(See, I don't think it is! I think this is exactly in fact how you get individuals to display group behavior in a natural population!)
Wanton Destruction!

In #76: re: "benefit more": if nothing else than considering time spent playing losing positions. This temporal consideration drives a lot of this quantification in the game theory, I have noticed.

I consider this to be a game of wontons because he drops a piece, I drop a piece, he drops a piece, but towards the last, I make a (blush) fairly sweet play with a seemingly hung knight that, nope, quite isn't! Enjoy!

And here, the blood bath ends with me passively sacrificing both knights like paired bait (while up) for the easy snap mate. Check, position, material, in that order, every time.

Check out this possible Ma16 with continuous checks beginning with 26. Nf3! (He resigned, down 6 pts., before it happened.)

I found a flaw in the 16 chain (still makes it to 13 though, so far; some tough analysis actually):
39. Ka1!
(then e.g., 39... Nd3
40. Qc2 Qxf3
41. Rb1)
On other hand: I'm pretty sure I can back up once and not draw, here, right?
39. Ka1 Nb3+ (threatens but does not hit draw with 37. Ibid.)
Though where the knight would go after that would be? Is there some sac I'm missing to keep it going? Only need 3 or more checks. Maybe impossible with 39. Ka1 but checking can be a bit exhausting.
Since you have disabled non friends from contacting or posting anything, I'll just leave this here. About our drawn game and your response on my page about having no respect: I was waiting for the inevitable stalemate. You obviously showed you couldn't close. You could have easily, and I mean EASILY, checkmated me FAR sooner. That wasn't some 'mouse slip'. You just can't play end game. My king was alone with a few pawns for 20 moves. 20! You had two rooks and a bishop! Don't post garbage on my page when YOU are the one that lost it for yourself.

Whatever. Tired and bored with the ridiculousness of your position, I was waiting for you to resign. For all those 20 moves and more. You got forked, straight up, bungling your silly queen in the open to a checkfork early. In move 15. You clearly were in it for the remainder of the game out of spite. Have a little respect for other players instead of sitting there pouting when you've clearly lost. Do you like it when people do it to you? Then to come here and start smack is just more spite. Since you seem to want this game to be posted, I'll post it for you. It's appropos to this forum topic since this one is about the shame and frustration of people who just won't resign. This is my last response to this person. I have no further interest. Timewasters like to keep wasting time, and I won't oblige this butthat further (I type 300 wpm). Blocked.

You clearly were in it for the remainder of the game out of spite. Have a little respect for other players instead of sitting there pouting when you've clearly lost. Do you like it when people do it to you?
You may like it or not, but people almost never resign early in blitz. The reason is that time is an inherent part of the game. It's not sufficient to just get a won position; you should be able to checkmate within the limited time, which not everyone can.
You didn't have to take all opponent's pawns and queen your own when you already had an extra pair of rooks. Say, at move 34, a simpler mating plan consuming little clock time is Rf8-e8-e3, Rd2 setting up a lawnmower mate (or, if the king runs to g4, the lawnmower will be on the 5th and 6th ranks).
Another upside of setting up a mating net while leaving the opponent with a few unblocked pawns is that it reduces the stalemate risk - those pawns will be able to move even when the king already can't. (Recall that K+N+N vs K is a draw if the naked king doesn't blunder, but K+N+N vs K+P is sometimes a win.)
Also worth noting is that 25. Rhe1!? was a little (dubious) trap, as the d-rook got ready to pin your queen to the king, and you did fall for it (26... fxe4?) and exacerbated the material loss by not finding the queen-saving move 27... Re6.
When an opponent fails to resign a lost position, s/he gives you a valuable chance of honing your skill of winning won positions, which is, famously, the most difficult thing in chess (people become euphoric about winning too early and end up blowing the advantage).
Learn from your won games as hard as from the lost ones.

