I just hit 2100 on rapid for the first time. I know that it's both a high rating and not that impressive when you look at the masters. But I am very happy about it, and to kinda celebrate, I want to share what I learned along the way.
I am not sure what I did right, but I know a few things I did wrong. I will share below a few pieces of advice, some stuff I learned, and I wished I knew when I started my chess journey. I hope it can help someone. I read a few of those during the last 4 years, and I know it helped me.
First, a bit of back story:
I created my chess.com account in January 2017 but started to take chess seriously in august 2017, so it took me roughly 4 years. I think I touched 2000 for the first time after 3 years. Before that, I played roughly once a year with my father. I kinda knew the rules, but that's it. And I say "kinda" because I didn't even know "en passant."
When I started playing seriously, I had a few goals in mind. First, learn how to focus, both short-term on a single game and long-term, on a journey towards improving my chess level. I was starting a Ph.D. at the time and set to myself the goal of reaching 2000 on chess.com before the end.
Once I reached this original goal, I was hooked and kept playing and trying to improve because half of the fun of chess is improving for sure.
Without further ado, here is a list of advice and anecdotes on my journey in no particular order.
Play against a better player
One of the best decisions I took was to set my chess.com challenges to maximize the chances of playing against higher-rated players. If you go on chess.com settings, you can actually set the minimum and maximum range of elo you are challenging in rated games. The default setting is +200 -200, that is +-200 elo points around your level. But you can tweak it to be -25 to +infinity.
I would strongly advise anyone interested in improving to do so. On average, you will lose more than 50% of your game by construction (that's my case), but the learning experience will be worth it.
If you have the chance to mean a very strong player in real life, play with him as much as you can. I was lucky enough to work with a national master for a while, and I think that on something like 50 games, I drew him once. That's it. But I am still grateful for the time he spent playing with someone so clearly beneath his chess level.
Don’t care about elo:
That’s one piece of advice I read often and personally don’t believe in for a few reasons:
First, setting yourself elo based goals is extremely motivating and extremely rewarding when you reach them. The trick is not to put a time limit on the goal. i.e., deciding you gonna learn hard until you reach 1600elo is a great mindset, but deciding you have to do it before December first is a way to get stressed and sad when you failed.
Second, elo correlates with everything. Once your elo increase, so does the quality of your games, the beauty of your plans, the precision of your endgame… and, more importantly, the quality of your opponent. I like to see elo as the right to play against more adversaries. When I was 1300, I loved fighting against 1500s, and when I reach 1500, I felt privileged to be allowed to play (and get crushed) from time to time by 1800+ players. And when I reached the level to randomly come across (and get crushed) by fide masters or national master, I was to the moon.
If you view it as a marker of progress and of how likely you are to play against great chess players, thinking about elo is great and can greatly improve your motivation. If you believe it reflects your intelligence in some abstract sense, that’s a) wrong and b) the best way to get depressed and quit the game altogether.
Blitz, rapid or classical.
This is a controversial topic.
During my 4 years on chess.com, I played 11,896 blitz games (mostly 5-0) and 996 rapid games (mostly 10-0, but also some 30-0). I generally feel I learn more when I play 10-0, but it depends on my level.
In the beginning, I only played 30-0. I believe that when you start and can’t spot a tactic to save your life, playing 5-0 is close to useless, and even 10-0 is too short. Longer games make more sense. Once you progress, you can shorten it up.
You will hear many people telling you that the only way to truly learn chess is classical. I don’t believe for a second that it is true. I think I played a total of 7 classical games in my life and am definitely happy with my current skill level.
The key is to analyze your games. Suppose all you do is play one blitz after another without looking at your mistake. In that case, you are probably never gonna improve. But if you analyze each game, the game length becomes a trade-off. The shorter the game, the more you will play. This means you will see different positions and learn from more mistakes. But of course, the shorter the game means more silly mistakes, and the positions you analyze are less worthy of being studied.
In an interview, I once heard Magnus Carlsen that the most important is to play a time format you have fun in. Otherwise, you’ll just quit. For all of us who play for fun and not money, this is 100% true. But I’ll add below a few rules of thumbs:
- When you start the game, play only rapid. Classical makes no sense because you will focus for 4h to put your queen in front of a bishop for no reason on move 39. Blitz makes no sense because you will basically randomly move pieces.
- Don’t play a time format in which you give pieces. Start with longer time, improve your tactics, then you can play some 3-0. Before that, it may be fun, but it won’t help you improve.
- Don’t play a format where you get bored. If every time you play, no matter how hard you try, you end up with 25% more time than your opponent, you are most likely made for shorter games.
Joining a club
I tried to join two different clubs during my time, and both were useless.
Granted, I am not a very social person, and I am sure that my experiences are not representative of all clubs. But the ones I joined were full of ego clashing (over a game which I find a bit sad) and high talk about the nature of the game.
I met a lot of arrogant 1600-1800 who explained why I should learn an entirely new opening repertoire because Magnus would never play my opening… Or some other who explained with the same certainty that I should read books after books on endgame because nothing else matters. Everybody in clubs seems to know how to become a master, but curiously none of them ever pass 2000elo…
So to me, it was a waste of time, which is a shame because moving pieces in real life is so much fun. I am curious to hear if some of you guys have had better experiences.
