How I Finally Hit 2000 on Lichess and Improved My Rating
trapezemobile.com
trapezemobile.com
[0] https://github.com/ornicar/lila [1] https://lichess.org/features
The lead developer of Lichess, Thibault Duplessis, gave an excellent talk the other year if you're interested in the story behind Lichess. [0]
The frontend virtual dom was initially Mithril and has later been replaced with snabbdom [1]. No React or Vue.js.
Combined with a lightweight-ish frontend (Cordova build for native mobile) and, voila, impressive performance with a decent UI all while not having to maintain separate web, iOS, and Android apps.
Although maybe in the early days it was, for some reason, written in PHP; I'd be quite surprised if that were the case.
I’m also building some software integrated with lichess to help people get good at chess, fast. Any chess hackers interested in contributing, please email me! Contact info is in my profile.
I just got hooked on lichess after not playing Chess since I was a kid, it's a lot of fun, I also spend a fair bit of time in analysis...
Lichess matches people by rating, so on average, if I blunder my way into a large deficit, my opponent may just as well blunder me out of it.
Not a chance. People at lower ratings screw up what should be wins all the time. Your opponent may be hoping for a stalemate or to run out the clock or waiting for you to blunder.
It's not quite as polished as Lichess and if you've never played Tetris with a keyboard before that may be a significant learning hurdle (though it's actually a very good control scheme for the game). But overall it's a really cool site. (There's also Tetris Friends, but I just always found the general style of the site very offputting.)
Wait, whaat?
Are there pointy-clicky interfaces to tetris nowadays? Well, of course there must be.. I think I last played tetris as the original version made by that Russian guy (or was it so long ago that it was a Soviet guy?) running on DOS.
1. I can't even scroll the page horizontally on Chrome.
2. There are 2 menus with a completely different interface.
3. The Android mobile app gives you a completely different experience, and I think there is no lobby, so I often pick a match from Chrome and then switch to the app (and of course the "Open in Mobile App" button doesn't work for me).
4. In general I find the UI is very confusing:
4a. There are 3 places where to write something about a match: your personal notes (never used this, I guess the value proposition here is that your notes will be stored along the match, but I don't think almost anybody cares about notes about games played last year for instance), the chat (which works only between the current 2 players, and you can't access it from the analysis page) and the spectators chat (I suppose this has a purpose for highly spectated matches and live commentary, but I could also run a separate browser where I'm not logged in and read that too so it sounds useless to me).
4b. The match history at the bottom of the chessboard is useless to me 99% of the times, as it's completely empty.
4c. Often the same information is displayed in different parts of the page, I already mentioned the menu problem, but also when playing you can see some infos about the players both on the left and the right column. All these things complicate the UI.
To close on a positive note the feature I like the most of lichess is probably the analysis page, super useful and well done.
Certainly there are ways that it could be improved, but calling it "terrible" seems like a huge exaggeration to me and does practically warrant an illustrative alternative that would better suit the OP.
I've also never been of the belief that mobile and should have exactly the same interface specifically given the different restrictions of the mediums.
I'm not sure what that's supposed to mean. If you asked designers about the state of mobile UIs about 12 years ago I think many would have praised them, then it came the iPhone and they vanished completely.
I've only mentioned very clear examples in my post, would you, as a designer, put 2 completely different menus (with basically the same functionalities), whose buttons are visible at the same time, on an app? Would you want to have the app scrollable if needed or not?
> I've also never been of the belief that mobile and should have exactly the same interface specifically given the different restrictions of the mediums.
Of course, currently lichess.org is doing the opposite by the way.
My complain was about the mobile app being a _completely_ different experience. i.e. there isn't the lobby, the first thing you see after you go to lichess' homepage.
- I've got the core UI framework mostly ready, unfortunately still undocumented but most components should be ready for production [1]
- Some core structures, like a Hierarchical State Machine implementation I made, should be good enough [2]
- I've implemented some games like tic-tac-toe and chess, but focusing mostly on the logic of the game rather than the UI or anything else really for now. I have a usable chessboard component but it's not really any good for production use.
- Database-wise pretty much everything still needs to be developed.
- The server/cloud infrastructure still needs to be developed.
