Stories of my demise have been...ahh. you get the picture.
Ok, so I'm 3 rounds into our Fall tourney here at the local club. The strange thing is that the first actual game I played was last night in round 3. In the first round I took a half point bye because of work (.5/1). In the second round I was paired as white against a lower rated player (about 1200), well he was a no-show and so I racked up the full point and got a night of watching the top boards play (1.5/2). What this means is that in round 3 I was going to be paired way up, in fact against the highest rated player in the tourney, at 1850ish.
Well, I played hard and went down swinging. I'll post the game here later on today, but here are some short thoughts based on the game and the big 5.
Time management - I was pretty happy with this aspect of my game. I was up on my opponent on time for most of the game, finally getting below him at the end. There are some situations where I'm still burning too much time on things I should be able to figure out easier. But it wasn't really a factor in this game.
Tactics - I was very happy with my offensive tactical vision. In fact I pulled out a nice tactic about halfway through. However I'm still not seeing my opponents tactics as well as I should and allowed one to be executed on me as well. This relates to "hope chess" problems I have.
Activity - It's really what this game was about. My rooks inactivity at the end was a major issue that turned the game in his favor.
Thought process - This is still my major Achilles heel. It all still relates to hope chess. I don't see my opponents replies to my moves all the time, and it causes tactical and positional errors. In fact in this game I allowed a piece to get trapped because I didn't see that my opponent could attack it, and then in the endgame I allowed him to win a pawn and get another passed pawn because I didn't see his moves that took away my defenses.
General principles - I made a couple of errors here but mostly they were small errors in position and weren't what the game hinged on.
Game to follow soon. All in all a good effort against someone 400 points higher than me. In fact, I think I was better for a little while, but in the end, my lack of "real chess" is what hurt me.
More to come next week!
Tuesday, November 20, 2007
Sunday, October 21, 2007
Busy!
Well, I've gotten very busy at work right now, and while I still manage to get some simple tactical work in everyday, the 16 hour work days (7 a week)are precluding any long games until after Nov 10th. The good news is that my next OTB tournament starts a few days after that (at least it's not before). So I should have time to get myself back in the right mind set before I have to actually play games that count for rating points.
Unfortunately, that means not much in depth-posting until then either, but have no fear, after the 10th of November work gets easy and all shall be amended.
Good luck in your journeys all!
Unfortunately, that means not much in depth-posting until then either, but have no fear, after the 10th of November work gets easy and all shall be amended.
Good luck in your journeys all!
Friday, October 12, 2007
How we break habits
This post was sparked by work I do in my real life job as an instructor and also by some comments from BDK. One of the hardest things to do as an adult learner (in any discipline) is break old habits. When we first start playing chess or driving a car we develop habits based upon the most effective way to accomplish a task that we are taught or can discover on our own. Unfortunately this is almost never the ideal (or even approaching) way to accomplish the task, but it works well enough and so we do it. In order to improve we must learn to break old habits in order build new processes of doing things.
One technique that can be used to deal with habitual action, thought, and change is the Alexander Technique. You can wiki it for a more in-depth overview but it was created by an Australian actor to deal with habitual errors in vocal production. It is a process of self-analysis and methodical change. Here are the 4 basic phases of learning in the Alexander Technique.
Therefore we see the biggest deciding factors in whether we can improve are:
I think this is especially relevant to chess in the thinking process. It is one of the areas of chess that is a habitualized process. Of course the goal is to eventually not have to 'think' our way through the process at all (step 4), but Alexander shows us that we can't get to that point without following the steps between. In my journey I feel that I have achieved step 2 (I'm aware of the mistakes I making) and sometimes I can do Step 3 (Mentally fixing my process) well, however the lure of my past process is still very strong, and much more work has to be done in order to even be able to accomplish step 3 consistently.
One technique that can be used to deal with habitual action, thought, and change is the Alexander Technique. You can wiki it for a more in-depth overview but it was created by an Australian actor to deal with habitual errors in vocal production. It is a process of self-analysis and methodical change. Here are the 4 basic phases of learning in the Alexander Technique.
- Unconscious Incorrectness/Inefficiency - This is where we all are on almost all skills we practice that we aren't trying to improve. Breathing is a common example, as most adult humans breath with alarming inefficiency. The idea is that the execution of the skill or thought is habitual to us, and therefore we don't expend conscious thought on it. However, it also denotes that the skill in question is not ideal and therefore contains at least some element of error or inefficiency. Driving, writing, walking, standing, breathing, and talking are all skills that for most of us fall under this category.
