20 May 2012

Three Steps Backward, Two Forward

Training Log

Yesterday's Valley, Today's Climb
There's an old figure of speech concerning progress amid adversity: three steps forward, two steps backward. However, sometimes we fail to make progress. So it was with my marathon training session yesterday. I did not intend to spend more than four hours on Chess Tempo. My failure was grounded in initial success. My standard tactics rating had been hovering near +25 relative to the interim goal I was just under when I wrote "Getting Angry." My objective since crossing that mark was to achieve the next line at +50. I reached +45 after a relatively short session, then crashed. I ended the day at -46. By the end of the day I had correctly solved 125 problems in 273 attempts, an abysmal 46%. I must say that I was not enjoying the training. At least I had some solace in knowing that deliberate practice is not inherently pleasurable, but rather often painful. This morning, I refused to write my weekly training log until I had crossed back to the positive relative to mid-April's interim goal. I correctly solved 47 in 83 attempts for 57%. My rating is now +1. I have spent more than twenty-seven hours working Chess Tempo problems in the past six weeks.

For the week that ended yesterday, I solved 155/336 on Chess Tempo.

I continued working with Shredder's iPad app, but only ten problems.

1657 puzzles: 12987/16570 points 78%
last 10 puzzles: 97/100 97%

I also continued solving problems from the set of 300 that come with Chess-wise Pro, another iPad app. These problems are good quality positions, but only the first move is part of the puzzle. The first 110 offer a verbal clue upon request, but no other assistance. I have not been tracking these week-by-week, but have completed the first 134 since beginning in early March.

For the first time in several weeks, I spent more than one-half hour working problems from Lev Alburt, Chess Training Pocket Book II. Thus, two of my three resolutions were met this week.

Yesterday's slide began Thursday night. I played two of three rounds in an unrated thematic tournament at the Spokane Chess Club. Due to scheduling issues that others faced, I was asked to say a few words about the Dutch Defense before we began a tournament in which all games began after 1.d4 f5. It was enjoyable looking through some games to highlight for the club. Then, in the event, I played awful in the first game against Ron, dodging a bullet when he stepped into checkmate in one. Then, I made a simple opening blunder against Nikolay, giving him the initiative with Black. I defended well, and created some counterplay. At the end, he erred in giving me a choice of rooks to capture. The correct capture would have left me one pawn ahead, but I took the wrong one and had a minor piece for a queen. I left for home, while Nikolay went on to tie for second in the event with a draw in the final round. On Friday, I squandered some time playing online blitz in the quest for elusive redemption.


  1. The Standard Rating at CT does not use the Time you spend for the problem as factor for the rating. If you want to have a higher rating, thats easy: Spend ~twice as much time at every problem and your rating will go up ~ 100 - 200 Elopoints.

    1. Likewise, rushing things as I did yesterday, will just as assuredly make the rating go down.

    2. If you think half of the usual time your rating will go down ~100-200 Points. So the Standard ratingsystem is more "a watch" than a measurement of tactical power. The Blitz-Rating of CT compensate this thinking_longer - thinking_shorter effect astonishing well. You may read about these effects on Empirical Rabbits and my blog.

      Maybe it is useful to do CT-Standard for a training, to become better in tactics, but the standard rating is no indication for a progress ( or the opposit :). The Standard rating is in a higher degree showing how much (avarage) time you spend at your problems.
      If you want to judge your progress in tactics, i suggest to do some blitz at CT. You dont need to do these blitz puzzles "quick". This rating will show you better how your tacticalperformance/usedtime is.

  2. The trap with Chesstempo is that it's not offering textbook combos. Above a certain level, things become very complicated, especially in standard where you measure up with some guys who are ready to spend 15'+ / exercise making sure they get it right ! That's how you find some tricky puzzles rated under 1700 !

    So Axo's reco to try blitz makes sense. Just make sure not to rush too much !

  3. I know one guy rated OTB ~1700 who had a Standardrating of ~2400. He did spend more than 1 h !!! at each puzzle.
    Rating in Standard just tells you how much time you did spend.

    1. Understood. However, I am using the standard rating to measure progress against myself, not others. I also check the average solving time to make certain that I am consistently under. Rating itself is not the goal, but is a useful guidepost for self-improvement.

      I have started solving in blitz mode, too.

  4. if youraveragetime/averagetimeforproblem dont change, then! the Standardrating tells you something about your progress.
    If you always stay with your average thinkingtime for example at 0.9 times the average time for problems you solve... then its ok but your "improvement" ehh... will be difficult...