An Experiment in Rapid Chess Improvement

Record of my experience in undertaking Michael de la Maza's "Rapid Chess Improvement" program.

Monday, March 28, 2005

Circle Three Status Report

I thought I would update everyone on the third circle. Things are going well, but with a trip to Cirque du Soleil this past Wednesday to see Varekai, and my wife's parents in town for her aunt's 60th birthday this weekend, I've fallen a little behind already. (Can someone find that pause button on the remote control of life please?) Anyway, I've finished 630 problems so far and am 36 problems behind. I tweaked my schedule slightly to make days 6, 7, and 8 go from 70, 70, and 56 problems to a more balanced 66, 65, 65 since they are pretty much all level 40 (so why do 70 one day and 56 the next?). I've been keeping closer tabs on how long it is taking me, and through level 30 I was averaging between 1.5 - 2.5 minutes per problem. And that time includes trips to the bathroom, grabbing a quick snack, etc. so I guess I'm doing okay for now.

Anyway, here's a look at my current numbers:

Circle 1 % Circle 2 % Circle 3%
---------- ---------- ---------
Level 10 96% 99% 100%
Level 20 91% 93% 95%
Level 30 74% 80% 87%
Level 40 69% 74% 87% (in progress)
Level 50 67% 72%
Level 60 55% 64%
Level 70 60% 58%
Level 80 57% --
Level 90 55% --
Doing these problems can be surprisingly frustrating, especially when you miss problems that you got correct on prior passes. That sort of thing just drives me batty. I look at a problem, recognize it, and even remember the theme...but cannot find the answer in the allotted time. Also, I often find that about the middle of each level my scores start to fall off a bit, and then improve on the final third of the level (until I hit the "conjunction of tactical motifs" section of course). So I will miss problem after problem in the middle section, but through some freak of mathmatics I will end up with a higher percentage for the level. Of course there is the occasional pleasure of getting one right that I know I have missed on prior passes, but unfortunately the sweetness of this is usually more than offset by the bitterness of missing other problems. I should probably not get too wound up about it all since the point is to learn and half of the benefit is the process itself, but it is easy to forget that. And hell, my percentages are going up and I'm doubling the number of problems I'm doing per day from pass to pass, so I guess that means I'm improving...

Scheduling
One thing I just noticed was how my schedule differs from MDLM's proposed schedule on the later problems. I originally tweaked the original MDLM schedule since I was getting killed by level 30-40 in the first circle. Now it looks like the schedule I'm using is too light in the middle, which probably explains part of why I couldn't finish the level 70+ problems on schedule in the second circle. It may be too late to try to switch to MDLM's schedule for the third circle, but I may try to do that in hopes of having fewer problems to do per day in level 70+. I'd love to go back and finish off level 70+ from the second circle, but I don't think the laws of time and space will allow that to happen.

Wednesday, March 23, 2005

My New and Improved Plan

As noted in my prior installment, I was unable to finish the level 70+ problems in the second circle on schedule. Instead of taking extra days to complete all 1209 problems in the second circle, I decided to move on to circle three. I was considering only doing problems through level 60 on circles three+, but instead I decided to allocate time to the level 70+ problems, and just do as many problems as possible on the days allocated for those problems, and then move on to the next circle. I think this is a good balance between the need to move forward with the program and the benefits of stretching to solve very difficult problems.

From an implementation standpoint, I think this will work well up until the last two circles or so. After that, I won't have entire days dedicated just to the super-hard problems. I guess I'll figure it out when I get there. Until then, I'll just figure out the average time it takes me to do problems in each circle and allocate the same amount time for "as many as possible" in level 70+. We'll see how this works out.

Status Report
I did 200 problems yesterday on day 1 of circle three. I got 100% for level 10 (woohoo!) and 99% on level 20 so far. Here's my planned schedule:

Circle Three Schedule
Per
Day Date Day Total
--- -------------------------- --- -----
1 Tuesday, March 22, 2005 200 200
2 Wednesday, March 23, 2005 150 350
3 Thursday, March 24, 2005 110 460
4 Friday, March 25, 2005 80 540
5 Saturday, March 26, 2005 70 610
6 Sunday, March 27, 2005 70 680
7 Monday, March 28, 2005 70 750
8 Tuesday, March 29, 2005 56 806
9 Wednesday, March 30, 2005 56 862
10 Thursday, March 31, 2005 56 918
11 Friday, April 01, 2005 53 971
12 Saturday, April 02, 2005 50 1021
13 Sunday, April 03, 2005 50 1071
14 Monday, April 04, 2005 48 1119
15 Tuesday, April 05, 2005 46 1165
16 Wednesday, April 06, 2005 44 1209
Anyway, we'll see how this goes...

