Deep analysis exercises, or dare I say “Stoyko”?

I left a comment here about a puzzle I did this with. Now I did one with the game Timman-Hulak.

This exercise has improved my depth, accuracy and efficiency. At faster than G/90 time-controls, it may not make so much of a difference if the opponent is not letting you think on their time (the death of chess, fast games). At G/2 or longer, it should help a lot.

So far, this exercise has gotten me from looking three moves deep to five moves deep, but that is not good enough because there are some where seven moves deep is needed, and I think that is a healthy goal, seven moves.

I haven’t been using paper and notes, since one can’t use those at the board and must visualize. One can take notes mentally/verbally to ones self, however, and so this is how I am accomplishing that.

This is not “fad” stuff IMHO. The MDLM is really “study your tactics!”, just as much as this is about “analyze deep enough to where you truly reach a quiescent position!”

Incidentally, none of this is new. I am sure the concept existed before Dan H. was even born. hehe. I first head about this in the mid-90’s by an A player who said that a Master had told him that the secret to improving is not to go over a lot of books, but rather to take one position and study it for a couple hours – and I am sure that was an old quote even then. So no, the “Whoever said it on the internet first must have said it first” thinking is not correct (must be a “Gen Y” thing – whoever tweets it first wins!)

Think about it, back in Botvinnik’s day there was no Fritz, they HAD to do this. It’s easy to say “Oh, Botvinnik was a genius! (and therefore I can be lazy)”, but closer to the truth is that Botvinnik probably worked his butt off to acquire those skills the old-fashioned, time-honored way. He didn’t become World Champion by reading “Chess Life” articles, learning a new “trick” here and there, but rather by hard work at the board. And like they say, “you play how you practice.”

Advertisements

19 thoughts on “Deep analysis exercises, or dare I say “Stoyko”?

  1. Hey Linux Guy!

    I love doing those type of exercises (the long think over one position-Styoko if you will!)

    I DO write things down as I think this helps to clarify ideas I might have both wrong and right. Just as I remember something that I have written down better then something I have typed in. In fact the process of looking at a position has gotten better for me because I have been writing down the steps as it were.

    I have even been using this process when doing post analysis on my own games. On the most important moves I set up the board and check it out for awhile and try to figure out where I went wrong and what would be better.

  2. TommyG, I tried looking at one of my tournament games, analyzed it with a board once. It was very productive. 🙂

    Timman’s book is very good for doing deep/Stoyko analysis with.

    The thing that I miss the most are all of the pins and tactical devices that they throw in. I analyzed one position for a while where Timman planned on an exchange sac, and it worked! Later in the game Timman did sac the exchange for pawns. Now when I look at a rook I just wanna thing “Use one rook for exchange sac plus getting free pawns and infiltrating, then use knight, queen, pawns, with possibly last rook altogether now.” It’s like I just wanna use the rook as a sac device. It all depends on the pawn structure, that seems to be the key, it’s a great way to ruin a pawn structure and then invade.

    The stronger I get with tactics, the stronger my analysis should be, which makes me think I should probably go back to finishing the tactics section on pins, but it’s definitely a balancing act. If one only studies tactics, then endgames begin to take on a bit of a foreign feel, and analyses may become too shallow, or only “tactics deep”. It should also be pointed out that most of the tactics these GMs are looking at are tactics on defense, or how to sidestep tactics. Overconfidence in tactics where one only finds their own shots (if that is even done accurately) is the quickest way to a lost game. 😉

  3. Hey Linux Guy,

    One of the cool things at Chess tempo is that do have endgame problems and you can customize the sets (if you pay) and the database has material searches so you can search out positions in master games are like the ones you want to work on.

    And it is a balancing act! Tactics are great but it helps to get good positions to use them! Good positions are great unless one hangs a piece. 🙂 etc. etc.

    The endgame is the best though. I realized I had some problems with certain KG vs. KP positions if the other side’s pawns was advanced so I found the set of problems on Chess Tempo that deal with those and have been practicing those daily!

  4. That’s odd, I don’t know how they would make up an endgame exercise, so I’ll have to try it. Thanks for the suggestion, TommyG! 🙂

    You know, I have Nunn’s book on Beating the Sicilian, and it is great material in there, but I have to agree with your point-of-view that this opening study is mostly an exercise for Experts looking to make Master, or more likely Master trying to make IM, who doesn’t want to get caught napping vs. a lower-rated player.

