Tuesday, April 16, 2013

Non-Random Chess960 Trial Game 9: SP864

Here is the next in a series of trial openings we are playing through for very difficult SPs that black must face. The PGN database for these trials including comments is here.

Well for me personally, this SP number 864 was the toughest SP I have ever had to face playing the black side. I feel uncertain as to black's chances in this SP to draw and there would need to be more work done to convince me that black can find a draw (only my opinion please prove me wrong!)

The good news is that in practical play, black's opening moves are all principled and thus easy to commit to memory and white has to play very precisely otherwise all the initiative dissolves away. In any case even if black is in trouble in this SP, the game will play well beyond 25 moves. Therefore there is a lot of really interesting, creative and fun chess to be played in any case!

Not only is black's right to castle almost completely destroyed within only a few moves, but in this SP white's dark bishop sits in the corner readily deployed on the long diagonal firing at the kingside. Unlike our previous trial where the queen-side sequence was QBB, this time the queen sits on c1 in the classic "barbecue" BBQ arrangement (Weeks, 2012) and thus she has more freedom to move and exploit weaknesses from her c1 home.

Since we had already trialed the QBB version of this SP in trial 8, Mark found a very important improvement with 4.O-O. We both shared a kind of a laugh about our interpretations of this move! I thought of this early castling move as "super-solid" where I actually meant "aggressive". Mark thought I was implying that the early castling move was defensive. However please that it is not! After the castling move, white is free to gradually build up a very intimidating low risk attack on black.

The trial of this difficult SP concluded after 18 moves in some disagreement as to the likely outcome of the game. Mark chess expertise is significant. He gives black only a 10% chance of drawing after 18.d4. I think that this conclusion is harsh but it depends on a specific continuation starting with 18...Nd5 that produces a fascinating defense by black that we never got to play through unfortunately. So I present the hypothetical continuation after the 18th move just for fun.

SP864: How can black improve on this opening?

For me, I did not like this move I played:

SP864: Why 8...Nf4?!

Just from a question of logic, why threaten a pawn that requires white to respond in such a way that white's position is improved? Perhaps better was the immediate move 8...Qxc5. Black is actually trying to exchange off queens to take pressure of his king-side so that the king can lift to the seventh rank and finally connect the two rooks in a hybrid castling arrangement.

Since I feel uncertain about this opening by black 1.Ng3...Nf6, here is what I think might be a better opening:

SP864: An alternative opening idea

Black's idea is to give up another tempo to retain a king-side castling option and use the two military knights in the corner in a very challenging defensive setup that can protect the king. As we know, knights can be fantastic defenders if they are used well!


  1. SP864 BBQRKRNN, subscribed! I'll look at CCRL before playing a first move as White. - Mark

  2. The most popular CCRL move is...


    ...I see there are a few important differences between this SP and the previous SP (SP868). - Mark

  3. I wonder if we should just play out the standard line or should I try something out of the box like g3? I've gone for the standard line since g3 weakens f6 but I think it can work...

    The standout differences so far I am seeing is the knight fork on e2 against a castled king and queen totally alters the move tree. Another one is that the tactic Bb2 backed up by a queen on a1 is missing too.

    So having made my decision, check out the two standout games on CCRL for these first two knight moves. Critter 1.6a gets a draw against Houdini playing this system. Then we have Stockfish 2.3.1 castling queen-side as black with said same system!

    1.Ng3 Nf6

    ...Black is willing to give up castling after 2.Nf5 Rg8 because if white castles kingside black can use a rook on g8 as part of an attack on white's castled king. - Harry

  4. Does the Queen in the corner in the previous SP pack more punch than the Bishop in the corner here?...

    1.Ng3 Nf6 2.Nf5, if 2...Rg8 3.Nf3

    ...Following the CCRL most popular moves. I only give the 'if' sequence, because you indicated the intention to play it. - Mark

  5. Well it is starting to look to me like we have discovered a pattern in this class of SP thanks to your defensive ideas for black. The actual concept is pretty simple once you have experienced it and not hard at all to memorize or recognize for future generations of 960 players.

