| |
BGonline.org Forums
Reducing the draw percentage (and equalizing for Black)
Posted By: Nack Ballard In Response To: Reducing the draw percentage (David Rockwell)
Date: Wednesday, 12 December 2018, at 9:57 p.m.
I see a game which is won based on a small advantage as a good thing.
Hear, hear, David. You might also like this article: Who needs stalemate -- let's abolish it!. IMO, Matt Bishop overstates his case but he makes some good (if not overlapping) points.
Stalemate=win would overhaul endgame theory, changing a significant number of small material advantages from draws to wins. (I'm guessing the 57% GM-v-GM draw rate would drop to around 40-45%.) I believe the average number of moves would go up in a game, as players test to see how to what extent they can press a small advantage for a win, and that would be tedious, but eventually this average-move-number would settle back down as known theory adjusts.
I like Casper's suggestion of an interim score for a stalemate. By "interim," I'm not sure whether he meant for as long as it takes for players to get used to it, or if he was reiterating his proposal for 0.6 to the stalemater and 0.4 to the stalematee. I think the concepts could be combined. For example, starting in 2021 the stalemater transitionally gets 0.6, then 0.7 the following year, and so on, so that from 2025 on the stalemater earns a full point.
Regarding Chess960/Fischerandom or Shuffle Chess: As it stands now, GMs seldom make mistakes in the first 20 or 30 moves, because they've memorized the established openings. The existence of 960 or more different starting positions essentially eliminate opening memorization. Even engine vs engine in Chess960 reduces draws from 29% to 21%. In human play, I would guess that the 57% (current rate for GM vs GM standard Chess) would reduce even more: perhaps to 30-35% at first, then gradually make it back to 40-45%.
If draw reduction is deemed to be a paramount issue, it seems sensible to attack the problem from both ends: Incorporate Chess960/Shuffle (affecting the opening) and give scoring advantage to the stalemater (affecting the endgame). I estimate that both changes would initially cut the current 57% draws by more than half, and it would settle in the low 30s.
The other issue that I introduced in this thread (in a longer post) was the concept of leveling the playing field for Black (currently winning only 44% to 45% of decisive games).
In many ways a "perfect" solution (and my current favorite after considerable reflection) is to give Black draw odds, while giving White appropriate positional compensation (e.g., White moves twice before Black moves once?). This would not only equalize the colors but also eliminate draws entirely, and without revamping endgame theory (though it still works to incorporate stalemate=win, and for that matter 960/shuffle, if a refreshment is desired or needed). This was essentially the remedy that was so successfully adopted in Go: White is given a komi of 6.5 points as compensation for moving second (the half point ensuring that a draw is impossible).
Another solution that works on both problems simultaneously (by incentive) is a modified Bilbao system of 3 points for a win, 0 for a loss, and if it's a draw then 1 point for White and (say) 1.5 for Black. It should be noted, though, that this motivates sharper play only in tournaments, not in matches.
Every round being decisive by mini-match-as-needed is an alternative that could be applied to matches. If the classic (slow) game is drawn, play rapid and if drawn again then Armageddon (sudden-death blitz). Same thing for round 2, etc. Classic Chess is still the main determinant, but a round (each of which consists of 1 to 3 games) is never drawn.
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.