"I do everything I can to make my opponent's life miserable"
shockingly, that's not the right thing to do. You're taking pleasure in other people's pain and trying to seek validation for it? Just be the bigger person and end the game so everyone can go on with their lives, really strange that you even ask the question in the first place.
This all makes me wonder, how bad does the situation have to be to make resigning strategically rational? How unlikely does a draw or win have to be in order for resigning and playing more games against slightly worse opponents to be a better way to improve your ELO than playing every game to your king's last breath?
Moving your king for 49 consecutive turns just to stall as theirs is in prison is always fun. Their fault for not resigning
Apparently, you have drawn 13 games due to this! No wonder you like the strat so much...
I was wondering how many times I drew due to that haha cheers! And yeah, it's also a gamble of whether I can overcome my progressive myopia and avoid the stalemate
Don't mind the 1/1 though, stalling the game is the reward in of itself.
So both players are rewarded, one by getting a draw in a lost position, and the other, um... somehow.
1. Resigning an equal game has a cost of 8 ELO. We will assume that your previous rating was correct for the sake of argument.
2. Running out the clock looking for a draw or win has on average around half the clock for each side = one full clock.
3. If you do resign, the expectation is that 8 points of ELO difference means your ELO will climb at an initial rate of around 0.2 points per game. Meaning it will slowly approach your previous rating assymptotically. If it weren't assymptotic it would take around 40 games for your rating to stabilize.
So does this mean that you should be willing to fight for a draw if there's at least a ~1.5% chance you'll get one? I'm not sure what it means. It is worth mentioning that ELO actually bounces all over the place, and that in reality most players are slowly improving. Possibly faster than the 0.2 ELO per game assumed.
By the words of GothemChess, "never resign"
By the words of GothemChess, "never resign"
That advice is meant for children to teach them not to give up after a mistake and to give them experience playing difficult positions. Once past the beginner stage, you should respect your opponent by resigning when it's clear that you're going to lose. If you think you have a trick up your sleeve to get a stalemate or draw, or if your opponent is very low on time, keep playing; otherwise, cash in your chips.
I'm in the process of sueing an insurance company. They are going to lose but they are fighting a scorched earth fight to make me give up. In chess you should be able to finish off your opponent by knowing the endgame. If you don't know how to do that, you don't know how to play. Not resigning is just a psychological strategy to demoralize your opponent. I myself would resign out of respect for my opponent. But all's fair in love and chess.
I'm just here to read comments
Stop doing this or else you will get banned by violating the sportsmanship policy. Just saying)
I'm in the process of sueing an insurance company. They are going to lose but they are fighting a scorched earth fight to make me give up. In chess you should be able to finish off your oponent by knowing the endgame. If you don't know how to do that, you don't know how to play. Not resigning is just a psychological strategy to demoralize your oponent. I myself would resign out of respect for my oponent. But all's fair in love and chess.
Or it is because sometimes people blunder a draw, and they don't want to just resign when there's a chance they'll get lucky. It doesn't have to be psychological warfare or an attempt to insult anyone. Some have also been taught to never resign, so they never resign. I so commonly see people taking it personally, an insult on their intelligence and the likes. It doesn't have to be that deep. Not resigning isn't about disrespect, but dragging the game out to waste your opponents time is only about disrespect.
I concede your point. Not resigning just being psychological warfare was a hyperbolic thing for me to post.
By the words of GothemChess, "never resign"
That advice is meant for children to teach them not to give up after a mistake and to give them experience playing difficult positions. Once past the beginner stage, you should respect your opponent by resigning when it's clear that you're going to lose. If you think you have a trick up your sleeve to get a stalemate or draw, or if your opponent is very low on time, keep playing; otherwise, cash in your chips.
It doesn't even matter if you are a beginner or not(maybe GMs), you could be 1400 and down a piece by move 10 that doesn't mean you HAVE to resign. Even if you are in a worse position don't resign just because of "respect for your opponent" everybody makes mistakes. Don't just think "oh I'm down 2 pawns and my opponent can probably win 30 moves." Are they stockfish? No.
By the words of GothemChess, "never resign"
That advice is meant for children to teach them not to give up after a mistake and to give them experience playing difficult positions. Once past the beginner stage, you should respect your opponent by resigning when it's clear that you're going to lose. If you think you have a trick up your sleeve to get a stalemate or draw, or if your opponent is very low on time, keep playing; otherwise, cash in your chips.
It doesn't even matter if you are a beginner or not(maybe GMs), you could be 1400 and down a piece by move 10 that doesn't mean you HAVE to resign. Even if you are in a worse position don't resign just because of "respect for your opponent" everybody makes mistakes. Don't just think "oh I'm down 2 pawns and my opponent can probably win 30 moves." Are they stockfish? No.
2 pawns isn't completely lost at 1400. It's completely lost at like 2800. I'm not even sure a bishop and pawn is completely lost at 1400, and at 700 people come back from blundering queens on occasion.
I would say if Stockfish is down two pawns for nothing, or a super GM is down a bishop for nothing, or if a 1400 is down a queen, then we should be asking why they think they would be able to force a draw against their clearly competent opponent.
As for the 700, I am a 700 and I will have to agree with Gotham here. Never Resign. Your opponent is just as incompetent as you are. If you are bad enough to blunder a queen for nothing, so are they. No reason to respect your opponent's skill if you don't respect your own skill. I have won games down 20 points of material. And probably lost them with as much advantage.
