25 May 2012

Useful Blitz

I tried an approach to blitz this morning that differs from my frequent addictive obsession. Instead of playing game after game, I played one. After the game, I analyzed it without an engine. I did use the "novelty annotation" feature of ChessBase 11 for feedback on my opening choice. I played the novelty, which was neither worse nor better than prior games. However, several moves later, I blocked my own defense of a critical pawn, losing a pawn and rook for a knight.

Black to move
1k1r3r/ppq2ppp/2nbpnb1/3pN3/3P1P2/1NPBB3/PP2Q1PP/R4RK1 b - - 0 14

I played 14...Ne7?? 14...Ne5 would have been better.

My opponent picked up two more pawns, and then blundered in a manner that gave me a decisive attack.

White to move
1k1q3r/pp4p1/3b2n1/3p2P1/3Pn1Q1/1NP5/PP1B2PP/R4RK1 w - - 0 22

22.Be3?? gave the game to Black. I spent some time looking for improvements for White, and also examined my attack to see if I conducted it with maximum vigor. 22.h3 and 22.g3 both maintain a won game for White by neutralizing Black's threats. Later, I plan to verify my analysis with an engine.

After analyzing this game, I went back online and played two more. I gained an overwhelming positional advantage in the first, and then threw it away. In the second, I made an unsound piece sacrifice in the opening, managed to get pressure against the king. When my opponent failed to defend accurately, I gained back the material and then won on time after further errors on the part of my adversary.

White to move
8/2k2r2/p1pR4/1pP1P2p/1P6/P7/6PP/6K1 w - - 0 39

The simple 39.Rf6 maintains the two pawn advantage and offers a rook exchange that leads to an easily won pawn endgame. Instead, I played 39.g3?? Even then, there were possible improvements in my play that might have maintained at least a slight advantage. My opponent succeeded in swapping his a-pawn for my a- and b-pawns, and promoting his b-pawn.

Going back a few moves, I found a position that would be worth playing against the computer as a training exercise.

White to move
4k3/r4r2/p1pRp1p1/1pP1p2p/1P3P2/P7/6PP/3R2K1 w - - 0 29

I played 29.Rd8+, but 29.Rxe6+ or 29.fxe5 both deserve consideration. Training against the computer, all three moves will be played.

The third game emphasizes the point that online blitz games often create positions that are complex enough to deserve far deeper analysis that they get under time pressure.

White to move
k6r/pbp4p/3r1p1n/PBpP4/4P3/1R6/1PP3PP/2K4R w - - 0 25

After 25.a6!, my opponent was forced to return some material.

Rather than playing blitz to get my fix, playing and analyzing renders online blitz a useful training exercise. Of course, often online blitz is simply junk that is best quickly forgotten. But, even then, it might be worthwhile to quickly go over the game and identify what renders it junk.

6 comments:

  1. I have managed to trick myself to better focus when solving tactics problem to keep going until I have failed three problems. Three strikes and you are out! Maybe you can do something similar playing blitz?

    ReplyDelete
    Replies
    1. That's probably an excellent idea, Patrik. My problem, however, is that when I lose one, I feel a need to win five. Eventually the time for play is over, wholly wasted in idleness, and I must sleep or eat or go to work.

      Delete
  2. I use Fritz-Blundercheck after every ( blitz ) game. I look for "blunders" of 0.3+ centipawns. I see which move was better and i try to guess why its better.
    And i expand my personal opening"theory" after every game and during the time Fritz is analysing my blunders. This way i play a game in 10- min and analyse it 10+ min.

    ReplyDelete
    Replies
    1. I see the benefits, but lack the will. After I posted this entry on Friday morning, I ended up playing one forty minute session Friday evening on ICC. The morning started well with "Useful Blitz," but the day ended with rating-lowering garbage devoid of analysis.

      Delete
    2. you have these games still somewhere ? you can do that analysis know. Then you will have lost nothing. Put these games in an engine and work on your "faults". Its not to late ;)
      You may run the analysis while you are sleeping..

      Delete
    3. Thanks for the encouragement! A few years ago I analyzed every blitz game with Fritz. But, with more than 50,000 such games over the years, the analysis became tiresome. Going through the game immediately after playing, on the other hand, seems immensely useful.

      Delete