Saturday, July 01, 2017

Baseline

As you might remember, I'm on a mission at  Chess Tempo Blitz to get from 1700 to 2000 average. The past days I have set my current baseline at Chess Tempo: 1750. That's about the same average as a year ago, when I got an All Time High of 1806. Since Chess Tempo has no way to measure the average rating automatic, it is easier to take the All Time High, which differs about 50 rating points from the average. The next landmark then will be when the All Time High surpasses 1850.

I have assimilated a lot of new knowledge the past year, and since I haven't made any progress ratingwise, it is proven that you don't become better at tactics by gaining knowledge, once you have reached a certain plateau. Which is what we expected.

The coming time, I will work along two tracks: gaining new knowledge and put the acquired knowledge into practice. The latter I haven't tried the past year, for clarity.


7 comments:

  1. I just bought CT-Art 6.0 and I am loving it. It's the opposite of Chesstempo and my Combination Challenge! book. It tests you on how quickly you recognize tactics patterns, and drills them into you at the same time. With CT or my tactics book (although I have many) you have to forget everything, every pattern you ever learned, I feel, just to solve the problem.

    OTB, a lot of the little blunders are speed of board and tactical pattern-recognition. With simple problems set up like this by theme, you can drill for this; with ChessTempo, the problems are usually so complex that, if there is a pattern, it takes so long to find the pattern that it has already bypassed your quick pattern-recognition.

    The reason I got this CT-Art is because I have to move slow OTB, and then I will see deeply and avoid traps, but if I move quickly then I step into too many tactics that much lower-rated player may, consistently I might add, see faster than me. I understand that my strength/difference may come from my ability to out-analyze lower-rated players, but I can't make those "quick mistakes", which are the kinds of tactics that lower-rated players use to beat higher-rated players.

    I've never tried blitz on ChessTempo, but to do that for a rating sounds to me like banging one's head against a wall, because CT tactics are not pattern-recognition tactics, they are think it out and solve it type tactics. If I tried to blitz out CT tactics I would become demoralized quickly and all my move habits discipline would be quickly discarded. Not saying you go for speed and not solving.

    With CT, I have to do a material count, ask what last move and threats are, all of these systematic things that bypass instant pattern-recognition. I am going through all of the problems on CT-Art, including the easiest ones, don't care if it's thousands of problems. I want to make sure I recognize a mate in one, for instance, quickly, in 6 seconds or less instead of 9 seconds or more, etc. I've seen an Expert do this before, it's not as if everyone loves to solve nothing but the hard, slow problems.

    Another problem with hard problems is that you may end up only solving a few, versus a lot of easy tactics patterns in the same amount of time. So, in the end, you are improving your _calculation skills_ on CT, and may be leaving tactics _pattern recognition_ skills rustier than they should be.

    Ultimately, solving complex problems, and tactical/board-vision skills are both important, and both need to be addressed. If someone only solves hard problems on CT for instance, I think their ability to play blitz at a respectable level could actually suffer from that; e.g, looking for hard problems, but missing the easy stuff. Chess is like that, you can't train in one direction only, have to train in parallel directions. Endgames are an example of another direction, some are high-rated at blitz, for example, due to their endgame skills and training. Going over games solitaire-style (I'm going over 107 Great Chess Battles by Alekhine right now) is another type of training.

    ReplyDelete
    Replies
    1. http://takchesschess.blogspot.ie/2017/05/an-updated-list-of-ct-art-4-for-android.html?m=1. i think ctArt 6 contains a number of the programs including a very basic chess tactic program . the tradition 1200 ctart set which was ctart 2 and an additional 1000 mainly composed problem set that was added to later programs which became ctart4. Some of the higher level ctart problems I think you will find difficult , i look forward to hearing more as you work through them.

      Delete
    2. Good, point! I am sure they will get harder and am looking forward to it. :-) For now, I'll try to do the easy ones at night, when I am tired and before I go to sleep.

      It's funny that Shirov says he's a good calculator and thus loves endgames because I feel that is what all of those 2700 players have to be good at is calculating. All these tactics are window-dressing, eye-candy to show off their calculation skills. hehe.

      Delete
  2. If you haven't looked at the Help file for the 3 modes of tactics training on Chess Tempo, I suggest you do so. I didn't, and I had the wrong idea about which mode(s) to use for the training I wanted to do, until corrected by Aox. Here's the link:

    Blitz/Standard/Mixed

    If you want no timing, use Standard mode. If you want some timing, use Mixed mode. If you want significant timing, use Blitz mode.

    I suggest randomized selection of problems from at least two or more different themes/devices, rather than massed practice that is focused on one particular theme/device. It will not "feel" as if you are making as much progress, but cognitive research says that you will better retain what you have learned if you do NOT use massed practice on one particular theme/device at a time.

    ReplyDelete
  3. http://tartajubow.blogspot.ie/2017/06/shirovs-birthday-gift-to-tahl.html?m=1 for my fellow tactic loving friends.

    ReplyDelete
  4. Tactics is for sure important. One mistake - and all good planing, all good strategy was in vain.
    Having said that - very difficult tactics are rare (see chesstempo puzzle rating distribution).

    Assuming you have no so difficult tactics in your game - then your gained knowledge should result in a rating gain.
    Even though I trained tactics a lot at chess tempo, and for sure got a little bit better.
    (I mean "Better" by Blitz rating measurement - those who dont know what Blitz mode means shall just shut up, please. The word "Blitz" is misleading anyway.)

    And still, I went the way that I AVOID tactical positions. All I want is, that all my strategical efforts are worth something and not lost due to an oversight of a medium-difficult tactic.
    To learn "Strategy" is difficult, because there are different ways of how to play chess. I take the rather "slow" approach - with a twist: I start complications early, and then when my opening knowledge is gone, the position has stabilized at this point. So I chose the opening 1.Nf3 d5 2.c4 - after which there are several possible moves for black, and I need to know all of them. But it can be learned, and the ensuing complicated tactics are at least familiar to me over the course of the next 5-8 moves. It often flatens soon, and I am in strategy waters. We can not avoid tactics if your opponent really wants them. But seeking them early makes them less dangerous.
    And of course I try to calm the position down, and not poor more oil into the fire.

    This isnt really getting better in chess, you might argue. But it is making the most of what you have and what is possible for you to accquire. You can get a bit better in CT blitz rating, but not that much. But you need to find how to make your strategy knowledge pay out better. I thought for a while that the KID would be a good choice. It is not in our case! It makes it easy to play the opening, but the tactics start from move 10 on, and it gets more and more wild. This is the opposite of what you want. Instead - the reti requires you to learn all responses, but you can work on steering the game into known strategic waters. For black I think the O'Kelly is a good choice. The tactics are also early, but it is a good opening strategy wise from move 10 on. Example: 1.e4 c5 2.Nf3 a6 3.d4 cxd4 4.Nxd4 e5! (only that way played is the way it gets in favour of "our" way: seek complications now, but have the advantage to know the position. Then, after material gets reduced, you are in "easy tactics left only"-waters, and strategy becomes important. I also play a lot the Nimzo-Defense, and the Rat-Defense. Both see complications early, but become hopefully able to manage later - with little tactics left.
    I know, I know, this is not really becomming better. But it is about "winning more games". you need to understand the difference of "becomming better" and "winning more often without playing any better".

    ReplyDelete
  5. P.S. Sorry, got the O Kelly wrong. After 4.Nxd4 first 4...Nf6 5.Nc3 e5! (if white plays the common error 5.e4-e5? then 5.Qa5+ with Qxe5 next and a won pawn).

    ReplyDelete