Wednesday, June 14, 2017

Hanging examples on the Tree of Scenarios

My failures are used to hang as examples in the tree of scenarios like balls in a Xmas tree (see the links)
  • Tree Of Scenarios

  • target status
    • Is hanging or outnumbered
      • neutralize counter attack
      • defend after capture
    • Has lack of space
      • cover flight square
      • weakest defender
        • pry the box open
      • block flight square with his own piece
      • chase target into the box
      • squeeze the box
      • chase target into a line of attack
      • cut off from defenders
    • Is defending
    • Is target of duplo attack
    • Is B.A.D.
      • Remove defenders
        • Defender has other duties
          • deflect
      • are the defenders potential targets?
      • add attackers with tempo
      • can the opponent add defenders?
      • duplo attack it
    • Is attacking
  • point of pressure status
    • Is outnumbered
      • ready for use
    • Line of attack to PoP is blocked
      • clear with tempo
    • Is of 2nd order
  • attacker status
    • Has defensive duties
      • remove counter attacker
      • reach the loa while staying in contact with defense
      • defend with tempo
  • check for counter play
    • are there counter attacks?
    • is there a defensive move?

23 comments:

  1. Off topics:
    https://new.uschess.org/news/christopher-yoo-breaks-record-for-youngest-ever-master/
    From 0 to NM in 3 years.

    ReplyDelete
  2. Kudos to young Master Yoo!

    Excerpts:

    His first coach, Wei Liu, who got him to A/Expert level with his emphasis on drilling on tactics and calculation.

    When asked about advice for other aspiring US Chess masters, Yoo advised players to focus on tactics till they reached 1500, and “do them in your head without moving pieces around.” Specific recommendations include chesstempo.com, CT-ART and lots of repetition on relatively simple tactics, like the ones in Susan Polgar’s book.

    I’ve always loved chess (played some tournaments in college) and read books on chess history before Christopher was born, and I remember stories of prodigies like Capablanca learning the game at unbelievably young ages from simply watching others play. This wasn’t the case for Christopher. I tried to teach him chess at age 5 but he couldn’t remember the names of the pieces let alone how they moved so I gave up. A year later, when he was 6 years and 9 months old, we enrolled him in a once-a-week after-school chess program at his elementary school and after a couple of weeks I thought, great, I can now sit down with him and play a game. Unfortunately, that wasn’t the case. He didn’t know how the knight moved and his bishop refused to remain on one color. So I taught him how those pieces moved myself, but it wasn’t easy. But once he got the hang of it, he really took off. Within four months he was 1st grade champ of California. In another 7 or 8 months he was at 1700/1800. A couple of years after that, master.

    HMMM. . .

    (Have a large dose of "salt" before reading the following screed.)

    Nope, absolutely impossible to achieve master level in 3 years; I don't care how young and "talented" you are. We have it on the "good authority" of the "experts" that it takes at least 10,000 hours of deliberate practice and the internalization of at least 50,000 (maybe even as many as 300,000) positions to reach master level. That means young Master Yoo was putting in at least 9 hours per day, seven days a week, for those 3 years, studying and playing chess. (For at least the first few months, he was attending a ONCE-PER-WEEK after-school program.) He was also memorizing 5 positions per hour (if he stored 50,000 positions in his LTM), every hour he was studying. The number rises to 30 per hour if he stored 300,000 positions, or 1 every 2 minutes. Pretty darned focused for a 6-9 year old, if you ask me! (Yeah, I know: nobody's asking!)

    On the other hand, he must be a "genius kid" so maybe that's the differentiating factor. Nope, his Dad seems to think that he's NOT a budding Capablanca: just watching adults playing (poorly, of course, since they ARE adults) and then pointing out where they messed up. (Maybe we adults should try that approach for rapid chess improvement. You can call it the "Crazy Bob Improvement System™" or CRABIS™ (pronounced "Crabbies").)

