Can you capture the last piece in chess?

It should end as a insufficient material draw as soon as the position appears.

(Btw, Martin Stahl is wrong until FIDE fires its "experts". But he may be right by the rules of chess.com which are not the FIDE rules. FIDE has no "insufficient material" rules).

Not long ago I would have said it is a dead position by article 5.2.2. in the FIDE handbook which halts the game immediately but a bunch of FIDE experts have thrown confusion in the interpretation of the dead position rule. They claimed that the game continues when it is bound to finish in one of the known draw states: "5-fold position repetition", "75M-without capture or pawn move" or stalemate. The idea is apparently that the dead position rule only targets positions where the two sides could continue the game indefinitely if they wanted to. However, since 2018 the latter is impossible as all chess games end in either checkmate, or one of the 3 listed draw states.

If the verdict of the "experts" stands, then the dead position rule effectively no longer exist and you have to continue the game. However when the flag on either side falls then both sides still get half a point since neither side can checkmate the other!

Yet there is one difference. Without dead position rule you lose when you resign the game, even when you can't be mated. But with dead position rule before 2018 you would still get a draw rewarded after you resign in a dead position.

To avoid the risk of being shamed for not understanding the FIDE rules you are advised to emigrate to Mars, the nearest planet which does not recognize FIDE's authority on the chess rules!

5.2.2 ends the game. If mate isn't possible, there's no sense in continuing. Is there any FIDE published rule that says that the game should otherwise go on? I'm not aware of any

'5.2.2 The game is drawn when a position has arisen in which neither player can checkmate the opponent’s king with any series of legal moves. The game is said to end in a ‘dead position’. This immediately ends the game, provided that the move producing the position was in accordance with Article 3 and Articles 4.2 – 4.7.'
https://handbook.fide.com/chapter/E012023

However, you can capture the last piece, provided that piece is either a queen, or a rook, or a pawn.

I thought so! Yes, you missed the discussion between MARattigan, Jetoba and the "FIDE-experts" some months ago. The example entered was a position that was expected to end on the 74rd of 75 moves without pawn move or capture by "death" due to the impossibility to avoid the mandatory draw end on move 75.

It's not about the FIDE DP rule itself but about the apparent room for interpretation. In say 2015 everybody looked at 2.5.2. as a sort of anti-trolling rule which would end the infinite shuffling of pieces and the players refusing to claim. In that context there was no reason to declare a dead position before an inevitable (but short term) stalemate and the "experts"sort of concluded the rule was only there to stop infinite continuation. Already in 2015 they would not have declared a dead position for a sure stalemate 2 moves later - though they probably should by strict interpretation of 2.5.2. And things got worse in 2018 when the fixed draw 5rep and 75M rules were added to the laws. Using the same concept - only DP on possibly infinite sequences - suddenly all ground under the DP state had collapsed. Suddenly the chess system was made mathematically complete and DP would never kick in - by their interpretation of course!

FYI the composition domain is sleepwalking as well - in the sense it never noticed the other interpretation option. They even produced a DP-solving computer program based on "your" (and their) interpretation of DP.

I am somewhat amused since I won a second prize in 2003 with a proof game composition based on a refined distinction of the 2 interpretations of the death rule. Nobody noticed anything though the composition would have been considered flawed by half the chess community. But I already saw the 2018 extensions coming and wanted to witness the clash of the titans. Apparently it's now happening!

It doesn't matter what experts online are posting. If they are licensed FIDE arbiters, that's one thing but the rules are very clear, at least in my reading, and very explicit. The game ends in a draw as soon mate is impossible. There's no room for interpretation.

You'll have to check that with @jetoba. He is not only a certified arbiter, he has access to all the authorized case law verdicts. My understanding is that the issue was discussed during a physical meeting of referees (we had to wait a few weeks) but I can't say that for sure. The differences between a Zoom-meeting and a physical one are negligible today. Is your authorization somehow higher than his?

And there are things recorded in the referees case law which are not in the rules. Jetoba showed us e.g. that players can't resign anymore when the players continue after DP. They still get the draw. Very useful to know but not in the public laws.

I'm not a FIDE arbiter but unless the interpretation is officially published, it would seem as the public rules would take precedence and what most arbiters would rule on.

It doesn't make any logical sense to continue playing on in such an instance as the same result would happen regardless of the clock situation. It's pointless to continue playing to reach some other draw condition in a case like that.

Actually, I sympathize with your "mathematically logical" viewpoint as does MARattigan and - to some degree - the composition domain. There is a lot more to the apparently innocent DP drawing issues though.

First of all, from the day the DP rule was introduced (around 1990? have to look that up) many FIDE games will have ended in stalemate that shouldn't have. It is not uncommon that the last move before a stalemate is forced - like a forced capture after a check. The forced stalemate should never occur as a dead position should have been pronounced on the move before it. I bet all my millions on the Bahama's that no arbiter ever noticed this.