- I still haven't quite figured out what the best way to push updates to the app is, given that you don't want to disrupt current games almost at all but you still want to push bug fixes to everyone.
I once got banned because my cell phone lost connectivity as I passed a subway tunnel and in the meantime the other user “reported” me for abandoning the game.
This resulted in my account being frozen out from playing games for several hours. I wrote to the support email address and got no response.
This then happened two or three more times in the same month.
Furious with the ridiculous policy, I only will play in “anonymous” mode now, usually at chess.com with frequently changed accounts, sometimes still at lichess with no registration.
If your experience of internet chess can possibly be affected by intermittent disconnections or move delays by your opponent, then you are being the drama queen. That would be incredibly unreasonable.
If you agree to play a game against someone, you are implicitly accepting that one thing they may do is become disconnected / idle. That’s simply unavoidable. You’re always free to resign and leave the game if you don’t prefer to wait around, but locking the other player out when there’s no evidence they did something malicious is horrible.
> “If you're disconnects are significantly higher than that, you're probably not the target market for on-line blitz chess.”
It’s not very common and it’s not predictable or uniform (not even in subway tunnels). But if someone reports you for disconnecting, it autonatically locks you out from playing and emailing to the support email gets no response to unlock in false positive cases.
Also, just in the basic rules of chess, your opponent can do whatever they want with their clock time. You’ve agreed to abide by that rule by agreeing to play. If they want to take 4:00 to think about their sixth move in a 5/0 game, they are perfectly allowed, and may even still win if you blunder.
There’s no time limit per move beyond the first move and the overall clock time. Disconnections (when the player might return) are no different.
This is incorrect. At this point I'd like to point out that I was (and soon will be again) a USCF certified TD [1], and have run my own correspondence site for about 20 years [2]. Abandoning a game has the unique distinction of being mentioned three times in the USCF rulebook, and carries with it the most severe penalty specifically mentioned which is ejection from the tournament (with more severe penalties are up to the TD and USCF). FIDE has a similar rule, though they don't mention abandoment directly where a player has stopped trying to win under the "normal means". Not all rule violations have to be intentional, specifically similar to your case is "annoying behavior". When you fail to finish a game, for whatever reason, it's annoying and an inconvenience to your opponent, and some punitive measure is warranted.
You’re obviously not going to get punished if the “abandonment” is revealed to have been because you get accidentally locked out of the tournament hall through no fault of your own, which is the right comparison for something like online chess disconnection.
I don’t see a reason to feel that your comment is applicable.
This is the part that does not seem to be true. The way this works for federation-operated tournaments is not the same thing at all as basic rules of chess, and instead is highly specific to that physical format of chess. Moreover, the physical constraints of an in-person tournament would make its rules about abandonment less relevant for comparison with internet disconnections, not more.
On the point of “annoying behavior” — still, you’re not addressing the fact that disconnections are not the fault of that player and often are not anticipateable or controllable.
For things that happen in federation-operated tournaments that also are not intentional or controlled by the player, it would clearly be mitigating circumstances such that the rules you’re trying to cite would not be relevant anyway.
Separately from this, I personally actually think it’s totally backwards and unreasonable to ever, for any reason, care that an opponent disconnects. I don’t care at all if my opponent disconnects for any reason, whether because they are frustrated, their cell connection dropped, whatever. They have zero obligation to move fast, resign from losing positions, or anything else. It is internet chess with literally nothing riding on it, ever, for anyone.
If you click to play a game of length X minutes or whatever, that’s your commitment. It’s not your opponent’s. They might waste X-1 minutes looking at cat pictures before playing a move— that’s well within their rights and you agreed to play, meaning you can quit or disconnect if you don’t wanna wait around for the next move, or you can cede that your opponent can do whatever they want with their allocated clock time, whether that is resigning, disconnecting, or playing nonsense moves, or playing for real.
I think all the frustration that an opponent can degrade your experience by using their clock time to do whatever they want is unreasonable. That’s on you, not on the system or the opponent.
A few hours ban is a slap on the wrist from a moderation perspective, and it seems... odd... that they’d even hit you with that unless you regularly leave games.