    Or are there SOME 'received truths' that might, just might, NOT be true?!? Could it be that perhaps he is a genius computer whiz at MIT, and he's merely following in the esteemed footsteps of Dr. MdlM? 2 years from 1700/1800 to 2200 seems to be a rather "fast track" for anyone.

    I note, for the record, the emphasis on drilling tactics, tactics, TACTICS and calculation!

    I wonder what percentage of chess could be tactics? Could it possibly be as high as 99-100% at the foundation level?!?

    ReplyDelete
  3. His name in the Internet is yyoochess
    https://www.chess.com/member/yyoochess
    His rating at the tactics trainer is 3015 but he did only 7700 attempts there
    https://chesstempo.com/chess-statistics/yyoochess
    Attempts at chesstempo
    1428 standard tactics
    3098 mixed tactics
    498 blitz tactics

    He did from time to time an hour of tactics

    45 endgame
    139 Custom set ( like "Problems always failed" )



    The 10 000h rule is just a statistical average. Kids learn better. They dont need (so many) repetitions. They learn a language form 0 to 100% in "notime". Their brain is not fully developed and develop according to the environment they experience.

    These statistics are not suspicious for me.

    ReplyDelete
  4. @ Aox:

    Please note the following statement in my previous comment:

    (Have a large dose of "salt" before reading the following screed.)

    My apology. Perhaps that idiomatic statement above does not translate well from English. It has the connotation of being sarcastic, somewhat similar to "tongue in cheek".

    I wasn't questioning Master Yoo's accomplishment, which is fantastic! I was merely opining that the estimated statistical "averages" which have become established scientific "fact" are not quite as rigid as we "average" people might assume. (This is the difference between "wild ass guess" and "scientific wild ass guess" estimates.) Think of the Reinfeld value system in the same way. On average, it works. For any specific case, it sucks. Or, as Dr. Lasker put it, it is only true if we always accompany it with the Latin phrase "ceteris paribus" (everything else being equal). In too many cases, "ceteris" does not connect with "paribus" (to quote GM Rowson).

    The statistics you quoted are very interesting in highlighting the big difference between Master Yoo and almost all adult chess improvers. How many of us have done orders of magnitude more tactical problems than Master Yoo?!? Yet, we do not show that kind of improvement in that short a time. Crap, we don't show that kind of improvement - PERIOD! Definitely "food for thought" as to the reason(s) why not. Obviously, one of those reasons may be the relative neuro-plasticity of the "blank slate" younger mind.

    ReplyDelete
    Replies
    1. Master Yoo is learning chess for 4 hours a day thats 4400 hours in 3 years and he is """"only"""" NM. Maybe he will need an other 5600 hours to become FM ? ;)
      I checked a few games of his. I was thinking he might win these games by (tactical) skill ( young players are usually super strong in tactics ) and not by knowledge but his openings i checked are perfect.
      So maybe he did memorize 5 positions per hour...

      Delete
  5. This fellow did not learn chess till his twenties but was a competitive sportsman. http://www.chessgames.com/perl/chessuser?uname=fm%20avari%20viraf He seems to have accomplished much for a late starter.

    ReplyDelete
    Replies
    1. I note that he was born in 1946 (two years older than me).

      Thanks for the link! I do not have this book and had not seen it. I certainly am a believer in Petrosian's Law (as can be seen above).

      FWIW: Petrosian's Law

      Opponents always have some kind of weakness in their position, even if imperceptible. Against that you have to play.

      Thanks!

      Delete
  6. IM Emory Tate
    GM Melikset Khachiyan
    NM Bruce Pandolfini
    GM Cristian Chirila
    GM Daniel Naroditsky
    The Berkeley Chess School
    The US Chess School
    Greg Shahade
    IM Steven Zierk
    NM Richard Beale
    Dave Ceponis
    Bill Skog
    GM Jesse Kraai
    Stephen Shaughnessy
    GM Gregory Kaidanov
    IM Emory Tate
    NM Cameron Wheeler
    Lauren Goodkind
    Rochelle Ballantyne

    An impressive list of coaches, teachers and advisors.
    I wonder how we would do with such help.

