Your network blocks the Lichess assets!

lichess.org
Donate

World Champion Tiebreaks: A Counter-Intuitive Proposal.

@PRIYAMVAD said ^

The odds that TWPL will be exactly equal is almost impossible.

But as @elkress said, as players will play more engine moves, it will become slowly but steadily more and more probable. At one point, like in 2-3 yrs, it will just become a sea of players having nearly the same to the same TWPL. And we don’t want that. TPLV, however, is going to be harder than anything to equal out in.

But players can't play like engines. Engines are far better than humans. Humans make mistakes.

@PRIYAMVAD said [^](/forum/redirect/post/8FiMJqCf) > > The odds that TWPL will be exactly equal is almost impossible. > > But as @elkress said, as players will play more engine moves, it will become slowly but steadily more and more probable. At one point, like in 2-3 yrs, it will just become a sea of players having nearly the same to the same TWPL. And we don’t want that. TPLV, however, is going to be harder than anything to equal out in. But players can't play like engines. Engines are far better than humans. Humans make mistakes.

But they can start memorising, can’t they? Then there will be a new wave of prodigies who will be even better than Gukesh or Sindarov or people like them. Maybe true that it’ll be harder than anything, but no matter what, they’ll at least be able to memorise the first 15 moves if they can memorise 40-move games.

But they can start memorising, can’t they? Then there will be a new wave of prodigies who will be even better than Gukesh or Sindarov or people like them. Maybe true that it’ll be harder than anything, but no matter what, they’ll at least be able to memorise the first 15 moves if they can memorise 40-move games.

@PRIYAMVAD said ^

But they can start memorising, can’t they? Then there will be a new wave of prodigies who will be even better than Gukesh or Sindarov or people like them. Maybe true that it’ll be harder than anything, but no matter what, they’ll at least be able to memorise the first 15 moves if they can memorise 40-move games.

They can memorize the first 15 moves. That's just the opening. People can't play like engines in the middlegame and endgame.

@PRIYAMVAD said [^](/forum/redirect/post/wStcxce6) > But they can start memorising, can’t they? Then there will be a new wave of prodigies who will be even better than Gukesh or Sindarov or people like them. Maybe true that it’ll be harder than anything, but no matter what, they’ll at least be able to memorise the first 15 moves if they can memorise 40-move games. They can memorize the first 15 moves. That's just the opening. People can't play like engines in the middlegame and endgame.

Yes, so that’s why I said that we should use the TWPL from the latter stage of the game. I never said to not use it.

Yes, so that’s why I said that we should use the TWPL from the latter stage of the game. I never said to not use it.

As being the world champion that difficult I don't think so really I should try it. Yeah I think I'm very champion I know I am a champion I know I am a World champion. Thank you bye-bye everyone

As being the world champion that difficult I don't think so really I should try it. Yeah I think I'm very champion I know I am a champion I know I am a World champion. Thank you bye-bye everyone

i think there will eventually be a point where, in order to play chess you will need a computer chip in ur head

i think there will eventually be a point where, in order to play chess you will need a computer chip in ur head

@PRIYAMVAD said ^

Yes, so that’s why I said that we should use the TWPL from the latter stage of the game. I never said to not use it.

So TWPL later in the game and TPLV earlier? But I don't see the need for that. Why not just TWPL all the way? What is the exact problem with that.

@PRIYAMVAD said [^](/forum/redirect/post/p82KUJnT) > Yes, so that’s why I said that we should use the TWPL from the latter stage of the game. I never said to not use it. So TWPL later in the game and TPLV earlier? But I don't see the need for that. Why not just TWPL all the way? What is the exact problem with that.

@RuyLopez1000 said ^

So, that makes a little sense in 1v1,

I don't see a problem with it.

but it makes no sense at all in a multi-player tournament.

I edited the blog as I agree that TWPL may not be suitable for tournaments as solid players would be favored over tactical/complex players.

Even so, I find this tie-breaking criterion unreliable, even in the 1v1 World Championship, since when one of the two is losing, it would force them to resign prematurely to avoid worsening their chances of defeat. I can imagine players resigning when the engine gives -1.8 without even trying.

No because the TWPL accounts for the fact that in extreme positions, the win percentage loss is lower as changes in evaluation won't effect the outcome as much (e.g. +10 to +8 is not much compared to +2 to 0). A player who is losing will not resign prematurely as they have the chance to improve their TWPL by improving their position as the winning opponent's TWPL would decrease at a faster rate due to how the win percentage works. (e.g. +10 to +8 is not much compared to +2 to 0).

Losing games is worse than having a lower TWPL score. The tiebreak doesn't change the fact that the players want to win games. Having a lead in the classical section is better than leading the TWPL. If the player has a practical chance of defending then they will play on.

I edited the blog to address this.

Furthermore, it would be relative to use a conventional computer versus a 120-core supercomputer to analyze the game, since even Stockfish makes mistakes.

The strongest and most consistent engine should be used in world class events (Stockfish), with optimal settings. The engine used, depth, search parameters and hardware should be disclosed publicly when measuring the TWPL and these settings should be kept consistent throughout the event.

I still think it's not a good tiebreaker. Computers can flag as a mistake a move that's perfectly valid for the human elite, even if it has a difficulty rating of 3100 ELO and the mistake is so complex that only Stockfish could recognize it. A computer's win rate has almost nothing to do with a real (human) win rate.

