Freestyle Chess is Coming Home

[Ozymandias]

You're right on most of your comments, but the sad truth is, we Freestylers are such a small group that you have to negotiate with corr players and engine lovers. Even then, it's still unclear if enough participants will register, to justify a qualifier.

[Eduard]

So far, 18 participants have agreed to play under the existing concept. I am definitely not playing any more if I have to reserve up to 14 Sundays a year. I already had problems with 7 weekends, but 14 is a no go for me. In addition, 10 to 14 games are not enough for me. The concept presented at the beginning is feasible.

I have written following letter to Arno.
Translation (sorry for my bad english):

Dear Arno!
I plan not to participate in the Freestyle League games from 2022 under these new rules.

The already registered participants are suddenly confronted with completely new facts, I was not asked for my opinion. This is a pity. I was really looking forward to the tournament.

I can understand that some participants would like to play with pure engines. After all, the others in Centaur mode must first prove that they are better. The engine players should then play in engine mode throughout. I find it confusing that player X chooses engine mode against player Y, but would rather compete against player Z in Centaur mode. But well, if it has to be, then so be it.

More serious for me is my lack of understanding that some allegedly have fewer problems sacrificing up to 14 Sundays a year, when (for me) the entire Sunday is actually gone for other things because of 5 hours of play. I would be able to cope much better with 7 weekends a year, even if it means that the respective Saturday is wasted. Better for me than twice as many Sundays, when I'm already mentally at the Sunday game on Saturdays and would prepare myself mentally for it.

14 Sundays and then only half as many games as on the 7 planned weekends!

Not my thing. Ten to 14 games are not enough for me for such a year-round tournament. If you make even a small mistake (it can even be a mouse mistake), a whole year is messed up because you can't counter with so few games. If I don't have any practical chances after such a possible defeat, I don't even start playing. That's why I prefer 60m + 30s games. So you could play two games on Sundays, after which you get 20 to 28 games. If you can't do that, you'd better choose another discipline. The time pressure in the game is a failure factor, and whoever manages it better is a true freestyle champ. With 60m + 30s and with 60 moves, you actually don't have that bad 3 hours. Also, a fast player who has prepared well can pull faster and put pressure on the opponent. That's how tournament chess is! Some of the participants apparently don't like all of this. Some are looking for the perfect game. That's not freestyle, and that's how it should stay - my opinion. I think that given the huge amount of databases and fast engines, and super fast computers, it is more than enough to play 60m + 30s games. This level was even played 15 years ago, despite much worse engines or endgame databases. Stockfish is umpteen times better than Rybka back then, and you couldn't even dream of 7 Steiner Syzygy back then. I don't understand why you have to increase the level anyway. Still, I wouldn't mind if they could just play more games. As already mentioned, 10 to 14 games are just not enough for me, a mistake (can happen to anyone) and you are deprived of all chances for a complete year. There should be a chance that you can still get involved halfway through the route. Unfortunately, this is practically no longer the case.

Best regards,

Original letter:

Lieber Arno!
Ich plane, unter diesen neuen Regeln, an den Spielen der Freestyle Liga ab 2022, nicht teilzunehmen.

Die bereits angemeldeten Teilnehmer werden hier plötzlich vor völlig neue Tatsachen gestellt, gefragt wurde ich um meine Meinung nicht. Das ist schade. Ich habe mich auf das Turnier nämlich schon sehr gefreut.

Dass einige Teilnehmer gerne mit puren Engines spielen wollen, kann ich noch verstehen. Schliesslich müssen die anderen im Centaur-Modus, erst beweisen dass sie besser sind. Die Enginespieler sollten dann aber durchwegs im Engine Modus spielen. Ich finde es verwirrend dass Spieler X gegen Spieler Y den Engine-Modus wählt, gegen Spieler Z aber lieber im Centaur-Modus antreten möchte. Aber gut, wenn das sein muss, dann ist das so.

Schwerer wiegt für mich mein Unverständnis, dass manche angeblich weniger Probleme haben, bis zu 14 Sonntage im Jahr zu opfern, wo dann (für mich) der komplette Sonntag für andere Dinge eigentlich auch futsch ist, wegen 5 Stunden Spiel. Ich käme mit 7 Wochenenden im Jahr deutlich besser klar, auch wenn dadurch auch der jeweilige Samstag draufgeht. Für mich allemal besser als doppelt so viele Sonntage, wo ich doch auch an den Samstagen gedanklich schon beim Sonntagspiel bin und mich mental darauf vorbereiten würde.