Note that time control should have a similar effect to ELO too. Down 2 pawns in bullet is not completely lost unless you're Stockfish. Down 2 pawns in correspondence might be completely lost even well below the level where anyone is competent since nobody makes bullet blunders in correspondence and everyone is thinking several moves deep.
Larry Kaufman has apparently said:
"[T]he Elo equivalent of a given handicap degrades as you go down the scale. A knight seems to be worth around a thousand points when the "weak" player is around IM level, but it drops as you go down. For example, I'm about 2400 and I've played tons of knight odds games with students, and I would put the break-even point (for untimed but reasonably quick games) with me at around 1800, so maybe a 600 value at this level. An 1800 can probably give knight odds to a 1400, a 1400 to an 1100, an 1100 to a 900, etc. This is pretty obviously the way it must work, because the weaker the players are, the more likely the weaker one is to blunder a piece or more. When you get down to the level of the average 8 year old player, knight odds is just a slight edge, maybe 50 points or so."
Right, so if we just interpolate and extrapolate this a bit, we could say each point you're down changes your effective ELO to the next number: 5300, 4500, 3900, 3400, 3000, 2675, 2400, 2175, 1975, 1800, 1650, 1525, 1400, 1275, 1175, 1100, 1025, 960, 900, 845, 795, 750, 710, 670, 625, 590, 555, 525, 495, 465, 440, 415, 390, 365, 340, 320, 300, 280, 260, 240, 225, 210, 195, 180, 165, 150, 135, 125, 115, 105... To be fair I'm kinda implicitly assuming 8 year olds are around a rating of 260 here, which I'm not sure is accurate. And it doesn't even begin to account for sub-human performance from Martin, random move, or deliberately bad play.
Given that ELO is based on win percentage, a 1400 missing a knight is basically an 1100. Would an 1100 resign in the starting position against a 1400 because of sheer hopelessness?
Probably not. Given that the prediction is still 15% in their favor.
If we imagine 3 players with different resignation thresholds, for when we might resign, would normally resign, and when it would be dishonorable or disrespectful not to resign, 1 in 10, 1 in 50, and 1 in 250, corresponding to FIDE (Gaussian) ELO difference of 367, 587, and 758, then we can see what sort of material disadvantage would be early to resign, what sort of disadvantage is reasonable to resign, and what sort of disadvantage we should start to tell them they should resign.
At hypothetical super engine level, one pawn should be all it takes to demand the other engine resign.
At stockfish level, 1 pawn and we might resign, Not resigning after losing 2 pawns would be dishonorable.
At super GM level, losing 2 pawns is enough we should normally resign. Not resigning after losing a knight would be dishonorable.
At 2000 level, we might resign losing a knight, should normally resign losing a knight and pawn, it would be dishonorable not to resign down 2 knights.
At 1400 level, we might resign down a knight and pawn. We should normally resign down both bishops and a pawn, and it would be dishonorable not to resign down a knight and queen.
At 1000 level, we might resign down both bishops and a pawn. We should normally resign down a rook and queen, and it would be dishonorable not to resign down both rooks, a queen, and a knight.
At 700 level, we might resign down a knight and queen. We should normally resign down both rooks, both knights, and a queen. But there's not really a point where it's dishonorable not to resign. Or at least, that's what the formula says. I think at this point though we would need to take into account the compounding effects of being down 2/3rds of the board in material.
Incidentally this nonlinear equivalence between ELO and material advantage also probably means we can convert centipawns loss per move into ELO. Right, if a 3400-rated engine can manage to come up with 300 cp over a, say, 60 move game, then we might assume that 2400 is about 5 cp lost per move with rapid time control. 1800 is 10. 1100 is 20. 300 is 50. And perhaps around -500 every move blunders an entire pawn between equal players. Does this theory hold up in reality? Can we really estimate ELO from centipawn loss? I'm not sure. One thing is that really good players play more moves than normal because they play complex positions, and really bad players also play more moves than normal because they miss opportunities. Reasonably short sharp games should mostly be a middle level thing. So the cp of loss might be overestimated for really bad and really good players. Or there could be other factors at play.
Unfortunately, Stockfish seems to disagree with our centipawn loss estimate severely. Lichess stockfish tells me a 54-move 663 vs 660 rated, rather chaotic game I won as black, which we might assume has a 28 cp loss or so, actually has 124 for my opponent and 112 for me. That seems to destroy our estimate except that Stockfish doesn't think the same as we do and will happily tell us how a Knight odds starting position is like +9 or something and not +3. So this is probably the one case where we can safely say we know better than Stockfish and say that a ~depth 25 stockfish centipawn is a lot less than a real centipawn. Maybe 3 millipawns or something. Or we could do the opposite and just multiply our numbers by 3 or so.
Now I wonder how many centipawns Martin is losing every move against itself?
@psd314 I agree with most of what you said in the top section, but I haven't slept yet and think elo at lichess is a bit.. inflated.
@pds314 that's interesting and seems to explain it very well. Surely there's no-one here so proud of his ability at about 1100 level that he gets all upset if an opponent, a knight down, doesn't resign? If so, that's a sheer lack of logical ability, which explains, more than likely, why he's 1100.
Also I'm getting some passive-aggressive feeling from this, resign whenever you want but not for something dumb.
Has it occured to you that the person who refuses to resign might be the one trolling you, by wasting your time?
Having fun is never wasted time.