    I have a theory on queens in the corner. If she is backed up by a bishop on the long diagonal, we have a massive potential battery ram that can be used to smash the opponents king safety to create a weak square that becomes a forward outpost. However there is a perfect opportunity cost in that the queen is restricted while she is in the corner.

    If the queen is in the corner but no bishop on the long diagonal, she then creates mating threats in cooperation with knights that can move forward and find ways to penetrate on the opposite wing again with the same opportunity cost.

    What I found in our last game was that the Q+B battery ram was completely ineffective because without knights the weakness created by slamming through with the battery could not be exploited...

    In this SP, the Q+B battery ram takes too long to assemble. The essential balance of this and the last SP is that all of the diagonal cross the board attack on the king is mutual and black can exchange off B for N in order to render white's king attack ineffective because white cannot afford to push pawns on the kingside as an alternative to attacking with knights, since that would weaken king safety.

    At the same time as you have shown previously, any other weakness despite a king attack can be removed by black since the positioning of the pieces are such that a king attack is really the only possible threat.

    1.Ng3 Nf6 2.Nf5 Rg8 3.Nf3 Ng6

    ...Develops a piece. Black does not need to commit to any pawn moves yet. Black is simply in damage limitation mode - Harry

  6. Some commentators think that Queen in the corner means a slow game. I think it all depends on the initial position of the other pieces. It's also worth noting that all of the BBQ ('barbecue') positions have a corresponding QBB position...

    2.Nf5 Rg8 3.Nf3 Ng6 4.O-O

    ...The thematic move. If White doesn't play O-O immediately, 4...Nf4 will make it problematic. - Mark

  7. Yes completely dependent on initial position. However in order to reduce the complexity of predicting how soon the queen will contribute how about thinking something like how the minor pieces are distributed on the back rank relative to the queen.

    Back to the game, it is interesting that you are playing super solid with white as you did with black. Not extending yourself too quickly as I have done time and time again. This plan is really stealthy.

    3.Nf3 Ng6 4.O-O b6

    ...Preparing to attack the g2 square and hopefully open access to the h3 square as well. Black is reluctant to play anything other than rock solid moves that try not to induce a second weakness into black's army - Harry

  8. I didn't receive a message for the latest comment (it happens once in a while) and just saw your move. I'll need some time to think about it. - Mark

  9. You think I'm playing 'super solid'? That implies I'm playing defensively. In four moves I've managed to develop two pieces, to castle, and to stop Black from castling. That's not what I call defensive chess...

    4.O-O b6 5.c4

    ...White turns the attention to developing the Queenside and addressing the center. - Mark

  10. Good one yes relative to SP518 it does imply that.

    But we need new words to describe 960 openings! In this I meant "super solid" as actually more like the calm before the brooding storm where there is nothing much I can do but batten down. Your improvement with the early castling has actually increased your offensive power because almost every line is loosing for me. Is that not "super solid?"

    4.O-O b6 5.c4 d4

    ...Black must deal with the problem coming of d4/b4 which would damage the queenside beyond economical repair and even threaten the king - Harry

  11. I assume you mean 5...d5, rather than 5...d4, as in...

    4.O-O b6 5.c4 d5

    ...Re 'super solid', I try to make good moves following classical principles like developing the pieces, controling the center, and watching King safety. I've tried early b4/g4 ideas (...b5/...g5) in other games and they just don't work for me. Nakamura plays like that and does very well, so there is something to be learned there.

    The main concern for White now is to open the game without allowing mass exchanges of pieces. Every piece exchanged takes a little pressure off the Black King. - Mark

  12. Thanks for correcting Mark. Yes ...d5. I await your move then. There is some fantastic variations coming up including the possibility of an exchange sacrifice relatively soon.

  13. Greetings from a sunny Mayday in Brussels, when we do our first BBQ of the year...

    5.c4 d5 6.cxd5

    ...Avoids ...d4, exchanges a side Pawn for a center Pawn, and opens a couple of lines. How will Black recapture? - Mark

  14. May is a beautiful time of year here as well. The heat is finally starting to drop off and we can look forward to some cooler days and nights over winter (10-25 celsius). Enjoy the barbecues but unfortunately for me this particular barbecue BBQ on the board is looking like black's meat will be underdone because the flame has gone out.

