What's the quickest possible draw?

Okay, thanks for these extra examples. More remarkable stuff from Labelle!

Excellent problem! My first awarded triple repetition problem was built of the same ingredients. Not as a solution though, but as a cook. The stip was to "game over - drawn" the 3rep diagram but the diagram was drawn by DP one move earlier! I reported it myself but the PB editor let it go as the awareness of those interactions was low around that time (somewhere near 2009). Also I had a feeling that FIDE would still modify the interactions as they had no clue what they were doing. But @jetoba will make them look at it! The 2015 Codex change would have saved my creation had it eliminated (by default) the dead position rule for all compositions - but unfortunately the PG is of the retro-type. So I was bitten and severely injured by my ow favorite chihuahua (huahuahuahua)!

But ... as a problem constructed as a DP /3rep problem your creation is of superior quality - especially with the nice way in which all alternative repeat attempts are refuted by "loss of castling right" and the e.p. trick!

Btw, this is not the ultimate complex interaction. What Is still available is the combination with "premature repetition" which would give us (3rep +DP + premrep + others?). They don't bite another but I never completed the PG I started on when I fell ill. You are welcome to fill it in!

@anselan Amazing stuff! Thanks.

@arisktotle & @Hans_GOAT_Niemann thanks for your kind words.

I'm sorry for posting so much content in a rush, but there was a lot to respond to from the extensive email thread. I would like to assemble an article about this, and if I have time maybe I can do it in time for the next issue of our online chess magazine, for next week! There are other compositions to show, and a chance to respond on the questions arisktotle raises. However, I must finish other articles first.

In the meantime, here are three recent links:

(1) if you want to follow up on maximum number of pieces in DP, I recently posted (as Laska, another nickname) here: https://chess.stackexchange.com/questions/18699/maximum-number-of-pieces-on-board-in-dead-position

(2) The stipulation hdp (analogous to h# and h=) invented by Per Olin has proved fertile, and 52 such problems using promotion were published 6 months ago. See https://www.thehoppermagazine.com/gallery

(3) Miguel Ambrona's superb engine for determining DP status, and solving many classes of otherwise-unenginable problem, including help-retractors and hdp, is at https://github.com/miguel-ambrona/deadpos

So deadness is very much alive!

@anselan: Thx for the links!

Here is one you'd want to read related to the status of 5R/75M deadness in FIDE when players play on and on and on ... It is a clarification for arbiters on what to do next, also when players resign or a flag falls. Very sane! https://rcc.fide.com/2021/11/08/questions-answers-november-2021/

Why does it matter? This is part of the FIDE queries placed around the deadness rule, most critically the ones @jetoba will add to it on the dependencies between death and 5R/75M. I wouldn't be surprised if FIDE decided to cancel the dependencies by law - like the dead rule would no longer be permitted to look at automatic 5R/75M termination (and perhaps stalemate)! That would remove the fundamental justification to retain those dependencies in the Codex. Our kids would ask "where does that strange deadness rule in the Codex come from? In games it is much simpler! Or worse, FIDE might cancel the DP-rule altogether and it truly looks like a random object falling from the skies in the implementation of orthodox chess in the Codex!

Reality check: FIDE arbiter @jetoba actually believed that the dependencies discussed do not exist until MARattigan's arguments placed him in doubt. Which might result in FIDE providing a clarification to that effect for all arbiters. They probably would not even perceive it as a rule change!

As always, just some thoughts

Just in case they meant it was a trick question. Sit down at board. Shake hands. Go home. That's fastest draw. 🤪

@anselan: That's amazing! While I was writing my last post, @jetoba came back with an answer on his query to FIDE officials. And it is the (first) one I predicted though I don't know if he takes it as final. So they "believe" the rules say (always said) that the dead rule does not look at unavoidable collisions with 5R/75M draw lines in evaluations of the DP-rule. Here is @jetoba's short answer: https://www.chess.com/forum/view/endgames/i-hate-the-threefold-repetition-rule?page=23#comment-98123317

Which effectively implies that FIDE says that the Codex interpretation of the DP-rule for 3R and 50M - which were the forerunners of 75M and 5R only with lower numbers - is conceptually wrong. Not legally of course, since the WFCC is autonomous and could redefine the whole of "orthodox" chess if it wanted to.

FIDE is wrong in this matter but I know why it happened. I leave you to ponder on it such that you can prepare for the shock when I tell you (later).

Btw, FIDE's earlier responses to the query on what happens after DP's were passed (and ignored) in a game are still wise and well-behaved!

The good news: when FIDE's interpretation is clarified/implemented in the Codex then my 2009 repetition PG is sound again! The bad news: compositions such as the superior last one I commented on, will have to be discarded as unsound.

In the Manhattan Open Tournament in NYC that I played in a couple times and used to watch even when not playing in it, there was a time when this guy who was guaranteed at least 2nd place and likely 1st place after the first five rounds, supposedly offered a draw to his 6th round opponent after like 10-20 moves because that guaranteed him first and the draw guaranteed the guy for 2nd (where as if he lost he wouldn't have gotten anything due to lower ranked players winning). So it was basically like a deal, the 1st guy was far enough ahead that even if his opponent won he still may have only gotten 2nd so it was like a little deal they made, not sure what the legality of this/minimum moves are.

Now just wait for it to be played OTB as a novelty

Nf3, Nf6 three times, 4 moves

Repeats and stalemates are specifically excluded by the OP!

50 move rule then, where the knights wander around the board without pawns moving or pieces being captured

That will take about 50.0 moves though. OP requested the quickest draw, which is currently thought to be 15.0. No way to avoid it: this question demands actual thinking, I'm afraid

Bruh only in 4

And only 3 posts after it was spelled out that it's not of any interest.

The players agree on a draw before the game.

it is not possible to do that until the game has started

On the contrary, it's rumoured to have happened quite often.

what if white flags? its a draw

Not under FIDE laws.