Thursday, September 06, 2007

Circle 5.1 Finished!

This may have been the toughest mini-circle of them all. The problems were harder, some way over my head it seemed (especially the counterintuitive K and P endgames!). About a third of the way through, I started to think that I was missing something. I realized I wasn't understanding the solutions very well, that I was likely to forget them by the time I started going through the problems again.

At first I thought this was because I was ignorant of some advanced set of combinatorial motifs, a new set of motifs beyond the basic tactical elements (e.g., fork, pin, etc), but motifs that all the great players know by heart just as an intermediate player knows forks and pins. I asked you all about this, receiving many comments, some helpful and some inexplicably pathological. Ultimately, we converged on the fact that the complicated combinations are built up out of a coordinated attack that combines multiple simpler motifs. This was extremely helpful.

At the same time, I discovered a great article from some psychologists that showed that beginners who self-explain a position are better able to reproduce the solution when they later see a similar problem. This, combined with the frustration of remembering the original position but not the solution, plans, or basic tactics involved, triggered a drastic revision in how I approached learning this set of problems (as described here).

I did about 15 problems in Circle 5.2 tonight. The problems for which I used the new solution technique were either easier to solve, or at the very least I more quickly remembered the correct solution and plans if I got the wrong first move. Because the utility of the new technique is so apparent, I predict my improvement between 5.1 and 5.2 will be significantly larger than that between 4.1 and 4.2.

In my games, I am already seeing some of the benefits of doing these more difficult problems. For one, I am making fewer quiescence errors, as I am getting practice thinking through longer forced lines. This is key. Also, I am starting to think combinatorially, thinking about how I can coordinate my pieces so that when things shift around I will be able to make multiple threats. I am by no means good at this yet, but just starting to think this way, and have begun finding some initially crazy looking moves that, when thought through to quiescience, turn out to be quite powerful.

My goal: finish the Circles within 6 weeks. Then, time for a cuban Cigar sitting on my back patio with my awesome dog (pictured above) taking a break from this insane prison that is the Circles.


# CirclesPercent Correct
Problem Set 11498-99-100-100-100-100-100
100-100-100-99-100-99-98
Problem Set 21590-93-96-99-99-99-99-99
99-99-99-99-99-99-100
Problem Set 3885-93-97-99-99-99-99-100
Problem Set 41373-87-93-96-98-99-99
99-99-99-99-99-100
Problem Set 5169
NOTE: Circles undertaken with CTB.

11 Comments:

Anonymous Anonymous said...

Cool dog. Congrats with another circle.

9/07/2007 02:38:00 AM  
Blogger transformation said...

nice photo.

as russian-armenian self realization philospher--and le enfant terrable--Gurdjieff once aptly said: "To learn how to get along with people, practice on animals first!"

and... hearty congratulations to a good man!

salute!

9/07/2007 03:38:00 AM  
Blogger takchess said...

Dude, 6 weeks !

for some strange reason your tactical prison sentence reminded me of this.

http://www.retroweb.com/prisoner.html

9/07/2007 05:33:00 AM  
Blogger Temposchlucker said...

Good show

9/07/2007 09:19:00 AM  
Blogger Pale Morning Dun - Errant Knight de la Maza said...

Check your email BDK. Scandy pgn present for ya.

9/07/2007 10:40:00 PM  
Blogger Blue Devil Knight said...

Awesome: thanks PMD.

9/07/2007 11:39:00 PM  
Anonymous Anonymous said...

Congrats BDK! Really got my mind spinning with these two posts. I know that as soon as I reach level 4 and 5 in "Combinational Motifs", I will more often than not stare at the solution being offered like a fool. They often stretch far too many moves down the line for me to find the solution on my own. I'll try your ideas when I reach there.

-Chris

9/08/2007 07:04:00 AM  
Blogger takchess said...

Even to us layman, I think we would all accept the premise that learning to understanding something would help in learning any topic even without scientific confirmation. It is funny that us chessplayers feel differently about it as if learning chess was a mystical thing.Then again perhaps learning chess is a mystical thing. 8)

9/08/2007 10:10:00 AM  
Blogger Blue Devil Knight said...

Even to us layman, I think we would all accept the premise that learning to understanding something would help in learning any topic even without scientific confirmation.

I disagree. Many players think you should just 'play a lot' or 'do a lot of problems' and not worry about all this intellectual crap. Just look at the strangely combative, dismissive, and patronizing reactions to my posts where I was seeking understanding of combinations!

This study showed such a pure pragmatism is not the best approach. As does the obvious helpfulness of the self-explanations for me in remembering problems.

9/08/2007 02:40:00 PM  
Blogger crow with no mouth said...

If I'm not mistaken, you are doing this a little differently than de la Maza recommends, right? i.e., repeating each chapter until you get 100% correct? Have you talked about why you decided to do it this way? (Tried to look through the old posts but couldn't find anything.)

Thanks.

9/08/2007 03:41:00 PM  
Blogger Blue Devil Knight said...

Crow: here and here.

9/08/2007 04:30:00 PM  

Post a Comment

<< Home