    ReplyDelete
    Replies
    1. "I wonder how we would do with such help."

      According to a scientific paper i read some weeks ago, there is a difference between younger and older chess-player.
      While kids benefit from chess teacher, playing chess, deliberate training and so forth, older player benefit only!! from deliberate training.
      So according to this paper: this help would have helped you 0.

      Delete
    2. http://www.psych.utoronto.ca/users/reingold/publications/PDFs/Charness.Tuffiash.Krampe.Krampe.Vasyukova.2005.pdf

      Page 160:

      Age subsets
      It is also necessary to consider how the chess activities investigated here may vary in their predictive strength across different age periods. Based on a longitudinal study of grandmasters (Elo, 1965) and an evaluation of cross-sectional trends in a complete cohort from the US (Charness et al., 1996), there is significant evidence to argue that the peak age of competitive chess performance occurs in the mid-to-late thirties. Thus, we partitioned
      players in the combined sample into those below versus above age 40, and re-computed the simultaneous regression equations with the six predictor variables. Results for the regressions of current skill ratings on the six chess activity predictor variables in each age sub-sample are shown in Table 5.
      In the young adult partition of the combined sample, cumulative solitary practice,
      cumulative tournament play, years of private chess lessons, and current serious study alone were significant predictors. In the older sample, only serious study alone was a significant predictor.

      Delete
    3. Given all sorts of training methods that have been exposed on this blog as being inadequate, the paper is probably right.

      Delete
  7. My Chesstempo rating and OTB rating are within a few points of one another. His Chesstempo is 2300. So...now...if I can just make Expert on Chesstempo then.....

    For three years of chess, that is not that much experience compared to say an older Class A player playing chess for most of his/her life. So, if he can overshoot the mark on Chesstempo, by a hundred points over his OTB rating, then that formula should work for me too.

    Nevertheless, it's sad how rating is more valued than experience in chess. It's weird to see a younger player think he knows some opening line because he is an Expert, then I go home and the computer always says I'm right and he's wrong, it's just a question of a simple tactical slip-up somewhere, and not losing games over simple strategic considerations (except to more experienced, i.e., old like me, players!) in my case.

    ReplyDelete
    Replies
    1. The chesstempo-standard-rating has nothing to do with our chess strength. A player rated 1600 OTB had a ct-standard rating of 2400, he did think about each puzzle several hours up to some days.

      This NM was doing : all few days an hour of chesstempo. I guess this was not his training but just a measurement of his progress, possibly a method to find weaknesses in his tactical thinking.
      From 2 or 3 hours of training in tactics even a 7 year old boy cant improve that much .. i think..

      Delete
    2. From 2 or 3 hours of training in tactics a week

      Delete
    3. I agree that 2 to 3 hrs of tactics training a week is next to nothing for the serious improver. However, the most important thing about tactics training seems to be to stay fresh, in the sense of practicing frequently rather than in big sessions where you really try to push your skill level. The biggest benefit of tactics is as a maintenance role for your chess rating, rather than to build incredibly from it. Nevertheless, Masters will be stronger at tactics than non-Masters.

      The thing that bugs me a little bit about studying tactics is the notion that every position is a play-to-win proposition, not a play to draw, and rarely is an answer an un-forcing move. It's usually about finding the most forcing continuations. Good chess is about good defense, so if you are always looking for the big-kill, and swing that haymaker and miss, then your opponent can take you clean out on the spot.