- Conscious Incorrectness/Inefficiency - In this stage the only change is that we are aware that we are not using an ideal process to perform these skills, that is, when we make errors or inefficient actions we notice. Once errors in our chess play are brought to our attention and we start to see ourselves repeating those errors we have reached this stage in the phases of learning. At this point we know we need to change but are as of yet not accomplishing the change. When you notice that your breathing is somewhat shallow and easily interrupted in mid-cycle, you've reached this stage.
- Conscious Correctness/Efficiency - At this point, with the mind utilizing it's conscious power, we can guide ourselves through a more efficient process or action, but it takes conscious mental action to accomplish the new steps and not revert to old habits. This by far the most difficult step in the process. It is very taxing, requires incredible repetition and the lure of falling back to old ways is always strong. Chess players at this stage can make the adjustment to play in the new manner, but doing so causes them to play very slowly and with great mental exhaustion. The key here is the application of mental energy to force the new process to be the one that is used. This leads to...
- Unconscious Efficiency/Correctness - Over time, and with much practice and repetition, the mind begins to integrate the new process into the body's mechanisms, causing it to become the ingrained pattern, and can begin to be used automatically without conscious effort. This is the goal of all habitual training, for the eventual release of the conscious mind. The amount of time that it takes to reach this stage cannot be understated, and is often measured in months on simple tasks that can be repeated often daily, and years on more complex tasks.
Therefore we see the biggest deciding factors in whether we can improve are:
- Realization of Weaknesses
- Application of New Processes
- Tireless, Accurate Work
I think this is especially relevant to chess in the thinking process. It is one of the areas of chess that is a habitualized process. Of course the goal is to eventually not have to 'think' our way through the process at all (step 4), but Alexander shows us that we can't get to that point without following the steps between. In my journey I feel that I have achieved step 2 (I'm aware of the mistakes I making) and sometimes I can do Step 3 (Mentally fixing my process) well, however the lure of my past process is still very strong, and much more work has to be done in order to even be able to accomplish step 3 consistently.
Saturday, October 6, 2007
CTS update
Never forgetting the "Big 5" mantra of balance, it's time for a look at a different element of "Big 5" strength, tactics. For well over a month now I've been doing 30 or perhaps a few more problems a day on CTS (Chess Tactics Server). When I first started I bottomed out somewhere around 1300 and now, after today's session I'm at 1416 having hit a momentary high about two weeks ago of 1447, and an end of session high of 1437 last week.
While the rating is an interesting indicator, I think more important is whether or not the CTS training seems to be accomplishing the goal of tactical pattern recognition. I have to say that so far I think that it is. The problems you are handed on CTS are always within a certain range of your rating, about 150 points above down to 200 or so points below. There could be whole posts ( and I'm sure there have been on other blogs) about the validity of the ratings (around 200 points below USCF from what I can tell, getting farther off as it increases up to over 400 points low around 2000), rating level of various problems, and the usefulness of the time-based ratings system....but.... the good thing is that you are handed a set of problems that utilize lots of basic tactical motifs, and that is what pattern recognition is all about.
I find that I've much improved on two motifs in particular: Long range queen double attacks, and piece trapping. When I first started I would often pick another move only to find the right answer was a long queen move that double attacked a piece and checked the king. The real problem was I never even saw the move before I looked at the solution. Now, I'm seeing them almost every time, and it even cropped up in an endgame I was playing last night, and I saw it there too....this has already been a tangible help.
Sometimes even when I fail the problem, I count it as a success in my mind....why?...because I may have seen 3 of the 4 parts of the position that I need to know and acted on that, where as before I might have stared at it blankly...now I just notice what type of motif escaped me on that problem, why I think it did, catalog it right quick, and move on, as opposed to having to work out the answer.
Again it's about instant (or near) recognition, not analysis. I just have to remember that during a game and not play the first thing that hops into my head, but actually do the checking process....but at least now some things ARE popping into my head.
While the rating is an interesting indicator, I think more important is whether or not the CTS training seems to be accomplishing the goal of tactical pattern recognition. I have to say that so far I think that it is. The problems you are handed on CTS are always within a certain range of your rating, about 150 points above down to 200 or so points below. There could be whole posts ( and I'm sure there have been on other blogs) about the validity of the ratings (around 200 points below USCF from what I can tell, getting farther off as it increases up to over 400 points low around 2000), rating level of various problems, and the usefulness of the time-based ratings system....but.... the good thing is that you are handed a set of problems that utilize lots of basic tactical motifs, and that is what pattern recognition is all about.