Sunday, March 20, 2005

Life Finally Caught Up

For the last three months or so I have been living in this fantasy world where I can spend every evening doing chess problems for 2-4 hours. In order to finish the second circle on schedule I was supposed to do seven days worth of problems Friday-Sunday so I would have a final set on Monday. Well...I didn't make it. I only finished four, two of which I did Friday (I actually left work a couple of hours early to do chess problems). At this point I am probably going to just finish the remaining four days worth of problems in four days and go from there. I was hoping to finish on time, but with guests coming over to the house at night, getting various things done around the house, etc., it just wasn't happening.

One thing that has made it difficult to catch up is that on the level 70 and higher problems, I am not doing them much faster than in the first circle. One set of 25 or so problems takes about four+ hours (versus two hours for twelve or so as in the first circle), so it is difficult to do an entire set in a day, much less more than one to catch up. At this point I am seriously thinking about switching to Don's schedule and only doing through level 60 on future passes. As a side benefit I think I'll enjoy the process a whole lot more since after level 60 I find that more often than not I have no clue as to what the next move is and have to take a total guess. If anyone has actually done the entire 1209 problems for each circle in something close to MDLM's schedule, I'd love to hear from you. On the last circle you are supposed to finish each problem in 30 seconds, but on some of these level 70+ problems, I don't think you could physically move the pieces on the board fast enough to do them that quickly given the length and breadth of the variations (and sub-variations). I sort of feel like I'd be giving up to back off the full schedule at this point, but ultimately I think I must succumb to the reality of time constraints.

The Stats
Here's where things stand at the moment:

Circle 1 % Circle 2 %
---------- ----------
Level 10 96% 99%
Level 20 91% 93%
Level 30 74% 80%
Level 40 69% 74%
Level 50 67% 72%
Level 60 55% 64%
Level 70 60% 58%
Level 80 57% 70% (in progress)
Level 90 55%
Finshed 1119 of 1209.

As you can see, level 70 was pretty rough and the second circle was worse than the first. In level 70+ there are a few problems where I remember the ideas (or the moves), but for the most part its a lot of guessing. As I mentioned in prior posts, I still think there is some benefit to doing the super-hard problems, but going forward I'd rather keep the 64-32-16-8-4-2-1 schedule rather than worrying about doing level 70+.

Thursday, March 17, 2005

Why I Hate Level 60

My Arch-Nemisis is Back
I have a particular loathing for level 60. In the first circle, I scored a miserable 55% in level 60, which tied for my lowest score with level 90. I'd be curious to hear from others if they had a similar experience.

I think part of the reason for it is that many of the problems have "enticing" initial moves that don't work out without some sort of preparation. In prior levels, if I couldn't see a very promising continuation for the tactics I saw, the first couple of moves were correct so I could play out a move or two and then figure out the remaining wrinkles. But in level 60 this doesn't seem to work as well. Maybe I am not looking carefully enough at all the opposite side's defenses. At any rate, I find that level 60 really takes the wind out of my sails.

For the record I have finished through 1020, with my current level 60 score at 66% and falling. I am supposed to finish this circle on Monday but I'm still four days behind. I hope to catch up this weekend so I can finish on schedule.

An Observation
In an earlier post I noted the increased calculation ability and the mental stamina I had noticed when starting the second circle. Having made it a significant way into the second circle, I have realized that in many of the problems (up until level 60 of course) I am seeing the initial tactics far better than before. In the first circle I would often have to go through a mental checklist of tactical elements to try to figure out what was significant in the position. Now, I look at many problems and almost immediately I notice the major tactical feature of a position, i.e. the enemy queen is very short of squares, undefended pieces in opposition with mine, captures that drag the king out into the open, etc. I'm not sure if this is a result of having memorized parts of the problems (I am fairly good at memorizing things) or if my pattern recognition is improving. I'm not sure it matters much at this point.

Congratulations are in Order
Congratulations to Chris Kilgore who just made class A! I hope to join you some day...

Saturday, March 12, 2005

Introducing the Chess Vision Trainer

Don is the Man
Congratulations to Don, who just finished the seven circles of the MDLM program! You are an inspiration to us all, oh noble knight! I aspire to join you and Chris in the exhaulted ranks of those who completed the program.