    5.c4 d5 6.cxd5 Bxd5

    ...The other choices ..Nxd5 and ..Rxd5 although playable created their own set of problems and so ..Bxd5 at least retains black's harmony for now. - Harry

  15. The time stamps on the comments don't make sense. For my previous post, I'm seeing 'April 30, 2013 at 11:59 PM', although I know I posted it on 1 May and I believe you are eight hours ahead of us. The time stamp on the file I use to record the moves says 1 May, 9:01 AM, which would be 7:01 AM GMT...

    6.cxd5 Bxd5 7.b4

    ...Opens lines for the Ba1 & Qc1 and prepares to meet ...c5. Black doesn't look to be in any immediate trouble, but the long term pressure will be uncomfortable. - Mark

  16. I've posted this at about 5:40pm 8th of May, about an hour or two after your email arrived.

    6.cxd5 Bxd5 7.b4 c5

    ...Simply a case of pushing on with development. Black hopes that the queens might exchange off even if that means 1v2 isolated pawns. - Harry

  17. OK, I understand now. The time stamp in the comment is California time...

    7.b4 c5 8.bxc5

    ...Opens more lines. White seeks to trade the temporary poor position of the Black King for something more permanent. What will Black offer to take the pressure off the King and bring the Rg8 into the game? - Mark

  18. I'm struggling to see the light for black. This is really tough. It will only be a few more moves and I will be beyond my skill level to save the game.

    I think that this SP might be harder than the last because since the queen is not in the corner, she can move independently levering weaknesses in a variety of ways keeping black on the back foot. Since it is the bishop in the corner instead, it is already developed and deployed not inhibited by the movement restrictions in the corner at all.

    This is a clear case of your theory that black must try to make best use of being effectively a pawn down. White has so many good moves and is under very little pressure as well as being the equivalent of a pawn up.

    7.b4 c5 8.bxc5 Nf4

    ...A typical theme in this setup. Black uses an intermezzo to allow a bit more flexibility down the track - Harry

  19. Did you see the Carlsen interview with Charlie Rose? When asked why he was the best in the world, he mentioned confidence. He said other players also play extremely well, but don't have the same confidence in themselves. A corollary would be, 'Don't be afraid of things that aren't there'...

    8.bxc5 Nf4 9.Rfe1

    ...Looks forced to me. Other moves lack punch. - Mark

  20. The corollary is great. Firstly we have to know that what we think we see may be an illusion. If we don't know that, the corollary is ineffective. The other issue is that years of experience reveals that the thing isn't there. Until then, the ghost really does appear to be there.

    Confidence is contested as well. For example Magnus may have confidence in himself, but what if it turned out that his extra grace given faculties cause his win rate to go up and his confidence is actually just a product of the faculty increasing the win rate?

    He claims it is confidence, but I could argue that he has extra natural faculty and the rest is an ability to keep his mind from becoming unstable and unable to utilize that faculty. That is actually not confidence then that he is talking about but the ability to control his mind.

    I'm not arguing with you. I largely agree it only appears that I am disagreeing. I actually thank you for giving me the confidence to go on with this very dark situation where all the evaluations are permanently negative!

    8.bxc5 Nf4 9.Rfe1 Qxc5

    ...Will white exchange queens and risk taking some pressure off black's king? - Harry

  21. I extracted the relevant portions of the Carlsen interview to my main blog...

    Carlsen on Confidence

    ...Confidence isn't just a matter of facing the future. It also stems from the knowledge that the previous moves leading to the current position were well considered and sound. Chess, after all, is a game of logic. I start with a reasonable position; I play reasonable moves; my position remains reasonable; I'm confident that I will continue to find reasonable moves...

    9.Rfe1 Qxc5 10.Qb2

    ...White wants more from the Queen exchange than to isolate Black's c-Pawn. Now the Black King and Queen are both objects of attack. - Mark

  22. Thanks for the link and it's been good think about.

    I'm trying to lift my knights out of the way so that I can play f6/g6 and finally do something about that king.

    Your d-e pawns look ominous, they really do but I will play on somehow.