Hiring a coach
At some point last year, I did 5 to 10 sessions with a coach. An IM contacted me on chess.com (I am sure it happen a lot to everybody) to offer his service. On that day, I was bored at work, so I answered, just change my mind. I was sure that I was gonna say no at the end, but the guy was a fantastic salesman.
He went briefly through my game and spotted a few weaknesses in my openings. He immediately (just in the DMs exchange) proposed a variation that was, indeed, objectively better. So I gave it a try.
The guy charged 28bucks per session. Which in my country is entirely acceptable and most people can afford to spend this once a week. We met over skype. He analyzed one of my games with me, answered some questions, proposed fun exercises, and helped me improve my openings.
All in all, it was great fun. The guy was nice and very efficient. But I stopped it after a few sessions because, after a while, all my initial questions were answered, and he just proposed an admittedly cool position to analyze with him or cool exercise. But nothing that I couldn’t have found on my own online.
Right now, I am thinking about contacting him again but only seeing him once a month or every two months. I think it can be useful to answer in bulk a few questions, which would take me time to google myself.
I would definitively recommend taking a coach once a week to people with motivation problems, otherwise others, just from time to time. But it’s for sure not a must to progress.
Opening part 1, how much to study them.
Another controversial topic. Some people will say it's useless for non-master, others view it as the center of chess, others just love to learn it all because they find it fun. I am somewhere in the middle of all that.
I firmly believe that people should learn a few openings very early for a few reasons. But the main one is the following: yes, it is true that learning opening will not be useful for beginners, probably absolutely useless for players below 1400, and only mildly useful until 2000+. BUT, if you are serious about chess, you know you are going to reach this level at some point, so why not start now? You should definitively not learn 20move deep theory at elo 800, but vaguely the direction and theme of some opening repertoire is probably a good idea. Then you'll be able to play games that have common themes and common patterns for you to learn and understand.
Opening for beginners (below 1200)
When you start, you should probably learn opening principles. You can read My System (Aaron Nimzowitsch) or just find a small video of opening principles. A good one should be able to explain all of it to you in 20' top. As said above, I would encourage you chose the embryo of an opening repertoire and stick to it, which means:
- Always play 1. e4, or 1. d4, don't vary.
- At some point, half watch a video about playing some "easy" opening for white, and one answer for black against e4 and one against d4. Remember the main move and concept (not the line) and try to stick to it.
Opening 1300 to 1600 (roughly)
Learning a bit more about your openings. Say 5 move deep will help you a lot… just because it saves time. Your understanding of chess is not yet good enough for openings to truly change your games, but it will help you to be confident and not waste too much clock time.
Basically, you should be able to play the first 5 moves being sure you didn't blunder.
Opening for 1700 to 2000
Now you need to understand the concept behind each opening. Start learning (still not lines) what the main ideas are in the middle game in each variation.
Depending on your repertory choice, you'll need to learn a few sharp lines quite well, but for the rest, you can just focus on learning the middle game theme. To do so, watch some master videos, and find some random master game on variations you don't understand.
Opening for 2000+
I am in the middle of it right now, so I have no idea if I am correct. But my feeling is that now, it's time to work on it a little better. Right now, I tend to either crush it or be losing out of the opening. I would say this happens 10% of the time, which is too much. It also means that: a) when I know a line well, I can use it to outplay my opponents, and b) when I don't, it's the other way around… So time to actually open the book and learn some tricks!
Opening part 2, building your repertoire.
What to chose? So many openings, all with cool names…
On the repertoire front, I made a few mistakes that I would advise against.
London temptation…
First, I played the London too long. For those who don’t know, the London opening looks great for beginners because it’s easy to learn, and it’s very solid, which means you will hold up to move 40 in most games, even against better players. When you start, it is very tempting, and it tempted me.
I met a guy who is a national master and played mostly the London system with white, so it’s definitively possible to become great with it. But the system has drawbacks… the main one is that the London tends to create very close positions, which shields you from tactical mistakes, and actually knowing how to defend in well. I realized at some point that I lost systematically when the game opens up. I definitively didn’t know how to attack.
The final proof of it (for me) was one of the exercises on chess.com, in which they have you played as white with a fully developed position against black’s original position. I highly recommend this exercise. The computer evaluation gives white at +5, even though material is equal. When I played the London exclusively, I couldn’t beat the computer on this exercise. Ever. Now that I switched to e4, I need to focus, but I can do it.
Again, I believe you can be good with the London, but it wasn’t my case. I was stuck at 1900 for a long time and only moved to 2000 when I learned how to attack.
Switching too much
My other big mistake was to try too many openings, especially with black. I started with the caro-kann, got scared out of it by some a**** at the club, move to french, then tried briefly e5, and finally back on caro-kann, which I love now… Against 1.d4, I started with the semi-slav, played the dutch for a while, and now I am on the KID. Every time I switch, I dropped elo… and I don’t think I learned enough to compensate. My advice, stick with your first choice.