    I should visit ChessTempo. 😉

    Okay, just visited there and got the first 6 or 7 that I just did right, 1727 rating at the endgame, is that good? 😀 Okay, 1762 now.

    I got a super-easy one wrong and that killed my rating. Only missed 1 problem out of the ten limit for 1702 rating. The only interesting problem for me was that I forgot about RollinPawns “building the bridge” lesson in the Philidor’s position, but it kept saying all of these other mates that I saw would take longer. I hope the tactics section doesn’t cut off like that.

    Appears there is no limit on their regular tactical section, which isn’t as fun, I got 1342 rating at tactics. My eyes detest the burning retinal scan required to solve a tactics problem by looking a screen. it is ten times more fun at the board, or even from a book.

  5. Hey Linux Guy,

    You can also set the preferences for hard, normal or easy puzzles. And as I said as a paying member you can even just specify endgame puzzles of KG vs KP for example (which I am doing right now)

    The other cool thing as a paying member is that you can then search the Chess Tempo database by material, or player etc, etc, and find many many types of positions for Styoko analysis! All without having to get chessbase! 🙂

    I agree it is more fun to look at a board for tactics but the reality is that during the school year I just don’t have that kind of time. So I save my board analysis for Styoko problems, my own game analysis or some problems for Chess gems.

  6. Ah, so you are another school-teacher who plays chess.

    I’ve been going over my round 6 game. The endgame possibilities were immense, so I spent over an hour shooting through lots of moves and variations with Crafty, which is enough endgame for me and very pin-pointed. 😉

    My real problem is that I didn’t finish him off with the doubling rooks on h-file tactic that I had originally planned but didn’t look through. It was my undoing not to put the game away right then and let a man, who looked mid-seventies and probably been playing chess for 50 years, into an endgame. An easy tactic would have bypassed the need for all that. I think I need to get stronger at tactics and study my own endgames where they arise. In general, the main way to improve at endgames is probably to save time for them on the clock! Getting better at tactics and getting in the best tactical shots early on will make the endgame go faster from a more decisive advantage.

  7. For the endgame problems were you doing the theory set or the practice set? I believe the practice set will allow you to play lines that are good, but not the best.

    I believe that within certain parameters that alternate wins are allowed in the tactics sets. There is an option in the preferences to avoid problems with alternate wins, and think I have run across some before.

    What kind of limit (or lack of in this case) are you talking about with the regular tactics problems?

    TommyG, What does the G mean in KG v KP ?

  8. I suppose it was the practice set. I am not really interested in that site because time spent counts against one’s rating that they give, and I am interested in getting a problem right, however much time that takes. If I am going to spend time, I prefer to look at a book or a board, not the screen that they give. When I look at their board visually, I am constantly getting the diagonals wrong.

    I am going through the book on tactics again, or this time I will go straight through and solve all of them. That works well for me. I do know the theme, but it doesn’t strain my eyes to look at a book, even when the diagram is small and low-resolution.

  9. If you go into the preferences and set the tactics problem set to standard instead of blitz (no idea why they default it to that for new users), the time will not effect the rating. It is still tracked and used in another formula they have for determining a very rough FIDE estimate, but that is all.

  10. That makes it useful to me then.

    I like the Comb. Challenge book that I have. There is a section at the end that doesn’t tell the theme, so that I can try to set those ones up OTB. Still, I am going back through the other sections. I am surprised how I can get rusty so quickly, probably because I haven’t or need to practice tactics for a year, or some other solid period of time.

    It can take a long time at first. The first couple may take me 12 minutes, from my book, but I can also get going to where I can solve maybe 15 within an hour on one’s I have solved before. MDLM was right that one has to have a strong tactical foundation to prepare oneself for playing a high-level chess game, particularly at today’s time controls. Not all games are tactical, but there usually are tactical shots whenever I am playing down in rating, Don’t want to draw those games, not just for ratings sake but more importantly for tournament standings. How am I supposed to play a G/60 if it is going to take me 12 minutes to find a fundamental tactical shot?