Sounds like they do it pretty regularly :/
2000 lichess is actually more like 1800 chess.com. I would know because that is my playing strength on both and because reddit.com/r/chess has literally dozens of posts of people comparing their lichess and chess.com ratings (my impression is the average difference is ~200). That subreddit also has dozens of posts of people 'reaching 2000' on lichess. It corresponds to about 95th percentile on lichess blitz, just as 1800 corresponds to about 95th percentile on chess.com blitz. So literally anyone switching platforms would 'improve their rating'.
Here's a blog post along the same line from a few years ago: https://www.chess.com/article/view/chesscom-rating-compariso...
Improvement takes really a lot of work, far more than I think the author appreciates. Going from 1800 to 2000 in a year would be an excellent outcome and would require hundreds of hours of work on chess.
The author dismisses the one method that is arguably most recommended by strong players: studying GM games. Chess is about playing good moves, and studying the games of the world champions helps you learn what good moves look like. Studying GM games is hard work with little tangible reward (no dopamine rush from finishing a puzzle or playing a quick game) but it seems highly effective.
Another thing that stuck out to me: the author dismissed tactics puzzles based on what looks like 30 puzzles done on Lichess. That's nothing. I've done 7500 tactics on Chesstempo standard mode alone and it did a lot for my chess. It was basically the only thing I did to go from 1300 to 1600 or 1700 or so (over the board). Tactics aren't a panacea for chess improvement, but neither is anything else.
I have some smaller comments (for example: re: "If I want to get difficult puzzles correct, I don’t rely on any pattern recognition. I just spend a lot of time and mental energy explicitly calculating out every combination of moves. I’m really not learning much, most people can get the same puzzle correct if they just thought as hard as I." -- pattern recognition is the basis for calculation skill so it doesn't make sense to say you don't rely on it when calculating) but the main point is that chess is hard and improving takes a lot of hard work, far more than the author makes it seem.
It seems within reason that one could move from 800 to 1000 in 16 days with very basic study and coaching.
> there is a lot of randomness especially when playing blitz and rapid chess
This is likely true, but this randomness appears to be averaged out by the sheer amount of games he played over the ~6 month period in the chart.
That would be correct if you'd written 200 points.
In blitz (5 minutes or less), time management is absolutely critical. When I went back and reviewed games I lost, it was frequently due to tactical opportunities or poor moves I made while short on time. You simply don't have the luxury of time to find the best move in a 3 or 5 minute game. If you can quickly find a move that improves your position in some way and doesn't hang material or create opportunities for your opponent, make it, and don't think twice, unless you already have a large time advantage over your opponent. My rating improved dramatically once I focused more on managing my time better.
Largely agree with most of the other advice offered in the article though. It cannot be overstated how much playing with strong players and reviewing your past games (wins and losses) will help you improve and develop a sense for tactical opportunities.
It's also worth noting that on lichess (and online chess in general) the fast time control pools are more competitive, so reaching 2000 in Blitz is much more difficult (for most people) than reaching 2000 in Classical.
1: https://lichess.org/blog/Wh9KWiQAAI5JrKVn/introducing-rapid-...
I played in anonymous mode for a long time experimenting and I realized that I was losing most of my games in bullet to time not to tactics.
For bullet: Just play average move each turn based on long term goals (I want to castle king side and pawn storm queen side etc), dont bother about tactics or traps or any other short term strategies.
Concentration != using n-seconds per move... More like not wasting n% of seconds per game with non-game stuff.
We tried to do it in Mithril like Lichess but eventually gave up and wrote our own animation framework for react-redux that uses a similar declarative, pure functional style: https://github.com/Monadical-SAS/redux-time/
For now you're welcome to sign up with a fake email, we don't require email verification.
The app is a mix of minigames and interactive tutorials. I moved through them quickly because they're all bite-sized enough to do in spare minutes while waiting in line etc. The tutorials are short, single topic, interactive lessons that introduce a concept, give an example from tournament play, and sometimes quiz you about what the next move should be.
The minigames seemed dumb at first, (and I still hate Dream Escape) but after playing them a few times in order to continue advancing through the lessons I came to see them as well designed "wax on, wax off" style drills. 'Poker Face', looks like a memory game but it's actually teaching you to quickly evaluate which player would benefit more from exchanging pieces. Beach Bounty helps you instantly visualize where a piece could be n moves in the future. Flight Control, where you try to quickly tap the appropriate square when shown the coordinates in standard chess notation, seemed pointless until I tried reading books about chess and found the notation slowing me down. Once I'd drilled the coordinates enough that they were second nature, I could visualize them as I read instead of constantly referring to a diagram and counting the squares.