Chess Vision Trainer
In 2003 I wrote a little Flash application to help do Chess Vision training drills per MDLM. I never really used it much, but if you are in the middle of doing find-the-fork drills I hope it will prove helpful. Been wondering if you found all the possible forking squares? Well wonder no more, just click the handy Next button in the program and see for yourself.

I threw together a skeleton chess page to go along with it at fussylizard.com, so check out the Chess Vision Trainer. Feel free to drop me a line on the blog or via email and give me your feedback. Bug reports and suggestions for improvement are welcome.

Status Report
Well, it's been a rough week with all my travel and such. I'm still pretty far behind (I just finished through problem 829 when I should be done with 970 by now...5 days behind). At any rate, I'm playing catch up a bit this weekend, so I hope to make up 2-3 days.

I finished level 40 earlier today and began level 50. Here's the stats so far:

Circle 1 % Circle 2 %
---------- ----------
Level 10 96% 99%
Level 20 91% 93%
Level 30 74% 80%
Level 40 69% 74%
Level 50 67% 81% (in progress)
Level 60 55%
Level 70 60%
Level 80 57%
Level 90 55%

Still showing small improvements over last time. Hopefully the trend will continue, but I find it hard to believe that I'll be able to do levels 60-90 in the allotted time and still get a reasonable score. Time will tell, I suppose...

Update on Sunday Night
I finished through 916 tonight, so I'm four days behind (just over 100 problems). I'm slowly catching up. Level 50 is hanging at around 72%.

Sunday, March 06, 2005

Status Report

Given everything that has been going on lately, I'm falling further behind. I just finished through 670 when I should be done with 829 by the end of tonight (ouch). I was able to do some problems on the plane this afternoon, and I think I can do another 30-40 tonight in the comfort of my hotel room. At any rate, I still have a bit of catching up to do.

I wrapped up level 30 a day or two ago, so here's where things stand:

Circle 1 % Circle 2 %
---------- ----------
Level 10 96% 99%
Level 20 91% 93%
Level 30 74% 80%
Level 40 69% 80% (in progress)
Level 50 67%
Level 60 55%
Level 70 60%
Level 80 57%
Level 90 55%
My percentages have been rising, though admittedly I was taking longer than I should have on level 20 which probably hiked up my percentage a bit. I'm doing pretty well at keeping to my planned time for my recent sets (find the first move in 1 minute, 45 seconds and complete the problem in 3 minutes, 15 seconds).

I am remembering fewer of the problems in level 40 than I expected, but I'm still doing much better than in circle 1. One thing I'm getting better at is seeing when there is nothing more to be gained in a position (i.e. I'm attacking the enemy king but there just isn't a mate there) and instead of (fruitlessly) continuing the attack, I find the (correct) "bail out" move whereby I end the attack with a material advantage. Often the bail out move captures an enemy piece that has been hanging for five moves or so. In the past I was more inclined to either try to take the hanging piece too early, or try to keep on attacking when there was nothing more to be gained. This is probably the result of being able to calculate better and seeing more in the position which translates into me trusting my calculations more. I cannot always see all the lines, but I'm getting better at evaluating when a line has promise or is a dead end.

Looks like Don is in the home stretch! Go Don go!!

Update at 11:25 CST: I just finished through 706. I was tired, but still managed 79% for this set of 36 (total time: 2 hours, 37 minutes). I'm catching up slowly but surely...

Tuesday, March 01, 2005

Suffering

As I work through level 30 in circle 2, I realize I did a poor job of scheduling circle 2. After doing levels 50, 60, etc. I forgot that level 30 was still no picnic. In making my schedule I thought I could do 70 per day of level 30, which is simply not the case at this point (without using up more time than I really can make for chess study- about two hours a day). So the first couple of days of level 10 and 20 were cake, but moving into level 30 was like hitting a brick wall. I finally got caught up through the first week of 70 problems per day, but right now I'm over 100 problems behind (I've finished through 562 but I should be finished with 670). My schedule has me doing 36 problems per day through Thursday, then it drops to 29 per day for the week after. But come hell or high water, I will get caught up...

So the moral of the story: don't take it too easy at the start or you will suffer for it later. If I could re-do the schedule, instead of doing 70 problems a day for a week I would have done something like 110-100-90-80-70-60-50-45.

I will be doing a bit of business travel at the end of this week and the start of next week. I hope to get lots of problems done sitting on the plane. FWIW my current solve rate for level 30 is at 83% and falling, but hopefully I can keep it above the 74% I scored in the first circle.