    9.Rfe1 Qxc5 10.Qb2 Be4

    ...The beginning of a plan to build a kingside fortress with f6/g6 - Harry

  23. This is the point where we normally suspend the game. For personal reasons I'll be taking the month of June off, so I suggest we continue playing this game until the end of May. Then, if you're still interested in continuing these experiments, we can start a new position in July. What do you think?...

    10.Qb2 Be4 11.Bxe4, if 11...Nxe4 12.Ne3

    ...Black has lots of options for the 12th move, which will determine the future course of the game. - Mark

  24. I've enjoyed this a lot and am happy to give it a break for as long as you feel is necessary for whatever reason. It takes two to tango!

    We make a good team because you are able to play through what I see as almost intolerable positions when I am actually the 100% Chess960 optimist!

    At a guess I think there are 40-50 difficult starts for black? We have already covered roughly 10? So we are quarter way there?

    I think you might have let me off the hook here what do you think? The move I was really worried about was 11.Ng3.

    The thing about 11.Ng3 is that:
    1) it keeps the e-d pawn files clear
    2) prevents me getting the free move ...Nxe4
    3) It delays even further the ability to connect rooks
    4) If I exchange bishops there is no relief in tension

    Could you take a look at that and give me your opinion before we continue? If you think that what you played is still best, then I continue with:

    11.Bxe4 Nxe4 12.Ne3 f6

    ...a really important move in the process of connecting the rooks - Harry

  25. I didn't look much at 11.Ng3, because it pulls the Knight away from the center and loses a couple of tempi. The retreat 11...Bc6 looks comfortable enough for Black...

    11.Bxe4 Nxe4 12.Ne3 f6 13.Qb1

    ...Skewers the Knight against the h-Pawn. Where is Black's counterplay? - Mark

  26. Sorry for the delay my brain turned to mush when I got a cold this week. Temperatures have dropped into the mid-teens and this is too much for an Aussie. However now I am drugged up to the eyeballs with anti-histamine and so am ready to fight.

    12.Ne3 f6 13.Qb1 Ng5

    ...a nifty little reply to the tactical skewer - Harry

  27. I don't see the 'nifty' bit, so you'll have to show me...

    13.Qb1 Ng5 14.Nxg5, if 14...Qxg5 15.Qxh7

    ...White seems to win a Pawn. - Mark

  28. It was a way to complete development efficiently and that it did. I'm trying trade down to a drawn endgame (unless you made a mistake like allowing me to mate you in one that is!)

    ...But you have that one little innocuous move that I always under estimate almost every 960 game I play...


    14.Nxg5, Qxg5 15.Qxh7 Kf7

    ...Black completes development with good activity all for the expense of just one edge pawn - Harry

  29. I don't know, Harry. A lowly edge pawn in the middlegame can become a mighty outside passed Pawn in the endgame...

    14.Nxg5 Qxg5 15.Qxh7 Kf7 16.h4

    ...Take a look at this:-

    Chess 960. Love it? Or not really?

    An Argentine player named morgondag mentions the Barbecue openings. (He also makes some good points about piece development.) - Mark

    P.S. After my next move I'll be taking the break we discussed earlier.

  30. Enjoy Mark see you whenever we meet again!

    If you exchange queens here, at least the position is pretty playable for black. Agree white is better but without queens a rook endgame a pawn down is not that serious. Yes your 5v3 pawn island is stronger than my 2v1 island but my king is already an active piece and yours is not.

    I feel like I can actually play this position with some confidence (assuming you exchange queens) which is better than some of the horrors I felt I was seeing a few moves ago.

    I think we have a disagreement about the alternative 11.Ng3 which is interesting.

    Also 1.Ng3...g3!? would be interesting to try! Gives up a tempo to stop Nf5 and requires ...f6/Nf7 to even have a chance at working. However introducing a structural weakness on f6 may not be a good idea.

    You definitely improved on white's play I think with 4.O-O!

