Does anyone know of a game that was drawn under the mandatory 75 move rule?

Just that.

I looked for a while without finding anything.

I think that the rule must have come about as the result of the 269 move game played by Nikolic and Arsovic in 1989 when FIDE had changed the rule from 50 moves without a pawn move or capture to 100 moves.  It was king and rook versus king, bishop and rook.

Same here. I'd also be interested in games that finished within a few moves of being terminated under the rule.

I would bet such a game does not occur seriously in the next 100 years while both players are aware that the 50M mark was passed. The gap between the 50M and 75M rule is outrageously big. You won't find 2 players in the world both fighting to win for 25 moves without progress and without any desire to validly claim a draw - after making no progress either in the 50 preceding moves! The automatically drawing 75M rule only makes sense when the 50M claim is first removed from the rules. I believe the 75M rule now is only there to prevent players from trolling tournament organizers. Any games with the event will be fake games like similar games with hundred of moves or ridiculous amounts of underpromotions. 

I wouldn't take the bet on, but mainly because I think unless one player missed a mate in 1 on the 150th. ply the game would have terminated in a dead position before reaching it.

As for the rest, Nikolic and Arsovic didn't simply agree a draw in the game @notmtwain refers to and you could say Nikolic made progress because the KRBKR ending started off as a draw but on move 241 (and similarly after 238) Nikolic could have played

(Unfortunately he progressed backwards instead.)

In the game shown in @notmtwain's link the legend Three time repetition claimable appears off the bottom right of the board at move 102 and continues to appear for the rest of the game. Is that a bug in the site software? My understanding is the repetition can only be claimed when it is about to occur or occurs and the opportunity to claim disappears if it is not made exactly then.

Currently no time to react to all issues but this one is of special interest to me. The 75M draw will commonly be overtaken by dead reckoning but not as often as you suggest. The possibility of checkmate is just one of the conditions frustrating premature death. The other ones are: the possibility of a pawn move and the possibility of a capture.

The Nikolic - Arsovic game is actually a good example. Let's assume 241. Kc5 Rc8 242. Bc6 are played and now black plays 242. ... Rb8 and the position is drawn by 75M. No time for checkmate. Was it dead before? No? why not? Well, black could have played .... Rxc6+ resetting the 75M count. Both walk, no one dies!

A highly significant anecdote. 3REP - automated in the composition environment - is more potent in demonstrating these issues than 75M. I actually created a magnificent problem ending in a 3REP draw (somewhere near 2007). And then to my disgust I discovered the solution was flawed by premature death. I reported it to the editor of the magazine who supposedly forwarded it to the judges. To no avail. In spite of my resistance it was awarded 2nd prize! Though it was all in the rules, nobody was much impressed with the idea of applying all rules simultaneously rather than one at the time. This is common amongst legislators. They only think one lego piece at the time. The rest is case law. I hope you understand the significance of this example. It predicts something about your chances of convincing FIDE and WFCC officials about the formal implications of the current rules. God forgive them, they do not know what they have done.

@Arisktotle

Yes, of course you're right. Didn't think of that.

I'd be surprised if much software handles it correctly.

I don't see any reason why most positions would be dead before reaching 75 moves. Most drawn positions are only a draw with best play and there are plenty positions where you can make 75 moves without capture or pawn move which are not dead.

And I don't understand your example. If the position is dead, then you don't need to apply the repetition rule. You just reach a draw due to dead position before the repetition occurs.

Edit: I get it now, just the story after the example is weird, because dead draw and repetition rule do not interfere with each other. You can't have 3x repetition in a dead position, because the game is already drawn the first time the position appears on the board.

Indeed, they do not interfere on the result but they occur in positions 1 move apart. I didn't give the context of the stipulation so here it is for clarification:

The problem was a diagram with the stipulation "SPG, drawn in 41.5 moves". The intention was to let the solver find a draw by repetition resulting in the diagram after 41.5 moves from the start of the game. However, this is impossible because the postion was dead after 41.0 moves. So the shortest proof game leading to the diagram (while drawn) does not exist.

Note that draws by agreement do not exist in chess compositions; neither by claim when the claim rules are replaced with automatic draws. The automatic drawing conventions are pretty similar to the new competition rules on 5REP and 75M.

Another note: The position is only dead if there is no choice in the next move manifesting the 3REP.

Third note: Just as dangerous as the premature death is the premature repetition. It occurs when the position preceding a 3REP is itself a 3REP. This is very common. It is possible to make interesting proof games featuring both premature death and premature repetition plus a decent solution! I let this research go at the time because I was drawn to other areas of retro logic.