I find that I've much improved on two motifs in particular: Long range queen double attacks, and piece trapping. When I first started I would often pick another move only to find the right answer was a long queen move that double attacked a piece and checked the king. The real problem was I never even saw the move before I looked at the solution. Now, I'm seeing them almost every time, and it even cropped up in an endgame I was playing last night, and I saw it there too....this has already been a tangible help.
Sometimes even when I fail the problem, I count it as a success in my mind....why?...because I may have seen 3 of the 4 parts of the position that I need to know and acted on that, where as before I might have stared at it blankly...now I just notice what type of motif escaped me on that problem, why I think it did, catalog it right quick, and move on, as opposed to having to work out the answer.
Again it's about instant (or near) recognition, not analysis. I just have to remember that during a game and not play the first thing that hops into my head, but actually do the checking process....but at least now some things ARE popping into my head.
Sunday, September 30, 2007
Easy thoughts, and not so easy...
So, I'm about a month into rebuilding my thought process, time for an update. As a matter of reference my goal was to have a mostly corrected thought process in 5 months, so I still have lots of time, and that's a good thing.
Here's a quick look at each step in my new process and how easy it is, and the success or trouble I'm having with it:
Check his move: I check what his move does, can't do, leaves, takes, threatens, etc. This is an easy part of the step to do, early in the game for me...but it seems as the games go on I get too excited/tired/etc. and begin to omit this step. It really isn't hard to look to see what his move does, but I get caught up in the flow of the game it seems. But, I'm happy that I can seem to do it for at least 20-30 moves, now I just need to do it on every move.
Generate Candidates: Make a mental list of all my candidate moves that are apparent in the first pass. I find the 'act' of doing this rather easy, it's rare when I don't actually consider the best move (I may reject it in the next step, but that's an different problem, not part of this step). However, it's the discipline of doing this part and not jumping straight into analysis that I have the most trouble with. Again, I get caught in the flow of the struggle and revert to my old, whatever I see first process.
Analysis of Candidates (Real Chess): Create variation trees and evaluation of said trees. This is the bug-bear. Dan Heisman's big challenge is that of 'Real Chess' and this is that step. I have a major deficiency in being able to see my opponents replies to my candidates, and to be honest I think it is simply a difficulty in seeing from the 'wrong' side of the board. It is almost a sub-class of 'board vision' failure...I simply can't 'see' some of the moves. I've decided to work on this aspect alone for awhile before I re-integrate it into my full thought process. I'm going to try playing games from the wrong side of the board...If I'm white, I'm going to play from the black side of the board (you can do this in Chessmaster, or over a set of course). I think this might help me, since I will have to see all my moves from the 'wrong' side of the board, and make my opponents replies easier to find on the 'right' side of the board (which should train both the process of doing it as well as the ability to see the moves at the same time).
This doesn't even take into account evaluation issues, but I understand success in that area is a function of experience, so it will be improving as I go. It just isn't my primary focus right now as, although I may not always evaluate perfectly, I do evaluate with the right process.
Sanity Check: A last check before making the "Final Move" to see it from the surface one last time, fully present in the reality of the board. While this is a very easy step to 'do', I find I often leave it out, on the order of 40 or more percent of the time. I get very confident about my move and play it without the Sanity Check. Again, it's a discipline issue, and breaking my old habit. I need to build a physical trigger I think, to help me remember...like closing my eyes, or standing up, something to ritualize the process... like dribbling twice before a free throw in basketball.
So, while I feel I have made some progress in the last month, there is a long way to go. But I feel good about this self-analysis process, and I think I'm making changes based on my weaknesses as I find them. I continue tactics training every day, but thought process is my true Achilles heel right now.
What have you all found hard or easy about relearning your thought process or playing "Real Chess"?
Here's a quick look at each step in my new process and how easy it is, and the success or trouble I'm having with it:
Check his move: I check what his move does, can't do, leaves, takes, threatens, etc. This is an easy part of the step to do, early in the game for me...but it seems as the games go on I get too excited/tired/etc. and begin to omit this step. It really isn't hard to look to see what his move does, but I get caught up in the flow of the game it seems. But, I'm happy that I can seem to do it for at least 20-30 moves, now I just need to do it on every move.