Here's an example of my game where I was atrociously (given that it was a corr game) imprecise at finishing the opponent off
I caruanised him in many senses: 1) played solidly in the opening (as solidly as it could be in Alekhine's), 2) exploited his mistake when he lost focus, 3) played too fast after the winning move 20 and failed to find a faster win (an uberstandard mate in 2 in a side line that wasn't played, then a semi-standard mate in 3, both with rook sacs, and then one more forced mate with a bishop sac).
I blitzed the 24... Rxf2+?? out because I wanted to finish the game fast and knew that the opponent would log in once a day and blitz out all his 50-smth pending moves, so the best chance to get in as many moves a day as possible was to play a reply very fast and set up a chain of conditional moves (which I did twice - chains of 5 cond moves 25-29 and 4 more at the very end did get played).
Only just after submitting the 24th move did I have an anxious feeling that I had missed Rg1+ (though not knowing that it was as good as mate in 3). I was considering 26... Qg5+ too, but again, failed to see the needed sac.

Still looking at 2nd post but re:
"Also worth noting is that 25. Rhe1!? was a little (dubious) trap, as the d-rook got ready to pin your queen to the king, and you did fall for it (26... fxe4?) and exacerbated the material loss by not finding the queen-saving move 27... Re6."
I knew that it was, I was willing to trade material. I hadn't totally abandoned the idea of simplification. There's a certain human element to the scenario, agreed.

As this consideration: in playing a string of games. I agree optimization is a virtue, but the effort sometimes takes over.

Still looking at 2nd post but re:
"Also worth noting is that 25. Rhe1!? was a little (dubious) trap, as the d-rook got ready to pin your queen to the king, and you did fall for it (26... fxe4?) and exacerbated the material loss by not finding the queen-saving move 27... Re6."
I knew that it was, I was willing to trade material. I hadn't totally abandoned the idea of simplification. There's a certain human element to the scenario, agreed.
The queen is a powerful mating piece that's helpful in time trouble, you shouldn't give it up that easily.
That's why he didn't resign - he was expecting you to 'trade material' like this... and of course to stalemate in the end as he said.
Surprisingly, people stalemate in totally won positions even in correspondence games. That's actually how I've got into round 2 of the 27th chess.com tourney - my main opponent in the group gave stalemate in a K+Q vs K ending when he had mate in two instead.
That was a relief after I failed to draw a K+2P vs K+Q ending in an earlier game in that group: one of my (White) pawns was on h7 and I dropped the king to h8 instead of a safe square and mated myself - I forgot that I had another pawn and that the standard stalemate resource wouldn't work (when one has a rook pawn on a pre-queening square, supported by the king, vs K+Q, they can easily draw by putting the king on the queening square whenever possible, unless the opponent's king is close enough to allow queening but mate on the next move).

There was no stalemate. I built the box and had it. Qc3 was the winning move, clearly. What of this??

My argument is this: of course (!!) there are tons of possible mates there, with the queen bungled in 15. What of it? It wasn't that great or interesting of a position to begin with. I can't hardly think of another feature that might mitigate what a stupid board it is. The stick in my craw about this fat chewing of it is I chose the path of least resistance with the proper calculation: promote the one pawn while the king is boxed in to two moves and won't be stalemated by promotion and had a free check which would be mate, period. Do you see another move than 50... Qc3# that would apply? Yeesh. A rant. My apologies.

My point is that if you fail to train the tactical vision to the point where you see faster mates in confidently won positions, you also won't see tactics that would save you in lost ones. When you think hard only in lost and equal positions or analyse only lost and drawn games, you learn slower than if you analysed all your positions properly.
Note that in a game theory perspective, which I am consciously thinking of here, I am thinking of how players quantify the cost of hubris in games. In other words, note that in spite of the fact that people benefit more from a proper resignation in terms of a long series of random internet match play, people still refuse to resign. In terms of, say, finding the set of ESS, how often does this occur? Is it really somehow idiosyncratic?