    I am sure that many of the local 1400 players love the faster time-controls as they get a chance to see a player a couple classes higher whiff on a tactic, and then miss one of theirs as well.

  11. I am reading this “Rapid Chess Improvement” book for the first time and _the obvious_ thing struck me. He says you have to do this many in this many days, then this many, and time yourself not to take/spend too long.

    It’s obvious that to sell a book, you have to put up a faux goal or faux adversary. Why not instead of doing this plan, simply DO TACTICS EVERY DAY? Take your time, solve them thoroughly, there is no need for early outs, and if you can see more than one solution, then that’s great too. What’s the rush? It is about your knowledge-holes after all.

    People want a goal, because the first thing they want to know is when they can stop doing it and simply BE a Grandmaster. lol. Probably any sincere tactics study will help before a tournament.

    Oh no, I am on page 66 of his book, first game, it is going through his thinking process. He sees that 13.Bxg5 fxg ruins Black’s pawn structure and plays it because his “g5 bishop unprotected”. This is way funny. Not a bad way to think at some point during a move, but not how one should base their move. I can see that 13.e5! wins the d6 pawn cleanly by force (tactically, I might add!) in all variations. I mean, if I can do this casually, then I am already suspecting that a big component of his rating was due to superior clock management.

    I’ll post the position if someone wants to see it, but I don’t know that anyone follows this blog/rant that closely. I have to switch to my other computer to post it because Ubuntu doesn’t offer me any chess analysis engine, so I can’t even enter a game on Xboard.

    I take that back 13.e5 doesn’t win a pawn, his move was better than 13.e5, but 13.f4, threatening 14.e5 looks almost crushing. For someone with “book knowledge” it’s not hard to see that Black is not well positioned for anything about to happen on e5, plus Black cannot play ..e5 for tactical reasons.

    BTW, the Bg5 is not “hanging”. For instance 13.f4 Nd5, 14.exd BxB, 15.d5xc6 Qxc6, 16.Qxd6 wins a pawn, and the Bg5 is shooting at air. Black’s queen is, too, for that matter. +/-

    Wow, pg.75 paragraph 3 defends his simple-minded style of playing basically by saying that at the class level it is better to devote more time to studying tactics (than post-game analysis or in-game analysis of variations).

    The game was interesting. Basically, as White, he was trade-trade-trade K.I.S.S. No wonder his opponent became simple-minded themself during the endgame and let White slip in a mate threat, which clinched it by winning a couple pawns to avoid the mate. Wow, what a psyche-job. hehe.

    He even admits after the game that he lost his advantage during moves 16-so and so, and that Black did likewise later, and that they should both try to improve this part of their game. My guess is that Black was in time-trouble, but White had time to find a threat other than just winning a pawn.

  12. Blindly following his plan will result in someone hitting the same wall he did. He basically gave up on competitive chess once his tactics training had taken him as far as it could go.

    Anyone thinking of buying the book should just go to chesscafe.com and get the original articles The book is basically the articles plus fluff from what I have seen.

  13. Hey Snits,

    G in KG was supposed to be a Q as in King Queen! 🙂 My bad.

    Hey Linux Guy,

    I do the practice and standard settings for endgames and tactics so time doesn’t matter. But I don’t really care about my rating at Chess Tempo as it means nothing. I just track my percentage I get correct.

    I have found my tactics get better when I do about 5 problems a day there (on top of other tactical training)

    And I always thought the problem with MLDM’s book and plan was the end goal thing! As a musician and college music teacher I ALWAYS tell my students to practice their rudiments (scales for drums!) but I never ever try to get them do some one day blow out of their rudiments as the end goal. The end goal is just to get better.

    I think the end goal is the reason a lot of MLDM followers experience some sort of burn out.

    I DO think tactical repetition on certain basic patterns is sound as it is the basis for learning any skill that requires in the moment responses. (sports, music, chess)

  14. My advice when it comes to tactics is to take your time and try and get it right. If you need a book to learn the patterns, then use another book where you don’t try and “learn the patterns” once again, but simply aim to get it right that time.

    The lessons have to be conceptualized. If one is only remembering patterns in short-term memory then what happens after 1,000 tactics in day? Surely an immediate big cash tournament, ala The World Open, as that huge effort is going to have a short shelf-life.