Even with white, I switched too much. After London, I tried a d4 system, gave up, and went back to London. Switch to e4, gave up again. Try a hybrid (play e4 sometime, London the rest of the time), gave up, and finally, I made the switch to e4. I have much more fun with it.
And whenever I tried to switch my main white system, I dropped 200elo points… this is scary but was definitively worth it for me in the long term. But again, don’t do it like me… Start with something and stick with it as much as possible.
First move, e4 or d4?
1 e4 is more aggressive, and d4 is more positional. My only advice here is… do the opposite of what you like. I had a more positional inclination, and switching to e4 was great for me because it forced me to work on my weaknesses. In both e4 and d4 systems, you will have dynamic and close positions from time to time, so you can’t afford to be only good at one of the two styles, hence my somewhat counter-intuitive advice.
Main lines or secondary lines?
I would say I prefer secondary lines, but both make sense.
If you choose a very popular system, it will be easier to find resources (which is important), but it means people, on average, will be more prepared. To me, the pleasure of finding obscure variation and surprising opponents with it definitively outweigh the pleasure of knowing that my variation could be used (by someone else) to beat a world champion. But to each his own!
Also, one piece of advice is to ignore considerations like: “is it playable at top level.” Maybe yes, maybe not, but who cares? As long as a gm can play this in a game, it’s good enough for me. For example, I play the king gambit (which super-gm rarely if ever use in classical), and I have lots of fun/success with it.
One system to avoid
I would strongly advise against learning the french… It’s fun and elegant, but… you need to learn just as much as if you chose the Sicilian, and the opponent can always just play the exchange to make your life boring as hell…
How to learn openings
I planned to expand a bit on learning strategies, but I realized I wrote already waaaaay too much, so if somebody is interested in advice on this, just ask.
Endgame
Learning endgame is definitively important. But keep a few things in mind:
-
From the point of view of a computer, everybody sucks at endgame, even GM, so don't get too sad when you look at analysis.
-
Some patterns are absolutely essential. I would highly recommend Anna Rudolf's chess.com video on endgames. However, the advanced fancy ones are useless. Keep in mind that they are developed and analyzed by super-gm, for super-gm, and by bored geniuses who like to have fun with theoretical problems. I learned what the Philidor position is and trained it. In my 10,000+ games on chess.com, I think it came out once…
For the beginner, what you need is just the main mate patterns (queen king, rook king). For the intermediary, you need king pawn tricks (the square, the airplane, the opposition), and after that, a few more tricks when you reach 1800, I would say. Endgame tends to be my strong suit, and I didn't learn much stuff by heart. Maybe it helps other people, but I don't think it's a must.
Tactics
I will not repeat here how crucial it is. Until you reach 1500, I would say it’s the only thing that matters. Don’t get me wrong, you should always study all aspects of chess because you will need them at some point. But below 1500 elo, games are determined by tactics 99% of the time.
Here is just a few pieces of advice and a weird pattern I notice.
First, whenever I do a lot of tactics, I lose elo in games. Especially when I improve on a tactic trainer. I am not sure why. I think it’s because the brain switch, and you start spending wait to much time calculating tactics when you should simply play a quick move. If you have the same thing, don’t panic. You are not alone, and, long term, the temporary elo loss is worth it.
Second, puzzle rush is fun but useless. It may be good to become a bullet player, but not for rapid or 5-0.
Third, if you want a true challenge, go do tactic on chesstempo…
Forth, keep tactic fun. I see many people explaining how to improve by never trying a move until you are 100% it’s the correct answer. I think it’s probably true, iff you can do it. But most likely, such a hard-core approach to tactic will get you done and make you lose motivation. So however you like to do puzzles, just do it that way, the most important is to do a lot of them. As an order of magnitude, I recently hit 2700 on chess.com tactic and spent 239h on this tactical trainer alone (not counting other sites I sometimes use).
Analyzing games
I read a lot about the importance of analyzing by hand, going over the game without the computer, and then put the computer on. I am sure it can be great, but I have never done it. I tried once and got bored instantly.
What I do (and recommend) is used the excellent chess.com analysis tool on all my games. At a minimum, all my losses. Just briefly go over it, check the mistake, and especially the opening mistake to correct them next time.
Full disclosure, I recently downloaded all my chess.com games (with some python script if anyone wonders how) and put them in ChessBase, which allowed me to go over all the games I played with a high enough elo and tried to analyzed where I was bad out of the opening. I ended up analyzing some of them by hand and writing down some notes. But I reach 2000 without ever doing this.
Helping a friend with lower elo
This is a very random anecdote, but I spent some time helping a friend learn chess. He started around 800 elo and is now quickly reaching 1500.
I think that if you have the chance to help someone with a lower elo, you should. Not talking about the “being nice” aspect, I believe it can help you improve. You will be forced to put words on the concept, formalize your learning strategies and your chess understanding.
Play often, and have fun
My final advice is simple and summed up by the section’s title. Just have fun, and play often. Do a bit of everything, tactics, games, analysis.