In my lay opinion, Magnus Trainer is a brilliant bit of educational technology. It's not teaching you anything you couldn't learn elsewhere, but it felt easy and fun to me. If I hadn't stumbled across it I don't think I ever would have delved as deeply into chess as I did.
[1] https://www.youtube.com/user/AGADMATOR [2] https://www.youtube.com/user/STLChessClub
For actual playing I recommend something fast like 3min/0increment or 2/1. It kinda of resembles a video game at that point, where your knowledge and speed is helpful, you get to play a lot of games since they're short, and you don't have to live with your mistakes for very long if you mess up
I should play longer games I guess. I do much better on puzzles when I have time to think through the move. Donno if the lichess ratings are comparable but my puzzle ratings are much higher than my blitz ratings...
There are a lot of good suggestions (based on my admittedly limited experience) in this thread: https://www.quora.com/What-are-the-best-books-for-a-beginner...
There are endless resources on YouTube including lectures from grand masters on various topics from various chess clubs.
I've mixed lichess training, actual playing, and their tutorials with YouTube videos and I'm having a lot of fun and improving slowly. I can recommend Ben Finegolds' lessons as fun and instructive and various other members of the club he belongs to also have good lessons.
Just be warned, it's very addictive!
https://catswhisker.xyz/log/2018/12/16/instructive_chess_you...
Also, at least in my opinion, the very limited nature of the total set of states relative to the total set of states in something like an RTS lends itself to repeat engagement and analysis. You can, and likely will, see exactly the same board position in more than one chess game. You almost never going to see exactly the same game state in multiple SC games. That changes the dynamics of post-game analysis and skill development, IMO.
(Don't get me wrong, I enjoy both- but I do enjoy them differently.)
This makes it reactive, rather than proactive. Reactive judgments are driven more by depth of knowledge, habit, and memorization. Whereas proactive judgments are driven more by predictive/entrepreneurial reasoning. Both seem equally driven by calculated risk with a little psychology involved.
Ancillary skills are important in start craft, I'll give you that. But at least in DotA those skills could be learned by anyone relatively quickly. I don't play video games any more so no other examples come to mind unfortunately.
True with 1 exception, 1/0 (all moves in 1 minute per side) bullet, some of the really strong players at that use high end/high accuracy gaming mice because with so little time even a slight decrease in move time pays off.
I don't play anything less than 15/15 or 20/10 though and your point holds for that.
I don't know, but it is very addictive. It's a 'fight to the death', but in a realm of ideas. There's a neverending amount to learn, so many different types of situation can arise in a game, so many varieties of strategy and tactics, unlimited, it seems.
I'm not sure what you mean 'these have all the same Dynamics plus some. They are so dynamic'. There is a fairly common chess variant, Chess960/Fischer Random, where the pieces are shuffled before starting, and you will rarely see the same opening move. But it hasn't taken over from normal chess, which seems the better game to most people. The opening move won't be new, but after a few moves you're in uncharted territory.
In terms of books, Play Winning Chess by Yasser Seirawan is a great introduction to the game and first layer of strategy, tactics, openings, etc.
Fundamental Chess Endings (Muller and Lamprecht)
Fundamental Chess Openings (Sterren).
A good games database - The best free one is this http://caissabase.co.uk/ (It's my project so I might be biased but it is the best free one I've found, it exists because I couldn't find a great free one that was always updated at least monthly) and Chess DB software (ScidVsPC is open source).
That and playing and analyzing your own games is all you'd need (apart from a massive commitment to improvement) to get to Category A (1800-2000USCF.
If you feel like throwing in another (cheap) book I like https://www.amazon.co.uk/Mammoth-Book-Chess-Books/dp/1845299... as it's a weighty dense cheap book you can throw in a bag and not care about getting beat up.
[0]https://lichess.org/qa/6/how-does-the-rating-system-work-on-...
> I looked at my win rates by opening and I won far more as black than white.
Normally he should have been winning more as white, not less.
Also, ah, guess I hadn't gotten that far in divinity!