14 Sonntage und dann trotzdem nur noch halb so viele Spiele wie an den 7 geplanten Wochenenden!

Nicht mein Ding. Zehn bis 14 Partien sind mir zu wenig für solch ein Ganzjahr-Turnier. Macht man auch nur einen kleinen Fehler (kann sogar ein Mausfehler sein) ist ein ganzes Jahr versaut, weil man nicht kontern kann bei so wenigen Spielen. Wenn ich nach solch einer möglichen Niederlage keine praktischen Chancen mehr habe, fange ich erst garnicht an zu spielen. Mir sind deshalb 60m+30s Spiele allemal lieber. So könnte man auch Sonntags zwei Spiele machen, wonach man auf 20 bis 28 Spiele kommt. Wer das nicht schafft, sollte besser eine andere Disziplin wählen. Der Zeitdruck im Spiel ist ein Fehlerfaktor, und wer das besser bewältigt ist ein wahrer Freestyle Champ. Bei 60m+30s und bei 60 Zügen, hat man effektiv auch nicht so schleche 3 Stunden Zeit. Auch kann ein schneller Spieler, der sich gut vorbereitet hat, schneller ziehen und Druck auf den Gegner ausüben. So ist Turnierschach! Einigen Teilnehmern gefällt das alles offenbar nicht. Manche suchen das perfekte Spiel. Das ist Freestyle nicht, und so soll es auch bleiben - meine Meinung. Ich denke dass angesichts der Unmengen von Datenbanken und schnellen Engines, und superschnellen Rechnern, es völlig ausreichend ist, 60m+30s Partien zu spielen. Sogar vor 15 Jahren spielte man diese Spielstufe, trotz viel schlechteren Engines oder Endspieldatenbanken. Stockfish ist zig mal besser als damals Rybka, und von 7 Steiner Syzygy konnte man damals nicht einmal träumen. Wieso man die Stufe nun trotzdem erhöhen muss, verstehe ich nicht. Ich hätte dennoch nichts dagegen, würde man nur mehr Spiele austragen können. Wie schon erwähnt, 10 bis 14 Partien sind mir einfach zu wenig, ein Fehler (kann jedem passieren) und man ist aller Chancen beraubt, für ein komplettes Jahr. Es sollte eine Chance gegeben sein dass man bei der Hälfte der Strecke, noch mitmischen kann. Das ist praktisch nun leider nicht mehr gegeben.

Viele Grüsse,

[Steamhammer]

Hello Eduard,
I understand your reservations, especially when it comes to the number of games. But as far as the reflection time is concerned, all experience has shown that centaurs need more analysis time if they want to deviate from the engines' main variations or just want to explore the chances. It doesn't have to mean that someone wants to play a "perfect game". Even the occasional checking of the engines' main variations (where it makes sense and in complex positions) often leads to latent time trouble against someone who strictly gos for the engine moves without wasting time, and the chances of success are quite a bit compared to today's top engines less than a few years ago.
As far as pure engine players are concerned, they are hardly attracted by tournaments in which the centaur mode is generally required and it is only permitted in exceptional cases to play in pure engine mode, which is more likely in the case of time problems, illness or other extraordinary circumstances to provide relief.
Let's see how the qualifying tournament, read the first test case, looks like. After that, there is still room for improvement that will appeal to the broadest possible group of the relatively small freestyle fan base.

[Eduard]

Dear Arno, of course, I'll be part of the qualification because I enjoy it. But for me it is almost certain that I won't play 14 Sundays in 2022. Why can't you play two games on a Sunday, for example? Then 8 Sundays are enough for 16 games. That would be a compromise for all those who want or can sacrifice fewer weekends.

But well, let's see what comes out of the qualification.

[Eduard]

Freestyle Chess League – Qualification Tour 2021 - Rules:

http://www.infinitychess.com/Page/Publi ... spx?id=347

[jefk]

personally i find it surprising that when
Arno Nickel has posted a suggestion to change/improve
the rules for correspondence chess (addressing the
'draw problem') he apparently continues with this
freestyle chess of stuff.

[Eduard]

Personally, I would be interested in the following point:

a) number of "stalemate wins" or "material wins" against a bare king

In the event of a tie, such "wins" count at the top. Does anyone have experience with this type of rating? It is not unimportant. Since most of the games will probably end in a draw, it will depend on the fine scoring. It will therefore be very important which openings you choose! In any case, it will be interesting, I am very excited.

[Steamhammer]

@jefk - What's so surprising? I am an active Freestyle Chess promoter since the beginning in 2005, and in 2015 we did already experiment with "stalemate wins" on InfinityChess in the so-called "Centaur Weekends Tournaments". My general approach to chess (means also for o-t-b) is based on suggestions by Lasker and Réti, who both recommended to change the scoring system. You might perhaps know that also a grandmaster and developer like Larry Kaufman agrees to such ideas.
@Eduard - Because of two reasons very few games end(ed) in stalemate wins: 1. you need a remarkable advantage in order to force a stalemate win (btw, stalemating the opponent, not yourself), as e.g. a simple ending R+P vs. R isn't enough, if you cannot exchange the rooks; 2. engine-only players were not able to force a stalemate win.
Apart from the scoring system, it's a fine nuance in chess.

[Eduard]

Dear Arno!

I consider this innovation at the tournament to be great! I have no experience with it myself. I've already thought about which openings I should play, but I didn't take this new aspect into account. Now I have to rethink everything.

What do you mean by "material wins"?

There is one more point that is not uncommon for auto-games with engines: I have seen many times that engines purposely give away material when they access endgame databases. Example: Rook ending KRPPP vs. KRPPP. Here it often happens that my engine swaps KR vs. KRPP in an endgame. The opponent has two more pawns but cannot win. At the end the rook ending is swapped to K vs. KP, and KP can set stalemate.

[Steamhammer]

"Material win" is my modern term for the old German term "Beraubungssieg" (robbery victory). Practically speaking it means, you have either K+N or K+B vs. K, but can't force stalemate. As you correctly say, K+P vs. K usually is sufficient in order to force stalemate. So why not reward K+B(or N) vs. K at the same level ...? Lasker and Réti took the old terms from the ancient chess before the big change arount 1500, and they argued, it had been an error to abolish it with the new rules. For a very long time nobody missed it because there was no problem with a high draw rate.
Interesting your observation of engine behaviour. We will not have many games centaur vs. engine-only, yet, it would be exciting to see if anyone succeeds in forcing such a type of "dtaw". It also means that centaurs have to be careful following engine moves that enter such a play ground.