Tuesday, September 30, 2014

Openings That Don’t Work

     Over the years I’ve tried a few openings that no matter what never seemed to work as advertised. First it was the Stonewall Attack. I read a book by Horowitz and Reinfeld that showed how easy it was to play and, as demonstrated in the book, simply smash your opponent. It was appealing because the patterns were so simple. It never worked like they showed in the book and after losing enough games I finally learned my lesson: don’t play it.
     Another was the QGD Exchange Variation. Reshevsky played it frequently and he made winning look so easy. It was appealing because the patterns were so simple. It never worked like they showed in the book and after losing enough games, I finally learned my lesson: don’t play it. I just couldn't play like Reshevsky.
     Then there was the Budapest Gambit. It was what GM Robert Byrne called a twilight defense…not bad, but not good. I played it a lot but ended up putting a lot of effort into getting P back and not having any real attack to show for it. Things just didn’t work out for me like they did when Bisguier played it. I finally learned my lesson: don’t play it.
     Finally there was the Samisch Attack against the King’s Indian. It looked like the perfect way to meet the King’s Indian. I saw guys like Botvinnik, Tahl, Spassky and other famous GM’s play it and Fischer lost five games against it and so avoided it whenever he could. All white has to do is castle queenside and attack on the K-side by pushing g- and h-pawns. It never worked out as planned. In fact, in a recent game on LSS using engines I managed to lose a game. Apparently I still haven’t learned my lesson on this opening.
     I could name some other stuff that never worked out like they were supposed to…the Caro-Kann and Rubinstein French, for example. They were supposed to be simple and easy to play…they weren’t.
     On the bright side, one opening that did work quite well was the Torre Attack. I learned the patterns against all the different possible black setups (Sicilian-type, French-type, etc) and was happy with my results. The problem is that playing the same thing all the time got pretty boring and since I’m not playing serious OTB and when it comes to correspondence chess, nobody cares what your rating is and even playing for the CCLA, ICCF or LSS where you use your real name, you’re still, for all practical purposes, anonymous anyway and ratings don't matter.  Can anybody name the top 5 correspondence players on any of these sites, or for that matter, who is the world correspondence champion?  You play correspondence chess strictly for your own amusement.
     My solution has been to experiment with backwater openings and defenses on LSS lately. I’ve tried the Alekhine Defense (4 draws), QGD Chigorin Defense (1.d4 d5 2.c4 Nc6) which has yielded two draws, 1.a4 which resulted in +1 –0 =2. I’ve also tried 1…a5 against 1.e4 and lost but drew against 1.d4. The rather odd approach against the Slav Defense worked out well and resulted in a win: 1.d4 Nf6 2.c4 c6 3.Nc3 d5 4.e3 e6 5.Nf3 Nbd7 6.Qc2 Bd6 7.g4


     And, as mentioned in previous posts, the Urusov Gambit (1.e4 e5 2.Bc4 Nf6 3.d4) has been surprisingly successful even against engines! +1 -0 =1 and in two current games I stand at least equal.
     Here's a nifty little Tartakower game using the Urusov.  You have to kind of feel sorry for Steiner because he lost on a tactical blunder from a better position. There were a bunch of Steiners:   Herman, Lajos, Endre. There were also a Dr. Conrad Steiner and a Dr. Konrad Styner, but for our purposes they don't count.

2 comments:

  1. I have always had great success against opponents under 2200 with the Urusov. Almost no one plays the best lines or seems to know the Panov line -- though I think White actually does ok there with 4...d5 5.exd5 Bb4+ 6.Bd2 = or 6.c3 Qe7+ 7.Be2 dxc3 8.bxc3 Bc5 9.O-O O-O 10.c4 Re8 11.Nc3! and if 11...Bb4?! 12.Qb3! and Black cannot win the Bishop without getting his Queen trapped after 12...Bxc3 13.Qxc3 Qxe2 14.Re1 etc. White can also look at 6.Kf1! which computers tend to like a lot. You can see my analysis here:
    http://www.rci.rutgers.edu/~goeller/urusov/gambit/H.html#h4

    Not sure you have seen my full Urusov bibliography:
    http://kenilworthian.blogspot.com/2012/11/urusov-gambit-bibliography_26.html

    There I recommend the Modern Attack vs the Two Knights / Scotch Gambit line that follows 4...Nc6, which is definitely about equal. I think Black might do best with the 5.e5 Ng4!? line -- see here:
    http://www.kenilworthchessclub.org/games/java/2013/C55-Ng4.htm

    Since doing that analysis, I tend to think that White's best option might be the Max Lange as you play it, especially since the regular Max Lange Attack is pretty promising for White:
    http://www.kenilworthchessclub.org/games/java/summer05/horowitz-lange.htm

    Meanwhile, the Anti-Lange lines are not bad and Ian Simpson treats them fairly well:
    http://tws27.50webs.com/chess/scotch5.htm

    In the end, it is all about equal against well prepared opponents. But it is still lots of fun about 90% of the time. Glad you are enjoying it.

    ReplyDelete
  2. Thanks for your additional information! I have a new game in progress and again my opponent did not play the Panov idea. As you point out this probably because the line is generally unknown. It is all but ignored in the engine opening books I have!

    ReplyDelete