Freestyle Chess is Coming Home

[Eduard]

I had already thought about the material, but I wanted to be sure, that's why I asked.

These new rules are very exciting and make the game very attractive!

However, this also means that many playable and safe openings/variants with which one can easily make draw (e.g. with one pawn less) will fall out of practice. Who then wants to play a draw with one pawn less? Unfortunately, the opening abundance will suffer as a result.

Regarding the engine-games: This is where experienced players have an advantage when they compete against engines. From my own experience, I know that Engines would like to keep playing as long as possible. In an endgame, material is often offered to the opponent before the 50 move rule, if the new position is also a draw. A game like this can take many moves. In games like this, I will definitely continue to play to the last stone.

[jefk]

Eduard wrote

These new rules are very exciting and make the game very attractive!

earlier Steamhammer wrote

our top tie-break criterion in all tournaments will be the number of possible "stalemate wins" and "material wins" against the bare king, as we have already applied this rule in the previous "Centaur Weekend Tournaments"

ah ok, so apparently in line with some suggestions for correspondence chess,
some endgame rules now are changed for the Freestyle tour; but because i did not
participate in these previous Centaur Weekend tourns with those new rules
(although years ago i once did participate in 'Freestyle') i wasn't aware
of these new rules.
Could you please clarify such new endgame rules ?
and/or give a link where its specified precisely ?
Preferably with some examples ?
(a stalemate now is a loss ? and if you have a knight of bishop more
you get a full point counting as a win ?)
and how much will the draw rate now be affected you suspect?
thx
PS and then the obvious questions come:
1) are such new rules also useful (possibly as alternative for some years)
for the ICCF correspondence tournaments ?
2) is it feasible (easy programmable) for some engines to have an option
to incorporate such new rules (probably yes)
3) id. for egtb (endgame table bases), probably more
work, needing completely new tables again (no?)

[Steamhammer]

@ jefk

Could you please clarify such new endgame rules ?

At present, we are talking only about a tie-break rule, i.e. it does not affect the game scoring in the first instance. Stalemate remains a draw. But in case of a tie at the end of the tournament, stalemate wins and material wins are counted as bonuses. Conversely, someone who is stalemated receives a worse tie-break rating.
In the long run, however, it would be welcome if these tie-break results were also rated 3/4 point to 1/4 point in the first instance, rather than half a point.
I have written some articles about this, among others from 6 years on ChessBase (in connection with a survey among correspondence chess players):
6/2015
https://en.chessbase.com/post/correspon ... aw-problem
8/2015
https://en.chessbase.com/post/how-many- ... for-a-draw

Preferably with some examples ?

The first article above shows some examples, but I'm afraid the ChessBase archive does not present them as originally included. It's not necessarily about endgames, but that's certainly the main topic. K+P vs. K, if unwinnable, leads to a stalemate and is frustrating for the player with the pawn, who has been unable to convert an advantage, while the defender is happy not to lose. With "stalemate wins" and "losses" we recognize the achievements of both players in a somewhat justified way.

@Eduard

I don't think that a previously recognized pawn sacrifice in the opening will be unplayable under the new rules. Let's take one of the oldest gambits: the King's Gambit. I haven't tested it, but the compensation you get for the pawn in the form of initiative should generally be enough to prevent your opponent from forcing a K+P-vs-K endgame even with the best play. This should also apply to the Benkö Gambit and other opening variations. But maybe you have a particular variation in mind for which this doesn't apply?

[Eduard]

I let many variants play on the servers where in the end someone has one more pawn or material. As long as the endgame databases show a draw, all is well. In auto mode, draws like the PGN below are not uncommon! It doesn't matter if you have 2 cores or 64:

[Event "Rated game, 5 min"]
[Site "Engine Room"]
[Date "2021.11.09"]
[Round "?"]
[White "Gerasim1969, Stockfish 081121"]
[Black "Solista, Cfish 250621 64 AVX"]
[Result "1/2-1/2"]
[ECO "C89"]
[WhiteElo "2728"]
[BlackElo "2731"]
[Annotator "0.21;0.00"]
[PlyCount "92"]
[EventDate "2021.11.09"]
[SourceTitle "playchess.com"]
[TimeControl "300"]

