armageddon in norway chess

[carldaman]

IMO, a scoring system that disincentivizes draws and rewards wins more, especially with Black, should be better than messing with the fabric of the game (by introducing weird variants).

Such an alternate scoring system was discussed here a while ago - HGM made an interesting suggestion, but I don't remember all the details.

Disincentiviting draws with White makes sense, not so much with Black. Logically Black should play for a draw, White should not. Simply awarding 0.6 to Black and 0.4 to White in case of a draw (promoted by Ed Epp) would be an improvement over current system, but perhaps not a big improvement.

I always thought top-level chess made some great strides forward in the 1990s', when the new young generation of players moved away from the old Karpovian 'win-with-White, draw-with-Black' paradigm.
There is still a need to go for a win with Black, for example in Swiss tournaments, in lopsided match-ups ratings-wise and whatever must-win situations may arise. Many tournaments use wins, as well as wins with Black, as tie-breakers, and rightly so. I never liked the Karpovian paradigm.

Interestingly, nowadays it is White who no longer necessarily seeks the largest advantage from the opening, hoping to outplay the opponent later in the middlegame - Carlsen's new paradigm.

The problem isn't incentive, it's that at least with five hour games the game is too drawish. I think that balloted openings makes the most sense, with chess960 a close second. Opening preparation is just too much of the game now.

I agree with this. Faster time controls visibly reduce draws. The trick is to preserve the quality of the games.
I doubt chess960 with its hideous castling rules will seriously catch on, but I do like the idea of experimenting with balloted openings, for some events at least. This may have a better chance of succeeding, but I'm surprised it hasn't been attempted yet in OTB games. Not that I know of, anyway.

[lkaufman]

IMO, a scoring system that disincentivizes draws and rewards wins more, especially with Black, should be better than messing with the fabric of the game (by introducing weird variants).

Such an alternate scoring system was discussed here a while ago - HGM made an interesting suggestion, but I don't remember all the details.

Disincentiviting draws with White makes sense, not so much with Black. Logically Black should play for a draw, White should not. Simply awarding 0.6 to Black and 0.4 to White in case of a draw (promoted by Ed Epp) would be an improvement over current system, but perhaps not a big improvement.

I always thought top-level chess made some great strides forward in the 1990s', when the new young generation of players moved away from the old Karpovian 'win-with-White, draw-with-Black' paradigm.
There is still a need to go for a win with Black, for example in Swiss tournaments, in lopsided match-ups ratings-wise and whatever must-win situations may arise. Many tournaments use wins, as well as wins with Black, as tie-breakers, and rightly so. I never liked the Karpovian paradigm.

Interestingly, nowadays it is White who no longer necessarily seeks the largest advantage from the opening, hoping to outplay the opponent later in the middlegame - Carlsen's new paradigm.

The problem isn't incentive, it's that at least with five hour games the game is too drawish. I think that balloted openings makes the most sense, with chess960 a close second. Opening preparation is just too much of the game now.

I agree with this. Faster time controls visibly reduce draws. The trick is to preserve the quality of the games.
I doubt chess960 with its hideous castling rules will seriously catch on, but I do like the idea of experimenting with balloted openings, for some events at least. This may have a better chance of succeeding, but I'm surprised it hasn't been attempted yet in OTB games. Not that I know of, anyway.

So perhaps you would like what I called "chess 18", which is just the 18 position subset of chess 960 with the kings and rooks on normal squares so normal castling rules apply. Anyway, the Norway tournament finished with the Armageddon score of 15 for White to 18 for Black, suggesting but not proving that the 10 to 7 time odds were almost but not quite enough for parity. Maybe balloted openings plus Armageddon might work well if we pick openings where White actually wins about half the games, or else ones where he wins maybe 40% with some time odds to bring it up to 50%.

[carldaman]

Yes, chess18 is a better option than chess960. I admit I'm also intrigued by Bronstein's so-called pre-chess, a form of shuffle chess where the players get to pick their starting setup by sequentially placing pieces on the back rank.

[Ovyron]

What is the effect of castling? How would a variant where black is allowed castling, but white isn't, do? Or in the handicap game, clocks are equal and there's no draw odds but black isn't allowed to castle?