@RuyLopez1000 said [^](/forum/redirect/post/MA0F3tiW) > > > So, that makes a little sense in 1v1, > > I don't see a problem with it. > > >but it makes no sense at all in a multi-player tournament. > > I edited the blog as I agree that TWPL may not be suitable for tournaments as solid players would be favored over tactical/complex players. > > >Even so, I find this tie-breaking criterion unreliable, even in the 1v1 World Championship, since when one of the two is losing, it would force them to resign prematurely to avoid worsening their chances of defeat. I can imagine players resigning when the engine gives -1.8 without even trying. > > No because the TWPL accounts for the fact that in extreme positions, the win percentage loss is lower as changes in evaluation won't effect the outcome as much (e.g. +10 to +8 is not much compared to +2 to 0). A player who is losing will not resign prematurely as they have the chance to improve their TWPL by improving their position as the winning opponent's TWPL would decrease at a faster rate due to how the win percentage works. (e.g. +10 to +8 is not much compared to +2 to 0). > > Losing games is worse than having a lower TWPL score. The tiebreak doesn't change the fact that the players want to win games. Having a lead in the classical section is better than leading the TWPL. If the player has a practical chance of defending then they will play on. > > > I edited the blog to address this. > > > Furthermore, it would be relative to use a conventional computer versus a 120-core supercomputer to analyze the game, since even Stockfish makes mistakes. > > The strongest and most consistent engine should be used in world class events (Stockfish), with optimal settings. The engine used, depth, search parameters and hardware should be disclosed publicly when measuring the TWPL and these settings should be kept consistent throughout the event. I still think it's not a good tiebreaker. Computers can flag as a mistake a move that's perfectly valid for the human elite, even if it has a difficulty rating of 3100 ELO and the mistake is so complex that only Stockfish could recognize it. A computer's win rate has almost nothing to do with a real (human) win rate.

@Tactical-Attack said ^

I still think it's not a good tiebreaker. Computers can flag as a mistake a move that's perfectly valid for the human elite, even if it has a difficulty rating of 3100 ELO and the mistake is so complex that only Stockfish could recognize it. A computer's win rate has almost nothing to do with a real (human) win rate.

Someone brought up a similar point to you by asking if tactical players would be penalized.

It's a similar concept here. You ask if making a human move (which is not approved by Stockfish) is penalized.

The answer is no.

This is because TWPL is a relative measure.

We are only interested in whether one player is better than the other.

As an example, making a human move (which is not approved by Stockfish) doesn't hurt a player.

It only hurts them if their opponent manages to defend with computer defense. The opponent also has to cope with the pressure.

The opponent making errors will give them a worse TWPL than the other player.

Only way the opponent can have a better TWPL is if they defend against the move with an advantage.

In which case they deserve to have a better TWPL.

@Tactical-Attack said [^](/forum/redirect/post/xJnDJC23) > I still think it's not a good tiebreaker. Computers can flag as a mistake a move that's perfectly valid for the human elite, even if it has a difficulty rating of 3100 ELO and the mistake is so complex that only Stockfish could recognize it. A computer's win rate has almost nothing to do with a real (human) win rate. Someone brought up a similar point to you by asking if tactical players would be penalized. It's a similar concept here. You ask if making a human move (which is not approved by Stockfish) is penalized. The answer is no. This is because TWPL is a relative measure. We are only interested in whether one player is better than the other. As an example, making a human move (which is not approved by Stockfish) doesn't hurt a player. It only hurts them if their opponent manages to defend with computer defense. The opponent also has to cope with the pressure. The opponent making errors will give them a worse TWPL than the other player. Only way the opponent can have a better TWPL is if they defend against the move with an advantage. In which case they deserve to have a better TWPL.

@RuyLopez1000 said ^

Yes, so that’s why I said that we should use the TWPL from the latter stage of the game. I never said to not use it.

So TWPL later in the game and TPLV earlier? But I don't see the need for that. Why not just TWPL all the way? What is the exact problem with that.

Look, TPLV is crazy hard to equal because no matter what you do, there will be minute fluctuations. But in TWPL, people playing extremely accurately like Gukesh with his 14-move brilliancy will be closer to equal because as I said, for the games that end quickly, it would become cumbersome to calculate TWPL, but TPLV is a different story altogether. That’s why I am recommending to do TPLV for the beginning of the game. I know that it’s counterintuitive and a bit complicated, but this is just to reduce the possibility of getting equal centipawn loss.

I know that you will say it’s too complicated and confusing, but I’m just telling this in order to reduce the possibilities of equal centipawn loss.

(Yeah, sorry for all those long ‘c’ words :P XD)

@RuyLopez1000 said [^](/forum/redirect/post/1HfmrSOj) > > Yes, so that’s why I said that we should use the TWPL from the latter stage of the game. I never said to not use it. > > So TWPL later in the game and TPLV earlier? But I don't see the need for that. Why not just TWPL all the way? What is the exact problem with that. Look, TPLV is crazy hard to equal because no matter what you do, there will be minute fluctuations. But in TWPL, people playing extremely accurately like Gukesh with his 14-move brilliancy will be closer to equal because as I said, for the games that end quickly, it would become cumbersome to calculate TWPL, but TPLV is a different story altogether. That’s why I am recommending to do TPLV for the beginning of the game. I know that it’s counterintuitive and a bit complicated, but this is just to reduce the possibility of getting equal centipawn loss. I know that you will say it’s too complicated and confusing, but I’m just telling this in order to reduce the possibilities of equal centipawn loss. (Yeah, sorry for all those long ‘c’ words :P XD)