    Good advice Snits and TommyG! 🙂

    The goal of MDLM all along was to win The Word Open for $$$. He probably had bills stacking up, and I doubt that his real goal was to improve ratings points. He pulled out of some high-money tournaments once he lost the first round or so. IOW, it was about the $$$. The book, again, $$$ (not that it probably made him much, but that’s all relative to how poor one is at any given moment). The purpose of ratings-points was more likely an indicator to him that his plan was putting him on the right track to win a tournament. Sure, no one hates being called an Expert at chess, but it doesn’t seem as though that were the primary motivation to me.

    So, to sum up. If one does 1,000 puzzles in one day, a week later that was largely for naught. After all, a day is just a day, and he is the one to say that it takes so many thousands of hours and it’s some unavoidable hard fact. He was going for the performance boost “energy drink” of chess before a tournament!

    I just missed the quickest mate in a puzzle where one has to sac rook, then queen for mate. What is the pattern? There is no “pattern” here per se (double-attack?). It’s a concept, “Can I sac two pieces and still have enough to mate, what is the least and most precise material needed here for mate?” Does one get that speeding through an exercise looking for a pattern? If one can get and retain a concept long-term, that quickly, then they are well on their way, and doing it slowly, there is no penalty for coming to the same conclusion a little later! This is just for practice and learning. “Ah shucks, I have to move now” doesn’t serve a purpose while practicing for a mental game like chess.

    Again, even if it wasn’t for the money, the money much like ratings-points was used as a “validator” of the success of his training plan.

    He didn’t even study his games that deeply according to his book. He studied “some moves” and let Fritz auto-annotate the game, then got back to his tactics studies. Which only proves that no one has unlimited time and energy to do it all, not even the unemployed.

    It keeps crossing my mind that MDLM’s study plan is probably best for, if anyone, blitz players. It seems he is against blitz so as not to damage the thought process, but if he had been a little more open-minded about it, I think he might of realized that the style of his game and his study plan are really best suited to winning a blitz side-event, or a blitz championship.

    I just solved a problem #844 in my book (I jump around) that I had to visualize 14 plies ahead to verify that it works (it’s mate). The book gave the first 7 plies (I also had to inspect an 11 ply mate in a sideline not covered). I’d love to know what the “pattern” is to this problem other than work your butt-off looking deep into the problem (the first move is relatively obvious). 😀 Well, it wasn’t hard to do after a cup of coffee, I must confess. 😉

  15. I envy your patience and persistence. 🙂
    Analyzing the game without pen and paper is close to the real game, I am wondering how much more you are going to see. Sometimes I think that postmortem (which is close to this) is practically useless, except getting just a few ideas and lines from your opponent.
    I showed the game that I won yesterday to one girl and I didn’t see anything new. I came home, ran it through Fritz – here we go… The thing is, I am not sure if these Fritz’s ideas are then stored in my head.

  16. RollingPawns, thanks for your insight! 🙂

    A lot of what I am trying to do is form healthy, point-getting habits at the board. Funny how much more relevant that becomes when a board is placed in front of oneself. 😉

    First thing I have to do when looking at a position, okay I am lousy at this and I think RollingPawns’ brain is automatically wired to do this from all his blitz play, is:

    1) Threats for both sides – look at every pawn and piece that is attacked. Say the word “threats” to yourself, then look for them; yes, this is a checklist.

    2) Coverage – say “coverage” to yourself, now look to see which pieces and pawns are covering which squares. 5-10 seconds. Purpose of this is that it’s easy to see Rxh2 as a threat, but miss that there is a Rb2, or a bishop on the other side of the board that is actually covering that square. It saves time later on not to miss anything ‘obvious’ up front.

    That concludes the “looking at the position” portion, and noticing everything.

    3) Tactics – say “tactics” to yourself. You may have done the preceding two stages in as little as 10-30 seconds combined, yet it is still vital that you have done so). Tactics is the part where you look for loose pieces, deflections, interference, pins, back-rank mate threats, all that good stuff.

    The key here is it’s not about just winning material – you can see this part at first, let’s call it “Stage 1” – but more importantly then see if your opponent has any counter-threats, counter-attacks. It doesn’t help to attack his king or pieces if the opponent’s defense sets himself up for an even better or clarified attack on your position.

