unsure about.
Well, the Kings gonna have a 50% chance of getting killed per game 
(err, actually, a little less, due to stalemates/draws, I guess)
Kings always survive. They just lose and surrender. 
Top class players don’t usually play to the end of the game. They figure out that they are down to an insurmountable deficit in either position or material and give up, or they both agree the chances of winning are small and agree to a draw. So it’s hard to say what is most likely to survive.
Perhaps glee can enlighten us if there are actual statistics like this.
I’d guess probably bishops or knights. Pawns are largely treated as expendable, and rooks and queens are powerful enough that it’s hard to lose while you still have them.
I would guess opposite about the pawns. I’ve never seen an endgame that didn’t rely on pawns. But which pawns? I would guess either the a- or h- rank pawns, or the next two inside (b- or g-).
I was considering pawns as a whole, and while a couple pawns from each side usually make it to endgame, two survivors out of eight isn’t very great odds.
Yeah, it depends on how exactly the question is phrased. If we’re considering each piece uniquely, my guess would be that either the a-,b-,g-, or h- pawn (not sure which) is present most often at the end of a game (whether that be through a resignation or checkmate.)
Yes I was thinking about individual chessmen, rather than the group they belong to. Say that 2 colossuses were to play a game with human pieces. Neither one will resign or offer a draw, when pieces are captured they throw them off a cliff but at the end of the game they leave all remaining pieces to go safely back on their way. Assuming the first 2 humans bagged the king spots, and now you can choose where you want to stand where would you like to be put?
In endgames with one white non-promoted Rook/Knight/Bishop left in, is it pretty much 50-50 which one it is, or does the queenside/kingside have some extra likelihood of survival?
Not that I think this is crucial, but for completeness, if a pawn gets promoted and is still alive at the end of the game, is the pawn considered to have survived or is that counted as a queen surviving?
The latter, I’d say. It’s no longer a pawn, but a queen.
If neither will resign or offer a draw, it’s probably impossible to know, since that’s such a rare circumstance in real games. You’d also have to know something about the playstyles: If it gets down to a classic kings-and-pawns endgame, then whichever side is able to promote first (without the queen getting immediately captured) is almost guaranteed to win, but there’s a question of how that player will win. A king and queen all by themselves are more than enough power to force a checkmate, but you might be able to get one a little quicker by promoting another pawn, as well. Does the colossus mate with a single queen, sparing the remainder of his pawns, or does he go polygamous, and toss all the original humans off the cliff? In a normal game, that wouldn’t matter, since the opponent would already have conceded, but that’s off the table.
It’s still the same chessman
Actually, I thought it was the letter of the rule that the Kings are never taken off the board. Furthermore, when was the last time you slid a piece into a king?
Id go for Bishops over Knights:
1.Knights move like no other piece, thus they have escape routes other pieces do not.
2.Knights can jump over pieces.
I thought this question was going to be more along the lines of, as Chess it self advances, what pieces will we lose? [Much like the Queen, or the piece that became the queen is no longer with us, thanks in no small part to the fact that it could only move 2 spaces max a turn.
If that is the question, and if the standard game would allow for pieces other than king to be “Royal”, then I would have to say the king would be next to go.
In speed chess it is perfectly acceptable to take the King. And that ends the game. You are not required to win by checkmate or time expiring, and you do not have to announce “Check”.
Really, the whole rule of “the king’s never actually captured” is antiquated and unnecessary. It doesn’t really make any difference at all in standard chess, but it can cause problems in many variants. Another example: In any of the chess variants with more than two players, it’s possible that even if I can’t get my king out of check, another player might do so. Alternately, even if I have the opportunity to take an opponent’s king, I might not do so, if I have more pressing demands from some other player.
In standard chess, the “don’t capture the king” is sort of necessary for stalemate to make sense.
True but not relevant. Knights have compensating weaknesses: a maximum eight-square scope, often less; comparable value to the Bishop, which means they’re frequently exchanged for them; the obligation to change the colour of their square at each move; and the inability to maintain a guard on a protected piece if they move.
On an otherwise empty board a Queen can easily round up a lone Knight, and almost as easily two Knights - even if the horsies reach a mutually defended position, they lose as soon as one of them must move.
I’d guess that King, Rook and pawns versus King, Rook and fewer pawns is the single commonest ending in serious chess.
I agree. The rooks don’t usually get deployed until later in the game than other pieces – in many games, they don’t get deployed at all except for castling.
Yeah, if it’s not one of the pawns, then it’s gotta be one of the rooks. These should be some way of getting hold of a large chess games database and query for games that end in checkmate and see what pieces remain. The problem is, as mentioned before, that I don’t think there are very many chess games at at the highest levels at all that get to checkmate. I think an analysis of ending positions even at the end of any non-draw game would be interesting, too, though.
Short answer: you want to be a wing pawn. Long answer:
The following stats are from the Chessbase “Reference Big” database of about 2.5 million games. The “All Games” numbers consider every game in the database. The “Checkmate Games” numbers consider only games where the mate was played on the board; all draws, resignations, and forfeits (time or otherwise) are ignored. A pawn that promotes is considered to still be alive.
All Games (2,533,512)
Piece Survived %
----- --------- -----
Ph2 1,884,464 74.38
Ph7 1,852,071 73.10
Pg2 1,765,545 69.69
Pa2 1,723,238 68.02
Pa7 1,705,429 67.31
Pg7 1,700,360 67.11
Pb2 1,538,838 60.74
Pf2 1,535,959 60.63
Pf7 1,497,931 59.12
Pb7 1,463,792 57.78
Rh1 1,406,376 55.51
Ra1 1,405,640 55.48
Ra8 1,404,682 55.44
Rh8 1,384,502 54.65
Qd1 1,239,266 48.91
Qd8 1,223,919 48.31
Pc2 1,117,113 44.09
Pe7 1,079,619 42.61
Pe2 930,376 36.72
Pc7 923,757 36.46
Bf1 922,610 36.42
Bf8 905,625 35.75
Bc8 875,196 34.54
Pd7 848,931 33.51
Bc1 827,133 32.65
Nb8 721,810 28.49
Nb1 709,001 27.98
Ng1 687,765 27.15
Pd2 670,781 26.48
Ng8 670,574 26.47
Checkmate Games (85,022)
Piece Survived %
----- --------- -----
Pa7 59,866 70.41
Pa2 59,655 70.16
Ph2 58,581 68.90
Ph7 56,394 66.33
Pg2 53,905 63.40
Pb2 51,598 60.69
Pg7 51,161 60.17
Pb7 50,742 59.68
Ra1 48,846 57.45
Ra8 48,810 57.41
Qd1 48,432 56.96
Qd8 47,726 56.13
Pf2 46,488 54.68
Rh1 45,883 53.97
Rh8 45,373 53.37
Pf7 43,432 51.08
Pc2 40,330 47.43
Pc7 35,348 41.58
Pe7 31,429 36.97
Pd7 28,779 33.85
Bc8 27,652 32.52
Pe2 26,670 31.37
Bf1 26,244 30.87
Bf8 25,699 30.23
Bc1 25,254 29.70
Pd2 24,357 28.65
Nb1 22,759 26.77
Nb8 21,489 25.27
Ng1 19,074 22.43
Ng8 18,558 21.83