3 ms·
> So it actually is a draw! (sort of) There's been a tournament of top engines playing out the position after the draw -- I think black (Carlsen) won 11 games
by cjbprime 8y ago
> So it actually is a draw! (sort of)
There's been a tournament of top engines playing out the position after the draw -- I think black (Carlsen) won 11 games and white won 2 games, with the rest draws. So this declaration of the lack of advantage just isn't true.
> just take things to the tiebreakers where Carlsen has a significant advantage.
But he had a significant advantage on the board, too. The question is the magnitude of the advantages. Caruana has previously been a world blitz champion too, so it seems unlikely that any advantage would be more than 60%-40% to Carlsen. Which seems smaller than the advantage we see from the computer tournament results.
- rtkwe 8y agoJust take a look at their relative rankings though. Caruana may have previously been blitz champion but his current blitz elo is ~170 points lower than Carlsen who iirc is the current blitz champion. I'd like to see that same process repeated for all the other draws from the points of highest advantage and right before the draw too if we're going to be using it to evaluate the decision to draw. Not sure how closely the computer tournament reflects the actual progression we've seen from Carlsen and Caruana because if we just look at the pawn equivalent advantage game 12 wasn't the best game that's been fought to a draw in this series. (I know that doesn't take into account positional advantage but we don't have an easy number to sum up that over all the games).
- rurban 8y agoHis current low blitz ELO ranking only says that he plays less blitz than Carlsen. Possibly to catch up to Carlsen in the normal rankings, with playing lots of normal tournaments. ELO is not measuring the potential, just the rankings of wins and losses. Caruana's potential is probably the same Carlsen. One should never underestimate him.