    15.Qxh7 Kf7 16.h4 Qh6

    ...Encouraging the queens to come off, and if so black get's another life. - Harry

  31. So after ten months of steady play I find out that your Achilles heel is the endgame...

    16.h4 Qh6 17.Qxh6, if 17...gxh6 18.d4

    ...As they say, 'when ahead in material, exchange the pieces', to be followed by a regrouping for the endgame. I give Black no better than a 10% chance of achieving a draw. White will just keep offering piece exchanges, each refusal forcing a positional concession from Black.

    See you in a month or so! - Mark

  32. Greetings from southern France, where I'll be for the next few days. Have you seen this?...

    Chess 960. Love it? Or not really?

    ...'The blog Chess960 jungle has made some more serious attempts to clasify and analyze chess960 openings'. You're getting a name for yourself!

    As for more 'serious attempts', what do you suggest we try next? I have a few ideas jotted down from previous discussions. - Mark

  33. Hi Mark hope that your time in southern France involves good food and wine somewhere in your travels. You have inspired me to get off my ass and write up this very interesting game we played. Have a read when you have time.

    Will begin to think what we should trial next!

  34. I read your notes and stand by my assessment that White is winning where we left off. When one side wins a Pawn, as White has just done in this game, it is common that the other side gets a temporary initiative as you have demonstrated. White will now regroup his pieces, nullify the initiative, and then play to convert the extra Pawn. If White manages to exchange a pair of Rooks, as is highly likely, Black will simply be a Pawn down with little to show for it.

    As for improvements, your move 1...g6 is worth a try. I also think 4...e6, forcing the Knight back immediately, is better than the move you played. The last word has yet to be said for this SP. - Mark

  35. As far as I am concerned this SP we need to wave the yellow flag on it. I've taken a look at all the other SP's we have "waved the yellow flag at" but only SP376 shows a bit of a problem but it is a piece of cake compared to this SP.

    The problem with 4...e6 as already noted in our database of these trials and we actually experienced in our previous trial, is that the f6 square is weakened.

    4...e6 is playable but the teaming up of white's bishop and queen on the long diagonal against a weakened king-side is a challenge (but could be possible with a lot of analysis). I found an absolute amazing variation at the time after 4...e6 which I published in the database that leaves black with chances. It's a very "computerish" idea though.

    As far as I'm concerned we don't need to continue these trials until this SP has been sorted out by someone (or us). We might restart these trials if we come across an SP in our day to day life that we think is as bad or worse than this one. So far I haven't found any as tricky as this one.

    Another possibility is 1.Ng3...e6!?


  36. I think you're being overly pessimistic, in contrast to your usual enthusiasm. One game where White prevails isn't enough to condemn the entire SP.

    On 4...e6, you wrote, 'One problem with this is that it weakens f6 when black might need to play g6 at some point', followed by one 20-move continuation with a single branch. At the risk of seeming 'harsh' again, this isn't very convincing. There is an old saying in chess, 'Long analysis, wrong analysis', that probably applies here as well.

    I propose we switch sides. I'll play 4...e6, and you can use your attacking talents to show that I'm wrong. What do you say? - Mark

  37. Hi Mark. Done. Pessimism is good when it encourages further inquiry. Like wise, optimism is bad when it discourages further inquiry. I'll be playing pretty slowly! I have 10,000 words to write and four exams over the next couple of months so bear with me.

    I'll start up another trial page because this one is getting too long. Just realize that I actually want to disprove my pessimism and am glad that you will help do that.

    Just occurred to me that an alternative plan would be to post the position 4...e6 on Chess.com and see what responses we get? What do you think?

  38. I don't mind slowing the pace. You always play quickly, but there are often higher priorities.

    As for Chess.com, I'm not a big fan of groupthink on a chess position. Selecting a move isn't a democratic process. Having said that, we might learn something new about the SP.

    I play 4...e6, so the next move is yours. Give me a shout here if you start a new thread. - Mark

  39. Yes at this stage group think is probably not necessary. If we run out of ideas (which I doubt) we can ask the chess community.

    I've started a new post with a graphic of the initial position. Could you double check that this is the position you were thinking?

    If it is, let me know and I'll start thinking about my 5th move as white - Harry

  40. Got it, thanks. Following up your suggestion, I posted a challenge for SP864 & SP868 to Chess.com...

    Do you find chess960 challenging or fun?

    ...Let's see what happens. - Mark