I remember as a kid, learning chess, I never heard of the 'en passant' rule and thus never used it. When I played my cousin one day and he used that capture, I actually leapt up in livid outrage and scattered the board pieces at this 'cheating' move he played. When I learned it was actually a legitimate move, I was totally red faced and humbled.
It's so good in fact, I can tell the difference between playing late at night or during the day, after an expresso, or a large meal. It remains super consistent while my concentration and skill wavers.
I highly recommend it - I'm having fun and slowly improving my game without that horrible frustration that comes from an opponent that's just too good.
(I think it would be interesting to learn how it makes itself less skilled - it's not just about search time, as that's what older programs did and they were either trivial to beat, or impossible.)
He quickly dropped those 200 points right back. In chess there is luck. Not luck like you hit a 9:1 longshot on the draw from a deck, but in other ways. Ratings don't account for personal differences. There are plenty of situations, all the way to the world champion, where you will have 3 players A,B,C of about the same rating. But A crushes B who crushes C who crushes A. One of the most famous here was Kasparov > Shirov > Kramnik > Kasparov. Keep getting paired as 'A' to your 'B' and you can hit a major hot streak.
There are also simple practical things. It's not uncommon to run into a player that's drunk, exhausted, on tilt, or whatever else that's just completely throwing them off their game. Which probably leads to maybe the most important point. In chess there is a huge element of variance in your own mental ability. It can be short term - did you sleep well, do you have anything grinding in your head, etc. It can also be longterm for a countless array of reasons.
Current world champion Magnus Carlsen was recently asked who his favorite player from the past is. He responded it was himself from 3-4 years ago. His game has undoubtedly deteriorated over the years for no great reason. He's probably just started to lose the motivation he once had. Nobody would argue against him being the best player alive, and few would argue against him being the best player of all time. Where's his motivation supposed to come from now?
And while those examples are negative, they have perfectly equal opposites on the other end. Sleeping extremely well, very clear headed, highly motivated? Things like this don't necessarily last, but when they're there - they can give you a very substantial boost.
So the lesson? There's variance and randomness in places you'd never expect to find it. In order to find genuine improvement it's important to be able to appreciate the difference between randomness and legitimate change. It's not easy, but I find there is a pretty simple clue for it. When you find yourself not wanting to play as much because you might screw up your rating - you have already subconsciously (and consciously really) accepted that you're overrated. This is a really common phenomena that I think we're all guilty of at some point or another. As the author of the article states, "As you can see in the chart, I played far fewer games in all of March to November than those sixteen days in February." That's a red flag for cognitive acute overratitus. Fortunately the cure is simple and fun - play more!
Also, when I've been doing lots of coding (systematic thinking), my score goes up vs. when I'm doing non-engineering tasks.
A couple of years ago, I spent a good little chunk of time searching something out, and I think I recall seeing a few interesting candidates, but I didn't follow through and don't have those results at hand.
https://news.ycombinator.com/item?id=18876042
https://online-go.com/docs/about
Online-go.com is made possible by the generous financial support from hundreds of individual site supporters, the guidance and welcome friendly attitudes of the Go community at large, and by a large collection of volunteers that have helped translate Online-Go.com into a multitude of different languages from all over the world.
FWIW, I’ve found that on Lichess I’m rated at least 100 points higher than what I would be on Chess.com (the most popular chess site), and maybe 150 points higher than what I’m rated by the US chess federation, which mainly rates live, not online, games.
A player with a rating of 2000 by the US federation would be called an “expert”, just below the 2200 rating that earns you the title of “master”.
You can't really compare ratings from across websites or groups like FIDE since the pools are so different.
As a wild guess, amateurs who only generally know the pieces and play casual games infrequently are likely to be less than 1000 ELO, so it's a bit of an achievement to be in the 1800+ range.
I remember when I was about 9, reading that Mozart started composing at 7, and thinking it was too late. So I didn't compose much until in my mid-30s, and really wished I'd started when I was 9!
Anyway, when you play online, you don't play the best in the world. You play people about as good as you are. And it's super-easy to find them online. I mostly played on FICS, which is very friendly, and I played and chatted with people from (what must be) nearly every country in the world. Most people play for fun, not to be the best in the world.
Walked away somewhat disappointed. :-)