Generate Candidates: Make a mental list of all my candidate moves that are apparent in the first pass. I find the 'act' of doing this rather easy, it's rare when I don't actually consider the best move (I may reject it in the next step, but that's an different problem, not part of this step). However, it's the discipline of doing this part and not jumping straight into analysis that I have the most trouble with. Again, I get caught in the flow of the struggle and revert to my old, whatever I see first process.
Analysis of Candidates (Real Chess): Create variation trees and evaluation of said trees. This is the bug-bear. Dan Heisman's big challenge is that of 'Real Chess' and this is that step. I have a major deficiency in being able to see my opponents replies to my candidates, and to be honest I think it is simply a difficulty in seeing from the 'wrong' side of the board. It is almost a sub-class of 'board vision' failure...I simply can't 'see' some of the moves. I've decided to work on this aspect alone for awhile before I re-integrate it into my full thought process. I'm going to try playing games from the wrong side of the board...If I'm white, I'm going to play from the black side of the board (you can do this in Chessmaster, or over a set of course). I think this might help me, since I will have to see all my moves from the 'wrong' side of the board, and make my opponents replies easier to find on the 'right' side of the board (which should train both the process of doing it as well as the ability to see the moves at the same time).
This doesn't even take into account evaluation issues, but I understand success in that area is a function of experience, so it will be improving as I go. It just isn't my primary focus right now as, although I may not always evaluate perfectly, I do evaluate with the right process.
Sanity Check: A last check before making the "Final Move" to see it from the surface one last time, fully present in the reality of the board. While this is a very easy step to 'do', I find I often leave it out, on the order of 40 or more percent of the time. I get very confident about my move and play it without the Sanity Check. Again, it's a discipline issue, and breaking my old habit. I need to build a physical trigger I think, to help me remember...like closing my eyes, or standing up, something to ritualize the process... like dribbling twice before a free throw in basketball.
So, while I feel I have made some progress in the last month, there is a long way to go. But I feel good about this self-analysis process, and I think I'm making changes based on my weaknesses as I find them. I continue tactics training every day, but thought process is my true Achilles heel right now.
What have you all found hard or easy about relearning your thought process or playing "Real Chess"?
Thursday, September 27, 2007
My Rubix's Cube and How it relates to Chess
I was solving my Rubix's cube earlier today and something struck me.
First off, I picked up the Rubix's cube only about 3 months ago. It took me five days to solve it the first time, and now I can solve it in just over 2 minutes. Really that isn't very impressive, as top speed-cubers can solve the thing in about 15 seconds consistently. The important question is what is different then that can allow you to go from days to solve to minutes to solve in a short period of time.
Pattern Recognition.
I don't really 'think' about solving the cube now, I see patterns and my hands spin. As fast as I can see the pattern I can solve that face or edge and move to the next one. When I first started I had to find each block and place it, then find the next one, and think about which way I had to turn it, etc. Now the patterns are so ingrained I don't consciously expend thought on it.
It reminds me of something Josh Waitzkin said in his new book "The Art of Learning" about slowing down time.... to paraphrase, as your skill increases time seems to slow down for you....it isn't actually that time slows down, it's that you have less to look at because so much is automatic, therefore you see more details, details that other people miss in the blur of so much information. Now a cube is much simpler than the game of chess, but I think it's the same process. Moving the information from our active, conscious minds, to our subconscious where it can be done almost automatically, as a response to a pattern.
It's why they say GM's are so strong, not because they can calculate faster or more accurately (although they can, it isn't a substantial amount vs an Expert according to research), what they can do is play 10's of thousands of positions by heart, automatically. If there are five important ideas in a certain position, they probably already see all five, whereas I'm lucky to see one during my first pass.
Of course this all goes back to tactics mostly, and it's why we do pattern recognition work. So, what are we waiting for, lets load those patterns in and get automatic! Dr. Rubix's would be proud!
First off, I picked up the Rubix's cube only about 3 months ago. It took me five days to solve it the first time, and now I can solve it in just over 2 minutes. Really that isn't very impressive, as top speed-cubers can solve the thing in about 15 seconds consistently. The important question is what is different then that can allow you to go from days to solve to minutes to solve in a short period of time.
Pattern Recognition.
I don't really 'think' about solving the cube now, I see patterns and my hands spin. As fast as I can see the pattern I can solve that face or edge and move to the next one. When I first started I had to find each block and place it, then find the next one, and think about which way I had to turn it, etc. Now the patterns are so ingrained I don't consciously expend thought on it.
