Despite not having a Beatrice to guide me on my way, I have made it through the seven circles of hell. Not Dante’s, de la Maza’s. (Click here for details.) On March 5th I completed my first run through de la Maza’s Seven Circles of Hell program. I noted the variations I would make in the program in the previous blog post on this topic (found by clicking here). The big question is, did it work? I have to let the numbers do the talking. While I don’t have chess.com (blitz rating) data points that I can use to find what kind of linear regression is at work, the charts the site offers can make for a good estimate of what happened. Here are the charts of the progress.
What is most compelling about this graph is that my peak rating moved up from a previous 1668, which I only every briefly touched before tanking, and, for the most part, I remained firmly in the 1500’s (sometimes lower) prior to doing de la Maza’s program. My previous best win was 1712, but, during my peak, which lasted the whole last week of the last circle, I smashed numerous players rated well above 1700, my best being a win against a player who was rated 1819. Given the peculiar nature of this kind during this particular stretch of studying 1,000 chess puzzles in cycles, I have to think it entirely reasonable that much of the increase was due to the program. I especially think this is so, because, as the cycles progress and I did more problems, the rating increases. There were other variables in play. For instance, I read a book on endgames (by Pandolfini) and the early goings of a book on chess strategy (by Grooten), but I had less and less time to study these, as the number of tactics problems increased per day. Despite not studying those books, my rating continued to increase.
One thing to note about the result is that I only expected to gain around 63 points or so, since de la Maza’s claim was “400 points in 400 days.” In 63 days, I went from consistently playing in the 1500’s, struggling to get up to 1600 occasionally, to whipping 1700+ rated players. I even set my search and challenge settings to higher than they were previously. (I used to search for players rated -150 to +150 of my rating, but I moved it to -100 to +200 of my rating, since players below 1650 were posing no challenge.)
There has been a negative side-effect: exhaustion. A couple of days after my program was over, I felt completely exhausted, and I am still feeling it more than a week later. On top of the feeling of mental exhaustion, which hasn’t been helped by other intellectual endeavors (e.g., philosophy conference), I have been experience extreme chess blindness, which is the case when you look at positions and have a difficult time recognizing very, very simple aspects of the position. For example, I have been incapable, in blitz games, of determining whether I am leaving a piece en prise by making a move. However, I have experienced this before. Who knows what the actual cognitive science is behind this phenomenon, but I appear to experience extreme chess blindness during periods in which my brain is…consolidating information (?), learning (?), integrating old knowledge with new knowledge (?). To be completely honest, I don’t know what it is that’s going on, but my extreme bouts of chess blindness have always been followed by extreme leaps in ability. Maybe Aristotle felt that natura non facit saltum, but the nature of my in-brain learning processes certainly do, as evinced by the following graph and it’s combination of precipitous climbs.
(Note that, while I am using chess.com blitz ratings to assess the benefits gleaned by de la Maza’s program, the ratings represent very different information than this USCF rating graph. The online rating is blitz, i.e., 5 minutes per player, as opposed to the classic, 2+hrs per player rating by the USCF. Additionally, my classic rating tends to be 400 points higher than my online blitz rating, e.g., it was 600 when my USCF strength was 1,000, and 1050 when I was 1450 in the USCF.) Toward the end of my tournament career, I was a C-Class player with at least one A-Class performance, and this followed a major instance of chess blindness; so, while the chess.com graph indicates an immediate and precipitous drop in ability, I think this is entirely explainable by past experiences. If my blitz playing ability returns to 1700+ in the next couple of weeks, then the drop off will be understood as a relic of my learning process. Nonetheless, the fact that, even feeling so out of tune with the game and experiencing horrible chess blindness, consistently playing above 1600 means success.
I am going to do the Seven Circle program again before the World Open, but I plan on doing it in a slightly more fanatical way. The Polgar book wasn’t a great choice for the program, but it certainly did the trick. I am going to use 4,000 problems on this next set of cycles, and I will try to make it through the set as quickly as I can. I may not make it through all 6 or 7 times, but the hope is to get advanced tactical motifs down as much as possible. Polgar’s might not have been quite as effective in the tactical sense, since almost all of the problems in it are mates. (I do notice an increased capacity to calculate in slightly longer games, probably because of this.)