{Stockfish 081121 (4 cores): 43.3 plies; 3.049kN/s Intel(R) Core(TM) i5-2400
CPU @ 3.10GHz 3093MHz, Slowmaster.ctg, 1024 MB} 1. e4 {B 0} e5 {B 0} 2. Nf3 {
B 0} Nc6 {B 0} 3. Bb5 {B 0} a6 {B 0} 4. Ba4 {B 0} Nf6 {B 0} 5. O-O {B 0} Be7 {
B 0} 6. Re1 {B 0} b5 {B 0} 7. Bb3 {B 0} O-O {B 0} 8. c3 {B 0} d5 {B 0} 9. exd5
{B 0} Nxd5 {B 0} 10. Nxe5 {B 0} Nxe5 {B 0} 11. Rxe5 {B 0} c6 {B 0} 12. Re1 {B 0
} Bd6 {B 0} 13. d3 {B 0} Bf5 {B 0} 14. Qf3 {B 0} Qh4 {B 0} 15. g3 {B 0} Qh3 {
B 0} 16. Be3 {B 0} Bxd3 {B 0} 17. Nd2 {B 0} Bf5 {B 0} 18. Bxd5 {B 0} cxd5 {B 0}
19. Qxd5 {B 0} Rfd8 {B 0} 20. Qg2 {B 0} b4 {B 0} 21. Bb6 {B 0} Qxg2+ {B 0} 22.
Kxg2 {B 0} Rd7 {B 0} 23. c4 {B 0} Bxg3 {B 0} 24. Kxg3 {B 0} Rxd2 {B 0} 25. b3 {
B 0} f6 {B 0} 26. Ba5 {B 0} Rb8 {B 0} 27. h4 {B 0} Kf7 {B 0} 28. Red1 {B 0} Rc2
{B 0} 29. f3 {0.21/35 17} Re8 {B 0 (Rg8)} 30. Rd5 {0.08/34 6} Be6 {B 0} 31. Rd6
{0.05/34 9} h5 {B 0} 32. Bxb4 {0.06/36 6} g5 {B 0} 33. hxg5 {0.11/41 6} fxg5 {
B 0} 34. Rd2 {0.06/42 8} Rxd2 {B 0 (h4+)} 35. Bxd2 {0.05/44 6} h4+ {B 0} 36.
Kf2 {0.05/45 15} g4 {B 0} 37. fxg4 {0.04/45 12} Bxg4 {B 0} 38. Re1 {0.04/47 7}
Rxe1 {B 0 (h3)} 39. Kxe1 {0.03/57 7} h3 {B 0 (Ke6)} 40. Bf4 {0.00/60 5} Ke7 {
B 0 (Ke6)} 41. Bh2 {0.00/63 6} Kd7 {0.00/68 14} 42. Kd2 {0.00/70 0} Kc6 {
0.00/67 7 (Bf5)} 43. Kc3 {0.00/70 5} Bf5 {0.00/68 4} 44. Kb4 {0.00/71 3 (Kb2)}
Bb1 {0.00/65 6 (Be4)} 45. Kc3 {0.00/63 4} Bg6 {0.00/63 4} 46. b4 {0.00/61 0
(Bf4) Solista,Cfish 250621 64 AVX2 offers a draw} Bf7 {0.00/65 6} 1/2-1/2

Diagram:
[fen]8/5b2/p1k5/8/1PP5/2K4p/P6B/8 w - - 0 47[/fen]


Now I have to completely rethink! Variants like this one can no longer be afforded in Freestyle Chess.

[Steamhammer]

What's the problem? Not the new rules, but the auto draw in engine matches.
The position will most likely be a draw also with the new rules, because it leads to a blocked position with either 50-moves-draw or 3-fold.
Another possibility could be a draw with K+B vs. K+B.
Yet, chessplayers have to think themselves how to evaluate endings.Not so bad...

[Eduard]

I do not quite understand. If I play in Centaur mode against an engine where the engine would be black here, am I not allowed to continue playing?

In the position I see the possibility that Black has to sacrifice his bishop and White ends up with KB vs K, i.e. with more material.

What am i seeing wrong? I've had this position on the board several times. If I don't offer a draw, then Black has always sacrificed or had to sacrifice the bishop.

[Steamhammer]

You are right, I misjudged the position (overlooked White's a-apwn), sorry.
With the new rules a centaur with White could force a material win. If playing against an engine "auto draw" is disabled.
An engine with White (against a centaur) might also achieve a material win, though curiously without knowing.

[Eduard]

My problem is that in my books (and not just me, but also many others) I have an innumerable number of similar variants in them. Especially for games with black. I play thousands of engine games a month on PlayChess. That shaped my thinking. So far I haven't had to worry about material or stalemate. A draw was always the same draw. Now I have to learn to rethink. I think that is only possible through experience. I think I'll be smarter in a year. I do not rule out that avoidable mistakes will happen to me in the upcoming freestyle tournament because I still lack the experience. Freestyle is hard.

[Steamhammer]

Same situation for all centaurs. I don't think, you have a disadvantage. Engine-only players might have. Good luck and have fun!

[Ozymandias]

It is not easy to organize such tournaments (with several weekends). It shouldn't fail because of the entry fee, there are also good prices (up to $ 3,000).

In the previous Engine Masters, 30 to 40 people have always played. But of course, in automatic tournaments your own chess performance is almost zero, maybe 10% to 20% book tuning. The engine does the rest. Whoever has the thickest PCs has the advantage. I've seen a lot of games in pure engine tournaments where they won because of gross book errors. The rest ended in a 98% draw!!

It's different with FreestyleChess. There will be no book errors, maybe a few during the qualification, and then no more later. So I would be surprised if less than 30 people register. It shouldn't be that pure automatic engine games triggers more enthusiasm than the freestyle discipline? In my opinion this is the wrong way to go, and then I will soon stop playing computer chess. We will see.

Try less than 20! With Alvin dropping at the last minute, we are down to 19 registered players (there's something even I wasn't expecting).