It reminds me of something Josh Waitzkin said in his new book "The Art of Learning" about slowing down time.... to paraphrase, as your skill increases time seems to slow down for you....it isn't actually that time slows down, it's that you have less to look at because so much is automatic, therefore you see more details, details that other people miss in the blur of so much information. Now a cube is much simpler than the game of chess, but I think it's the same process. Moving the information from our active, conscious minds, to our subconscious where it can be done almost automatically, as a response to a pattern.
It's why they say GM's are so strong, not because they can calculate faster or more accurately (although they can, it isn't a substantial amount vs an Expert according to research), what they can do is play 10's of thousands of positions by heart, automatically. If there are five important ideas in a certain position, they probably already see all five, whereas I'm lucky to see one during my first pass.
Of course this all goes back to tactics mostly, and it's why we do pattern recognition work. So, what are we waiting for, lets load those patterns in and get automatic! Dr. Rubix's would be proud!
Sunday, September 23, 2007
My new thought process... in progress
Last time I showed you my old, pitiful thought process, so this time around you get to see the one I'm in the process of learning. A lot of this is directly from Dan Heisman's "Novice Nook" articles on thought process.
I've created a special score sheet for my long training games against the computer that has a not only the actual game moves but also blanks for his plan, my candidate moves, my final PV (principal variation), sanity check, and time remaining. I use it in my Game 90 training games to check my process, being careful not to fill in the blanks unless I actually do that part of the process. This way I can see what I'm weak on without mentally fooling myself.
While I've made great progress in seeing what my opponents moves do, and in selecting my candidate moves, I'm still weak in seeing my opponents replies to my candidates and in doing my sanity check (I usually get so excited I just play the move after all the analysis, I have to remember to take that extra 3 seconds to check it).
My training plan is as follows:
So far it is helping, I've been able to beat ChessMaster 10k on a 2152 setting ( 2 out of 3 times), my previous high had been about 1850. And last night fought into an equal endgame with CM10 on 2237 before getting beat in a pawn race. So I think Dan Heisman is right on the money with his "Real chess" ideas. On to more training!
- Write down my opponents move and then figure out the following:
- What can his piece do now, that it couldn't before.
- What it can no longer do on it's new square (space left behind)
- What other pieces might have been affected (discovered attacks, open squares, new mobility)
- What is my opponents plan
- Is there an immediate threat
- Find a plan/candidate moves that accomplish threats, defense, or activity. At this point I select as many candidate moves as I can find that do something positive (starting with checks, captures, and threats) and bank them for analysis
- Begin analysis of each candidate move looking for my opponents best replies (starting with checks, captures, and threats). I try to go in order of checks, threats to mate in one, captures, other threats, then active but quiet moves. Next, I see if I have an answer to my opponents best replies and if so continue analyzing to quietness. At some point during the analysis I can either discard the move as bad tactically, or evaluate the position as playable.
- As playable lines emerge I save the best one in my mind as my "final move" and check the next line in the previous process. The next line that comes up playable is compared to my current "final move" to figure out which I prefer, and therefore which is now my "final move".
- After all lines have been analyzed I do a "Sanity Check" to recheck the easy part of the final move again, making sure it isn't just a blunder.
- Play the move, hit the clock, write it on my score sheet.
I've created a special score sheet for my long training games against the computer that has a not only the actual game moves but also blanks for his plan, my candidate moves, my final PV (principal variation), sanity check, and time remaining. I use it in my Game 90 training games to check my process, being careful not to fill in the blanks unless I actually do that part of the process. This way I can see what I'm weak on without mentally fooling myself.
While I've made great progress in seeing what my opponents moves do, and in selecting my candidate moves, I'm still weak in seeing my opponents replies to my candidates and in doing my sanity check (I usually get so excited I just play the move after all the analysis, I have to remember to take that extra 3 seconds to check it).
My training plan is as follows:
- Do it with my training score sheet at Game 90 or 120 until I can do it well on every move that requires it. Then...
- Do it without my training score sheet, just a regular score sheet, until I can do it well on every move that requires it. Then...
- See if I can do the same in an OTB tournament setting.
So far it is helping, I've been able to beat ChessMaster 10k on a 2152 setting ( 2 out of 3 times), my previous high had been about 1850. And last night fought into an equal endgame with CM10 on 2237 before getting beat in a pawn race. So I think Dan Heisman is right on the money with his "Real chess" ideas. On to more training!
Subscribe to:
Posts (Atom)