    Some tips for calculation. I am already so used to doing this that I find it easy, but here it is:

    Your move is ply #1. When calculating, say each ply# after visualizing the move. An odd ply means that you know that it was your move that you just made, so the even numbered ply must be his move. Also, note where the calculation braches. If your opponent can either take with rook or queen on ply#4, then say “ply #4, either queen or rook recaptures” to yourself so that you remember to come back to ply#4, and just saying this will make you rest on the position long enough to remember it (no need to recalculate the previous ply again) and then calculate the other variation (reminds me of recursion in programming).

    I looked at some of MDLM’s tactics for winning the World Open, and the thing I was most impressed with is not just that he had become a fearless sacker in a calculable position, but that he must have figured out some initial tactics, took his opponent’s reply into account, and then tweaked his threat to where the opponent truly had no defense against it. It’s easy to go for a tactic that doesn’t work, without first looking harder at to see if it needs refinement based on opponents best defensive reply.

    The procedure I gave above is a lot easier than it sounds, calculating once instead of re-calculating many times. This already saves time many times over by calculating slowly and accurately one time. Another neat thing about counting ply is that you know depth of your analysis. For example ply 13 means, 6 1/2 moves, or 7 of your moves.

    Naturally, this is not all there is to a chess game, we still have to do the “mindless” strategy part right, openings, set up the position well, pick our spots carefully, make the most accurate move in a sharp position (which isn’t always about just winning material) etc.

    RollingPawns, I don’t think the Fritz ideas store so well, if it’s not something that we practiced finding (instead of simply being told). I feel like you about post-mortem analysis. It can bee good for the lower-rated player if the other player is +100 ratings points.

    In my last Scotch game, I remembered to attack with f6, but sorta barely, even as I had analyzed it previously with Crafty a long time ago, and tried different moves, and even then it starts out with an “I’m not sure, but I think I remember something about….” 😉

    Unfortunately, I can already tell you the results of my new method. If there is something, an obvious enough attack, one can calculate deeply and accurately (if one tries honestly, not lazily) so that it should be possible that one is not afraid to go through with a long attack.

    The downside is what % of chess is this? It may be some of the most important points of a game, yet still one has to be diligent. Then there are the other 90% of moves which are mostly about paying attention and finding the simple things. So paying-attention, you can’t eat this in a bar or drink this in a cup(?!) is still the most important requirement and 1900+ are usually quite strong and consistent at this, paying attention to all of the little things in chess.

    The nice thing about setting up a problem on a board is that you will probably do these thing and make it into a habit. Doesn’t have to be a problem per se, one could do this while going over a complete game. Sometimes when looking at a book or diagram online, there is a greater temptation to say “scr*w it, does this shot work or doesn’t it?”

    The other thing is that I am just as useless at faster than G/90 time-controls, although I have improved at G/90 time-wise. The thing is MDLM’s system is definitely geared for finding tactics when tired or with little time – I bet he never lost a game on time. It’s not all about “Kotov”, it’s about winning in the here and now with quick moves.

    I think the thing is to study tactics ever day so that they are easy and quick to find, OTB. But when is a tactic the thing? It seems like a tactic is that one move we missed when playing the lower-rated player, that sort of thing, and otherwise it is just about unhelpful knowledge. So, when playing, try and forget that you ever studied tactics unless you feel one is in the postion already. 😉

    It seems to me that the thing that takes time is calculating complications in general, and I don’t know how people blitz around this, other than by having a lot of openings knowledge. Some people can just blitz at a high-level, I guess.

  17. I think that one could study tactics everyday and continue to improve, or at least I could. The ‘Combination Challenge’ book is a little weird as I was noticing that it is missing a section on ‘interference’. Then I start on the section on Smorgasbord, do 20 of those problems (over a few days), and they are all of the interference theme, which goes to show it could have used a bit more editing.

    Once you are onto what the theme is (a big help), often the biggest part becomes calculating, or IOW, the details.

