Month: June 2022

I went on vacation recently, which disrupted my chess playing and chess study routines. But it also gave me a chance to reflect on my chess journey to date. I’ve been studying chess seriously for about six or seven months, taking group classes and private lessons. I analyze all the games I play with a chess engine and I also study tactics. But is my self-developed training program rigorous enough? Am I improving? Because chess has rating systems by which players can measure their progress, I thought it would be easy to tell how I was doing. But getting a rating, especially your first one, can be difficult. In the Before Times, people normally got a rating by participating in official in-person chess tournaments. These opportunities are beginning to come back, but during Covid most players have had to rely on lichess and chess.com for their ratings. For reasons too complicated to go into, my rating on lichess is still classified as “provisional,” and I don’t trust my chess.com rating because I have only played 23 games there.  

Because of these uncertainties about my chess progress, I decided to add another element to the mix. It is the new training program offered by Chess DoJo, “a self-described hub for chess players, improvers, and coaches” founded by three distinguished chess sensei (their word, but I like it). I have just begun to explore the training program, but its content and organization seem very promising. The program is customized by rating band, starting with 0-400, 400-600, 600-800, 800-1000, and then going up 100 rating points at a time all the way to 2400+. You get access to the contents of all of the training programs when you subscribe, although you have to choose one in which to be an active participant at any given time (that group becomes your cohort). The videos they are posting on YouTube describing each rating band are very informative; I watched How to Make 400 Using the Chess Dojo Training Program and was hooked. 

After I subscribed and chose the 0-400 band, I was given access to the “Training Program Scorecard” spreadsheet, which lists the various chess activities that the sensei have determined will help you to move up to the next rating category. You check off the activities as you complete them, providing some accountability. The required activities include the kinds of things I had been doing (and more), but they are more specific and quantified. For example, in the case of tactics training, the spreadsheet says, do the first 50 exercises on tactic X in book Y, which is a refreshing change from my floundering from book to book trying to determine which exercises were appropriate for me. 

The real meat of the program is, as it should be, playing games. When it comes to playing, Chess Dojo is also serious about time controls. For the 0-400 level, they insist on thirty minutes per side, which is quite a lot in today’s online chess world, where time controls of five minutes or fewer per side are the norm. They require longer time controls as your rating increases, which makes sense, because the more you know, the more possible positions you will be able to visualize, and you need time to do that. Another helpful function of the scorecard is that members of the cohort list their time zones and when they are available to play. Thus you can find opponents around your skill level who are seeking to play games with the same time controls.

But you don’t just play the games, you play and then annotate them. Annotate? I had heard of analyzing your own games using a chess engine, but not of annotating them on your own. In chess study, we use collections of annotated games, but these are games played by masters. I have begun learning to annotate my own games, and this practice is a revelation. I want to give it its due, and this is getting long, so I will discuss it another time.

Recommendation of the Week: The Candidates Tournament is under way, with eight players competing to determine who will be the challenger for the World Chess Championship 2023. There are many reports and recaps available, but my favorites are those by Matt Jensen of Chess Goals. His explanations are concise yet thoughtful in terms of explaining obscure chess terms (today it was the isolated queen pawn, or “IQP”). In addition to recapping that day’s play, Matt uses his background as a statistician to set odds for each competitor, so you always know how things stand.    

 

I’ve written about playing chess with a regular partner and about playing with strangers, but there is another option: playing against the computer. The two major platforms, lichess and chess.com, offer this choice. Here I describe my experiences playing against both and compare their features. But please don’t misconstrue this comparison as a recommendation to use one in preference to the other. Chess.com is a commercial product, while lichess is free, open source, and nonprofit. Chess.com does have a no-cost option, which one could, perhaps, compare with lichess more fairly, but the free version of chess.com strictly limits the number of times in a day that you can do most activities (game analysis, puzzles, lessons), and so I don’t think it makes sense as a tool for serious learners.

Not surprisingly, then, chess.com’s play-against-the-computer feature has many more bells and whistles than does that of lichess. These are apparent from the outset, when you choose your opponent. Chess.com has dozens of bots, complete with names, pictures, nationalities, chess playing styles, and ratings. At the top of this post you see Zara, a Malaysian female rated 850 who “loves to play creative chess games and come up with her own ideas. Be careful or she’ll trick you with her unique tactics.” If you scroll down past the many bot pictures, you reach the more austere option of playing against the “engine.” The chess.com engine’s levels number from 1 to 25 and come with labels, from beginner (rated 250) to maximum (rated 3200). On lichess you can play against a bot named Maia or against its engine, which has 8 levels.

In addition to level of play, both platforms offer other options for customizing your game, including color (white, black, random) and time control. Chess.com offers many more options than these, however, and this is where playing against the computer can become an outstanding learning experience. You can select various types of assistance, including takebacks, warnings, and hints (to name just a few), or you can play in ”Challenge” mode, with no intervention. Options like takebacks and warnings were very useful to me at the beginning of my chess journey, when I was blundering and making mistakes that would have ended a game in short order had I not chosen to be warned and able to take back a move.   

Chess players debate endlessly about what percentages of time they should give to the various aspects of their study of the game, such as playing, studying master games, doing puzzles, and so on. But all agree that an indispensable part of learning is analyzing each game you play, preferably right after you finish it. Both platforms incorporate this analysis ability into their play-against-the-computer option. The main difference in their functionalities is in presentation, with chess.com’s analysis being more immediately understandable and accessible. It breaks the game down into “key moments” and lets you choose to “retry” to find the best move, get a hint about it, or just display it. Lichess has a similar feature called “Learn from your mistakes,” but its presentation is not quite as user-friendly. In addition to this brief analysis of key points in the game, both platforms offer a much deeper, move-by-move analysis as well. 

There is one huge difference between playing against the computer on chess.com and on lichess: even premium versions of chess.com do not save a record of the games you play against the computer, as they do in the case of your games against human opponents. Lichess saves both types of games, and you can go back as often as you like to analyze them and slice and dice them in a multitude of ways. But if you play against chess.com and exit the game without using the analysis feature then and there, it is gone for good, and you have learned nothing. But you can do a workaround in lichess to address chess.com’s failure to save the games played against its computer. Before exiting your chess.com game, you have a “Share” option, which you can use to make a copy of the game in the chess computer notation format called “pgn.” Lichess lets you import the game in that format (this takes a little tweaking), save it, and do all of the same analysis activities that you can do on games played originally in lichess. 

But how does it feel to play chess against a computer rather than a human being? I find that it is less anxiety-producing than playing with a random stranger. Even though I know that I will likely never encounter that random stranger again—and can even block them if need be—the fact that they are a real person adds an edge to the experience. This means that when I sit down to play without a partner and have to decide whether to play against the computer or against a random stranger, playing with the computer feels like the easier, but more cowardly, option. Is this how it feels to others, I wonder?

Recommendation of the Week: On his Perpetual Chess podcast, Ben Johnson interviewed Stuart Margulies, one of the coauthors of the classic book Bobby Fischer Teaches Chess. Don’t miss this unique conversation, which provides a fascinating glimpse into the chess world as this bestselling volume was being created.