      Delete
    4. It seems that you did not read the article ( https://new.uschess.org/news/christopher-yoo-breaks-record-for-youngest-ever-master/ ) careful enough

      "When asked about advice for other aspiring US Chess masters, Yoo advised players to focus on tactics till they reached 1500, and “do them in your head without moving pieces around.” Specific recommendations include chesstempo.com, CT-ART and lots of repetition on relatively simple tactics, like the ones in Susan Polgar’s book."
      And Susan Polgars book has chapters about drawing tactics and so forth too.

      Delete
    5. I know we've had this discussion before, so. . .

      I think we are all in agreement that 2-3 hours per week spent on tactics is a relatively minimal amount of tactics training, if that's what it was. It seems "unfair" that Master Yoo might have get so much out of so little training time.

      I think it more likely that he gained much more in TACTICS from other areas of his training than from those few hours spent on Chess Tempo. I concur with Aox: this "training" activity was much more likely to be measurement of progress or identification of weaknesses.

      On the other hand. . .

      Quantity of problems solved is NOT correlated with improvement in tactical chess skill. (If it did, there are several of us who would be GMs by now.) I think if you checked all of the adults who have pursued the Seven Circles of Hell, you will find that the sheer number of problems "solved" provides little or no correlation to playing skill as measured by OTB rating. At the lower ratings, it IS possible to increase rating up simply by becoming more familiar with tactics. I estimate that progress will top out somewhere in the vicinity of 1600-1900 USCF (1500-1800 FIDE).

      I think there is much more to be gained from targeted tactical practice, aimed at specific weaknesses in our skills. Nothing is gained from repetitiously practicing the skills that have already been cemented into the mind. For instance, if you know how to gain/retain the opposition in a K+P vs K endgame and know the key squares and know the maneuvers to promote the Pawn, then you will gain exactly NOTHING from repeated practice by shuffling the three pieces into different positions and then practicing against another player (human or computer). If you can move a Knight around the board at lightning speed, nothing is gained from repeating that process over and over.

      In short, if you are not "seeing" something additional to whatever it is you already know when you are "practicing," then you are not making progress toward improvement in that skill. That is one of the essential factors in the "deliberate practice" paradigm. You must be working close to the edge of YOUR skills. That does not necessarily mean practicing problems that are somewhat higher rated (within 100-200 points) than your current rating. Although the problems may be higher rated, they may be so because other players have difficulty "seeing" the solution. You, on the other hand, may have already inculcated the important parts of those problems and find them relatively easy to solve. Ergo, even though higher rated, you gain little or nothing from solving them (other than an increase in your tactics rating) in terms of actual playing SKILL.

      Delete
  8. I've solved less than half the number of tactics than this youth has, and I certainly don't want to detract from his accomplishment. Nevertheless, with all this talk about neuroplasticity, I want to say that chess gets it backward. It should be an accomplishment for an adult to step outside of their boundaries to achieve something of artistic merit in chess.

    Instead, chess grabs in the young, the artists, the two groups with fewer established boundaries (outside the one's they mimick, like their parents) than the social norm. It's the fact that children have not yet formed all of these "social" boundaries, and figured it out yet, that makes them so pliable towards chess. Once those kids begin to form social boundaries outside of chess, they often leave it, or they leave it for other artistic endeavors such as music, poker, computer programming.

    ReplyDelete
  9. Master Yoo's Chess Tempo (yyoochess) stats:
    Rating 2330.1 (1428 standard tactics)
    Rating 2646 (3098 mixed tactics)
    Rating 2249.2 (498 blitz tactics)

    Robert Coble's Chess Tempo (crazybob) stats:
    Rating 1547.9 (2043 standard tactics)
    Rating 0 (0 mixed tactics)
    Rating 1616.2 (192 blitz tactics)

    According to Chess Tactics Server, I have solved 35,374 problems with 78.9% success rate with current rating of 1502 (highest rating ever is 1634). I've solved" several books full of tactical problems, and Chess King, CT-ART 4.0 and, ...

    Where did I go wrong?!?- ROTFLMAO!