    I think I’ll take a break from studying tactics here, need the rest and time for other things, but I can say that it was beneficial getting used to the idea of making sacrifices, and paying some attention to a clock. The main thing is “How long does it take you to find the idea?” Once you feel confident that something is there, it’s mostly about how much time you want to spend/have to calculate it right, which of course leads to much bigger material gains or mate if you do. A theme becomes easy to spot when you know what it is and have drilled on it for a bit.

    As of now, I could use more practice on some of these other themes, even simpler ones such as pins and knight-forks in particular. That just says that over the past year plus, I haven’t been spending enough of my chess time on tactics. Tactics is just one part of chess, but it’s the most assumed part, that one knows what they are doing here. Other parts of chess, such as understanding strategy, come more from experience, although one always has to validate strategy and positional play via tactics.

    Tactics AND a mental work job such as programming is too much in combination. I think it’s better if you can find blocks of time/energy where they don’t interfere, which may mean only studying tactics once a month, or on weekends, or after a slow work-day, that sort of thing. Doing tactics every day, I find is a burnout, but maybe it’s because it’s too much practice and not enough play, so motivation diminishes. So, I’d say to do it in blocks, throughout the year, as time allows.

    I think that I can study openings just by looking at lines in MCO, etc, not even using a board or computer, but just by seeing it in my head.

    I’ll end this post/topic with this thought. When I went from 1300-1500, I did study some from a tactics book, rather randomly. I would do it between rounds at a big tournament. The thing that helped me most was that it made me much better at calculating, particularly calculating in endgames! I learned the power of making an exact move, so that the opponent would have to keep up by not making arbitrary moves in response.

    The worst thing studying tactics does is encourages me to switch my brain off until a tactics light flashes over the board (or worse yet play hope-chess). By this time it’s too late as chess is more about calculation than it is about tactics when it comes to performance. Chess is a performance sport. One could perform badly and then spot the tactic, and I suspect that these types of players often fool their opponents into playing weakly in response until they get “caught” by the tactic. IOW, they could be 1300 at positional play then 2000 at tactics, and it fools their opponent, and by then of course it is too late in the game to come back as the tactician is if nothing else a “finsher/closer”, so that I could at least understand their making it 2000.

    I have played enough 1900’s to know that energy, and discipline, and will to win, belief in one’s self can get one past them. MDLM had a 2200ish performance rating, but it was against 1900 level and below in a very long tournament. He had already had lots of experience playing against yet even higher-rated opponents at his club, in which opportunities to play them were plentiful.

    Show me someone who goes 3/5 against 100+ higher opponents, and I will show you someone whose rating goes up after each tournament, particularly if they are on a tactics-studying regimen. This girl Alexa went from 1500 to 1650 in the last tournament 4/6. The rating process will frequently smooth out progress so that even if someone were playing at 1900+ on day one. If they start out playing in low quads, it will still take them many tournaments just to rise to their true rating.

    MDLM had multiple coaches and his recommended book list is quite extensive, like 50 books or so, and most are not on tactics! lo, even repertoire books, enough to say his training program, if he read them, would have been more along the lines of the late Ken Smith’s recommendations, who I was originally following (loosely)!

    Soltis was right, calculation is the inner game of chess, it just so happens that a lot of what one needs to calculate are tactical devices. Performance = calculations. Tactics = finding the cheapos and quick, easy wins. For a good tactician, the tactic is decisive, but for a good calculator they will often not come away decisively after the tactic, but can finish it in a strong endgame. That isn’t the ideal, but performance and calculation are still the #1 things in my book, particularly calculation.

  18. Yeah, 10 minutes a day is far superior to some complicated “Circles” nonsense.
    Ambitious folks can do multiple 15 minute sessions. I advise against sitting there doing tactics for longer duration than 15 minutes– there is clearly diminishing returns the longer one carries on.

  19. Hi, Katar! 🙂

    Yes, I agree, 15 minute session once a day is best, twice a day if ambitious. It opens the mind in some way, to the possibilities, but for more exercises than that one may as well set up the board and spend more time. This is because the key to a lot of problems is to factor in the opponent’s counter-play, not just find the trick.

    I admire MDLM for his stamina, but he was clearly training to be a tactical Hercules. It must have gotten easier for him after the first year, and he must have formed this plan early on (as evidenced by his first two sample games where he lost in like 9 or 11 moves, which perhaps game him the impetus).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s