Monday, August 29, 2022

Magnus opinion on symmetrical play in chess 960

Here is an interview with Magnus talking about chess 960 in the run up to the 2022 world championship:

He begins enthusiastically about 960 then becomes reserved towards the end. His initial thought is that playing symmetrically is the default way to play chess 960 in the absence of deep analysis.

From my experience the opposite is true. Playing symmetrically depends on the number of undefended pieces in the start position that will disturb piece coordination down the track if left unaddressed.

I think playing symmetrically is less beneficial in chess 960 then it is in classical chess because of the undefended pieces that occur more frequently in the start.

Yes, the rooks are undefended in the standard position, but they don't impact piece coordination in the opening phase. In chess 960, symmetrical players don't have the convenience of knowing that.

I wonder what you think?


  1. Nice find, Harry. Magnus makes the point that in SP518 it took a long time to go from Black responding symmetrically to Black trusting semi-open positions (against 1.e4) and Indian positions (against 1.d4). When I respond non-symmetrically on my first move as Black in C960, it's often because White chose a second-rate first move.

    In C960, we haven't been playing long enough to recognize which non-symmetrical responses can be trusted. Trusting it for one position doesn't guarantee that the idea works for other, similar positions.

    The general principle of taking a stake in the center often means that Black follows the same logic as White. That means a symmetrical response is the path of least resistance, especially given limited thinking time.

    Does that make sense? - Mark

  2. We have to ask two questions:
    Q1: How many moves of symmetry is aesthetically acceptable to the chess culture of the day? (even without knowing the theoretical win rate)

    Q2: Do 75% of the start positions necessitate breaking symmetry before answer A1?

    A2: if yes, c960 survives the culture.

    My guesses (shoot me!)
    A1: 10 moves
    A2: yes!

    1. re the asking 2 questions maybe you can't ask magnus but maybe you can ask wesley on lex fridman? I made a pitch:

      Wesley So on Lex Fridman podcast - here's my pitch. (There's also some news on Sergey Karjakin who said Wesley So is a hero, Hans Niemann might not have cheated and that Sergey likes 9LX.)

      Link to pitch comment:

  3. This comment has been removed by the author.

  4. Hey HarryO ! Glad to see you here after 3 years (not that there's been much 9LX outside St Louis anyway).

    1 - are you glider1001 from reddit ? I guess yes based on your chesscom profile

    Supposedly you said this:

    After playing Chess960, I realise how mentally unwell chess players become the more they play it. They know they will get the same ideal setup every game and they deny the fact that they will mess it up unless they memorise some other geek's ideas. In Chess960 there is nothing to deny. You know you are going to get a far from ideal start position and you know it is up to you to make the best of it and you know you will be playing moves nobody has tried before.

    Which is better for your mental health?

    2 - Do you disagree with Hikaru?


    'What is the greatest challenge in 9LX? Hikaru: Playing with white is surprisingly tricky.'


    2019 Champions Showdown | Chess 9LX: 960's Greatest Challenge?

    Hikaru is 1st in interview says --

    -- oh wait...

    i rewatched the interview now...ah hikaru doesn't say 'black can just play symmetrically sometimes.' hikaru says IT SEEMS like black can, so IT'S UP TO WHITE to prove it.

    Oh lol ok maybe false alarm.

    But anyway do you disagree?

  5. Hi Nic, yeah the thing about chess 960 I realise now, it is a slap in the face to "expert" knowledge itself which stifles everything it comes in contact with and often hides incompetence.

    Symmetry plays mokery to expertise. Black mimics white for free treating them as an expert until their "gut feel" tells them it is time to break out.

    Comically though, white is not trying to play expertly. They are trying to lure black into a copycat trap of thinking they know what they are doing! That can be tricky.

    At other times players simply want to go their own way and not play the bluff or focus on their own world of piece harmony (until they have to face reality)

    All of this happens in Fischer random without needing expertise even at the highest level. Yeah, 960 is good for mental health so long as you can handle not relying on expert knowledge.

    1. So yeah you're glider1001 from reddit and from chesscom right?

  6. Harry aka glider1001, do you agree with Ben Johnson from Perpetual Chess Podcast (Bennyficial1)?

    'tablebases from the endgame are perhaps a bigger threat to chess than opening theory'

    Also do you have twitter, email, discord or something? i'm and nicbentulan 4008 discord

  7. Hey Nic. Not enough info. Are they implying solvebility of chess as the problem? Are they implying memorisation of the endgame as the problem? If computers cause players to memorize the opening straight into a table base win, 960 will make that a lot harder. Chess960 wasn't invented for that anyway.

    Maybe players are worn out by the burden of a life of memorisation and repetition. They become conflicted. Can't let go of all that opening baggage so deflect the problem onto the memorisation of endgame theory.

    Bobby knew what he was doing. Chess should have just enough opening knowledge so that you can have some creative control at the start. Then into the sometimes nihilistically ugly wilderness you go, sometimes into the spiritually beauty. like it or not - never the same twice. If you survive, then find yourself in the raw truth that sometimes chess is a mathematically perfect game that has to be memorised to comprehend.