    ReplyDelete
  10. Have to leave right now, but I'll try and think of something when I get back. Didn't even know or forgot there was a Chess Tactics Server. My live USCF rating at this very second would be approx 1814, which is close to the Chesstempo rating (Chesstempo is hard!) Here are my Chesstempo stats:
    User: chessanalyst

    Total Solving Time: 47.7 Hours (excluding outliers)

    Star Ratings Made:2
    Comments Made:2
    Tags Added:0

    Stats for standard tactics
    Rating: 1821.4 (RD: 50.09) (Best Active Rating: 1906 Worst Active Rating: 1463)
    Active Rank: 1471/12378 (Better than: 88.12% Best Active: 858 Worst Active: 5254)
    Problems Done: 875 (Correct: 559 Failed: 316)
    Percentage correct: 63.89%

    Stats for mixed tactics
    Rating: 0 (RD: 0) (Best Active Rating: Worst Active Rating: )
    Active Rank: Not Active/1151 (Best Active: Worst Active: )
    Problems Done: 0
    Percentage correct: n/a


    Stats for blitz tactics
    Rating: 1500 (RD: 350)
    Active Rank: Not Active/2264
    Problems Done: 0
    Percentage correct: n/a

    Stats for endgame theory
    Rating: 1702.6 (RD: 584.55)
    Active Rank: Not Active/768
    Problems Done: 12 (Correct: 11 Failed: 1)
    Percentage correct: 91.67%

    Stats for endgame practice
    Rating: 1500 (RD: 350)
    Active Rank: Not Active/395
    Problems Done: 0
    Percentage correct: n/a

    ReplyDelete
  11. Okay, so, there's something that doesn't make sense in all of this, Robert. You have a high solve rate on that Chess Tactics Server site, basically 80%. On Chesstempo, I am basically 64% and have a higher rating than you on CTS. That Master Yoo kid, I noticed on his Chesstempo summary that he'd have like patches of ten in a row wrong that would count for like -1 or -2 rating points.

    One thing I noticed, wow, over 25 years ago at a math class (I sucked at math so take this with a grain of salt or not) is I saw a lower-math class come in, mostly adults, and they were asking more questions than the kids do at higher-levels, so I asked myself "Why is this?". It suddenly occured to me, that as people get older they need more "sure-ity". The older a person gets, the more they want to know something for sure, and if it's a sure thing, then the generally know how to use it better than younger people as well.

    Little kids don't have jobs, they don't need to know anything for sure, and so they find it acceptable to just do what comes to mind, living off their abilities. They don't need to know how they do what they do, they just need to do "it".

    As young players, 1.e4, the direct approach to checkmate, often seems the best option. Later, most players switch to 1.d4, and 1.c4 as they begin to realize that nothing is so sure, therefore, let's make our opponents even more unsure than ourselves, so that there is safety in no one having a feeling of sure-ity, the way you get that feeling with 1.e4.

    Okay, I rambled, only perhaps gave your question a glancing blow.

    ReplyDelete
  12. My apology: I should have explained WHY my CTS rating is so low. It's simple: I don't guess. I really don't care what my rating is because I'm training myself to "see" the important features in a position. If I can't "see" the solution immediately all the way to the end, I intentionally spill extra time working out exactly what is important AND trying to "cement" it into my mind. Since it is timed, and usually there is less than 10 seconds before you start losing points, I rarely advance above 1500-1600.

    Your points are well taken!

    ReplyDelete
  13. Oh, I didn't know CTS deducts points for time (I meant my % correct on Chesstempo, not CTS). You are right, that would be useless for a rating then.

    I've come to believe it's best to take lots of time solving on Chesstempo, etc. Once in an OTB rated game, it's ideal to whiz through most of the game relying on tactics, strategy, openings built up from slow-study. Then spend most time OTB navigating the endgame, seems to be a worthy strategy of play, worked for me last night against another player with nearly the exact same rating as myself.

    ReplyDelete