What probably did happen a few times as well is that a player shook the hand of his opponent believing he was checkmated while he could have captured his opponents checking piece leading to stalemate. When he noticed it a few minutes later he claimed a draw which was refused by the arbiter on the basis that he had already "resigned". But by the clarifications shown to us by @jetoba the arbiter blundered as the position in which the hands were shaken was already dead. Nothing after death can change the draw outcome except the progression of the tournament to such a point that changing it would mess up the preparations for the next round or similar practical obstacle.

Another issue which I will not expand on here as it may lead to an extensive discussion on definitions is that the formulation of the DP rule is poor due to the fact that "legal moves" are confused with "well-formed moves". The FIDE-rules ignore the mathematical concept of well-formedness which leads to nonsense in phrases such as "series of legal moves" in articles like DP.

Interestingly, DP has lead to more issues in the composition world and was recently cancelled for all compositions except retro-problems. The 50-move rule was already cancelled and the 75-move version and the 5rep were never adopted so the only remaining defense against infinity is the 3rep drawing rule. Fortunately in the composition domain it was already converted from a claimed draw rule to an automatic draw rule which assures mathematical completeness for the composition domain.

Plus the issue of jetoba's "experts" and the thousands of arbiters who never had a clue of what they are dealing with in the DP-rule! Jetoba's first reaction was that he disliked the DP rule and wanted it removed from the FIDE-laws altogether. And it can be done since the 75M and 5rep drawing rules assure every game will end.

'This immediately ends the game' How much clearer can they state it?

Stalemate and dead positions are different things and different rules. While it may be possible for a dead position be some forced stalemating line, dead positions, in and of themselves, are not stalemates.

As to someone resigning in a dead position, unless the Arbiter verified the position, the agreed game end result stands as far as I'm aware

That depends. If the aim is to always end the game immediately, then indeed, this description is perfect. But if the aim is to end the game immediately in the overwhelming amount of cases, then the same description may still be chosen by the scribes - on the basis that there are obvious exceptions presented in the other articles in the FIDE laws. The latter appears to be the viewpoint of jetoba's "FIDE experts". Like, the DP rule will only be invoked when there is no other law that will unavoidably draw the game within in a finite number of moves - e.g. the automatic 5-fold repetition draw law.

This weird issue has come into existence because not all FIDE laws were written in a single block but some were added incrementally leading to inconsistencies. The original DP-rule had no trouble with the existing 3rep and 50M rules since they didn't contain forced draws (only optional claims) but things went wrong in 2018 when 75M and 5rep with automatic draws were included. The effects of these rules on the DP diagnosis was unforeseen as they were added for a different reason altogether. Meanwhile, the whole FIDE, all arbiters and all players had overlooked the DP effects on the existing stalemate rule which already existed for some decades. An immense blunder which would lead to the beheading of the complete FIDE organization if it were politics and not just a game.

You missed the message. It's all about the position before the stalemate but where stalemate is unavoidable. Btw, countless retro-compositions have already been produced with this mechanism since the stalemate rule is the same for game and composition domain.

Example (under FIDE game rules):

The position in the OP and forced stalemate positions are different. I would stay the latter should be played out and if one of the players runs out of time, at that point it would be declared a draw as mate is not possible by any series of legal moves.

I would not call that position a dead position by 5.2.2.

@12

'5.1.2 The game is lost by the player who declares he/she resigns (this immediately ends the game), unless the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves. In this case the result of the game is a draw.'
https://handbook.fide.com/chapter/E012023

We were already far beyond the OP's diagram which has nothing to do with the stalemate discussion. Your stalemate comment could only have made sense in the context of my discussion of the various draw states and DP in later messages. Your last message proves you are 100% in the corner of jetoba's FIDE "experts" and against the mathematical logic supported by @MARattigan, me, the composition community and even @tygxc.

What you missed is that your viewpoint implies there are no dead positions. All positions that can't lead to checkmate will lead to an unavoidable stalemate, 75M draw or 5rep draw. It doesn't matter which one of the three, and it doesn't matter when that happens - on the next move or after 20 moves. It will happen.

You're on top of the Institution wall. You can jump off on either side. Be careful what you choose

@12 and @16. Both wrong! I have seen the clarification statements for the referees about resigning after DP and it supersedes the original resignation rules. And it is quite reasonable to assume that you can't finish the same game twice - once by 5.2.2 and once by resignation You can ask @jetoba for the link. I have it somewhere but my archives are not so great.

Wake up boys! It's the 21st century!

They will end in one of those but the game should not be played completely out in completely locked positions or ones where there would be insufficient material to mate on timeout. Ideally the players are going to immediately recognize that situation and end the game themselves as a draw but the arbiter should as well in light of 5.2.2 and in the absence of an official published ruling otherwise. There's zero reason to waste time at the board in those positions.

I see forced stalemates as slightly different, especially since there may be conditions where it's not immediately obvious so they should be played out, unless the players realize and just end the game as a draw in the latter case an arbiter doesn't need to get involved until the end anyway.

Again, I'm not a FIDE arbiter but once the game results are agreed upon and reported to the arbiter I don't believe they are normally allowed to be changed, which is what I previously said.