I realize that you are using a very awkward definition of a 'dead' position. If you were to use the term like it is commonly used, then you wouldn't be having these problems with 3 REP etc. Similarly like you can't win on time in a dead position, you also can't draw by 3 REP or 75M when you are checking if a position is dead. The 3REP and 75M are not basic rules and should not be used for 'deadness'. The goal of these rules is to achieve draws in clearly drawn positions. The goal is not to create more dead positions. That is why the dead draw rule says ANY series of legal moves. These rules are means to terminate a game which goes on forever and achieve the result which is considered fairest.

I suggest a new term for Arisktotle's view of a dead position. Let's call it a complicated dead position: The game is drawn when a position has arisen in which neither player can checkmate the opponent’s king with a series of legal moves which can be played in accordance with all rules of the game if the game would continue from that position.

Not only do you have to check for 5REP and 75M, but also if players have enough time to execute the checkmate. If you are the only one who can checkmate and you got 1 second left on the clock and the shortest checkmate takes 10 moves, then it is a draw due to complicated dead position because you can't physically make 10 moves in 1 second.

Though something is to be said for your views, you should be careful to describe mine as "awkward" as if there is an existing standard for this. The automatic draws are a recent addition to game laws and no one says that dead reckoning ever had an effect on claimed draws. However, automaticities existed in composition chess for decades and it is the only place where you will find any expertise on the subject. I was very happy last year to finally convince (I think) Andrew Buchanan of my views who is the father of dead reackoning in composition chess.

As far as I can see, MARattigan runs on my logic, since his approach to 75M is identical to mine in 3REP. You can't be on different sides in these two issues.

Essentially the understanding of analytical "legal moves" is the same as the understanding of "i" as the base imaginary number in arithmetic. You use it in analyis but in many applications the imaginaries have to disappear when the real outcomes are calculated. It is part of my 2000 page treatise on this subject which I promised to Andrew, but will probably never complete.

Note: what players can calculate doesn't matter. A position is dead or not irrespective of anyone making any calculation. In the future, engines will run alongside the players and declare a position dead, often in hindsight.

@Numquam @Arisktotle

The very awkward definition is courtesy of FIDE. You can't blame Arisktotle. 

If FIDE call it a dead position, then you have to suggest a new term for what you say is commonly understood to be a dead position, perhaps "a commonly misunderstood dead position".

It's a universal common convention (for all games) that all rules of the game apply unless otherwise stated in a particular circumstance. 

For example:

doesn't say:

but it's necessary to assume that, otherwise it would be OK to move your knight on top of your king.

Your example does not apply to the rule I wrote, because the main addition is that we look at what happens if the game is played out under the tournament rules. " if the game would continue from that position." In the existing rule, that doesn't happen so whether the 75M or 3 REP applies to the game is irrelevant.

I am sure the term 'dead position' has been used a long time before FIDE made the 75M and 5REP rules. Let's say you show some GM a position from a tournament game and ask him if the position is dead. Do you think that the GM would say: that depends on how many times position x has occurred? Of course not, that is not what people commonly perceive as dead position. You can't even use the term 'dead position' like you use it, because a position does not include its move history. You would have to say something like the game reached a 'dead' state. Usage of 'dead position' alone suggests that the move sequence leading to the position does not matter.

What I'm saying is that your addition is redundant, because it is only adding what I referred to as a universal shared convention about rules of games in general. It's exactly what is to be understood by the rule as it stands, just as my addition to the knights move rule adds nothing.

It's patently obvious that neither player can checkmate the opponent’s king with any series of legal moves if no such sequence can be played without contravening the rules in force. If the Competition Rules are in force the 75M and 5 REP rules are in force.

The competition rules are not in force for article 5.2.2. What you are trying to achieve can only be formulated precisely if you use a different dead draw rule for tournament games. I'll give a better formulation, my previous one was too generic and can give confusion. 

The game is drawn when a position has arisen in which neither player can checkmate the opponent’s king with a series of legal moves such that no draw is reached under article 9.6 while playing this sequence out.

The addition in accordance with all rules of the game. is not a universal convention. The universal convention is to use all rules which are explicitly mentioned and make as few assumptions as possible. You can't change previous rules by adding more rules. In that case you'd have to change the previous rules like I did with 5.2.2.

Competition rules are in force for the game, but that doesn't mean they are necessarily in force for a specific article, in this case 5.2.2.

Competition Rules are indeed not in force for articles 1.4 and 3.6, but that doesn't mean you can play Nf7#. 1.4 is about the objective of the game and 3.6 tells us what knight moves can be played if it is your move. However the game can terminate before the knight move can be played. That does not mean tournament rules are used to determine legal knight moves. The game simply terminates before any knight move can be played.