Hacker News new | comments | ask | show | jobs | submit login
How to Get Good at Chess, Fast (2013) (gautamnarula.com)
258 points by andyjpb on Dec 9, 2015 | hide | past | web | favorite | 64 comments

How to get good at anything (cribbed from the excellent Talent is overratted)

1) Practice just above your level (enough to be difficult and stretch, but not demoralizingly hard). Get a coach to help you find this range. Do this as many hours per day as you can tolerate, and the number you can tolerate should go up over time.

To quote the book, practice:

"It is activity designed specifically to improve performance, often with a teacher's help; it can be repeated a lot; feedback on results is continuously available; it's highly demanding mentally, whether the activity is purely intellectual, such as chess or business-related activities, or heavily physical, such as sports;"

2) Be clear about your goals and motivations. These will help you get through the difficult hours in 1)

The book is excellent and lays out the basics about getting good at anything. (Notice that I'm not saying anyone can be world chess champion. The book argues that it would be highly unlikely that any chess champion would get that way without applying these principles.)

A lot of what is discussed here is the what and how to practice. These are very valuable questions that I gloss over by saying, find a coach. Presumably a coach would know this. Knowing what to practice, and designing practice is a skill in itself.

Talent Is Overrated: What Really Separates World-Class Performers from Everybody Else Kindle Edition

by Geoff Colvin


$13.99 Kindle

an expansion of this article:

What it takes to be great


Any NFL fan will want to read the chapter about Jerry Rice, as the book puts it:

As every football fan knows, Jerry Rice was the greatest receiver in NFL history, and some football authorities believe he may have been the greatest player at any position. His utter dominance is hard to believe in a league where the competition is so intense and conducted at such a high level.

Yes indeed. Apparently, he was the hardest working player in the league [1], to which the book attributes to his greatness, and longevity.

[1] http://thegameology.blogspot.com/2010/11/game-training-compi...

Sorry. What is the book you referring to?

Is it this one? Talent Is Overrated: What Really Separates World-Class Performers from Everybody Else by Geoff Colvin

Stuck in the mid 2100s. The main difference I see when I play 1800s, and most levels beneath that, is still tactics. A 1200 lets you fork their king and queen. An 1800 goes into a variation that lets you fork their king and queen 3 moves in.

I make the same kinds of mistakes, allow positional concessions or just straight-up blunder because I overlook tricks in the lines I thought were safe.

Building up that mental repertoire of tactical and positional patterns by solving puzzles and reviewing well played games is essential, no matter what level. It's like building a huge in-memory cache of correct answers so you quickly solve problems at the board.

Interesting, I am myself stuck in the mids 1900 (used to be 2000+), and I feel that my tactics are quite good but it's my positional understanding that keeps me under 2100.

My experience is that tactics is a spectrum. You can always improve your tactics, no matter who you are. There is no player who never makes tactical mistakes.

Whenever you feel like it's positional understanding that holds you back, that just means that your tactics are at a higher level than your positional understanding, so your opponents are less able to exploit tactical mistakes in the games you play. Once you improve your positional understanding, your rating will rise until you are facing opponents who are once again able to take advantage of you tactically.

If you're someone who enjoys intense strategy games like Chess (and Dominion and other modern euro board games), but you're disappointed in Chess for some reason or another, I've got a game for you--- Go. Here are some chess players speaking about Go:

"While the Baroque rules of chess could only have been created by humans, the rules of go are so elegant, organic, and rigorously logical that if intelligent life forms exist elsewhere in the universe, they almost certainly play go." -- Edward Lasker, chess grandmaster

"Chess has only two outcomes: draw and checkmate. The objective of the game . . . is total victory or defeat – and the battle is conducted head-on, in the center of the board. The aim of go is relative advantage; the game is played all over the board, and the objective is to increase one's options and reduce those of the adversary. The goal is less victory than persistent strategic progress." -- Dr. Henry Kissinger, quoted in Newsweek, 11/8/04

"What's happening with chess is that it's gradually losing its place as the par excellence of intellectual activity. Smart people in search of a challenging board game might try a game called go." -- former Wold Correspondence Chess Champion Hans Berliner, The New York Times, Feb 6, 2003

Learn to play: http://www.usgo.org/learn-play

The Seattle Go Center has events multiple times a week: http://www.seattlego.org/

I think you lose a lot of potentials by comparing to (or outright bashing) Chess too much... That said I have come around to seeing Go as the better game after a coworker taught it to me, but it took Hikaru no Go and realizing that playing mirror moves is punishable before I decided to get into it more with books and try to reach my equivalent former chess level of 10 years ago, which I think roughly matches somewhere in the single-digit kyu range.

I've been to a few lectures at the Seattle Go Center, and can confirm it's pretty great. Unfortunately I live in south Renton and work from home so I don't make it up there much.

I, too, watched Hikaru no Go and then took up the game myself. Wound up playing a few games with guys at work, it's a lot of fun.

Huh, even if the bashing quotes came directly from chess grandmasters? I figured it was a good appeal to authority that a discouraged chess player would respect more.

Well the biggest obstacle to getting good at chess is finding a solid reason why it's worth pursuing, as opposed to getting good at something else (or doing nothing).

Good point! Some reasons: 1) large and international community of players meeting and playing all the time, both in person and online

2) its a game you can play at every age, so your investment in the game pays off over your lifetime

3) all your games can be easily recorded for future analysis

4) fun for hackers: there's lots of software to analyze and manage your game history

5) also fun for hackers: there are massive databases of games at every level

6) hundreds of years of published analysis that will become interesting as you get good

7) playing well requires you develop and exercise a wide range of cognitive skills

I'm going to a tournament tonight at my local club. I'm terrible (USCF 1224) but playing is a rush.

You can do all those things and more with other games. Eg: street fighter.

I'm just agreeing with xyzzy4; finding the game fun is more important than anything else.

Personally, I don't like the fact that luck is a bigger factor that chess players are willing to admit. Sometimes you make a move that anyone would consider strategically good, but little did you know, that move would put you in a bad situation many moves later, and only a machine or maybe a grand master could know it.

> You can do all those things and more with other games. Eg: street fighter.

While you're right (all ages can play Street Fighter), it's hardly something that one can expect to be common in, say, 45 years. Chess will certainly still be played ubiquitously.

Furthermore, Street Fighter requires good reaction times, hand/eye coordination, etc. A better counter-point might be something like poker (or another card game).

That said, you're right: if you don't find chess (as a game) interesting, you probably shouldn't spend time pursuing it :)

Street Fighter has been here for about 26 years, and it's only growing. It's going professional, with sponsors for the players and everything :p

And by the way, I too thought the reactions needed were insane. But after some time, you realize all that is just muscle memory, and the hard part is the strategy. It's a conversation between two minds.

I'd like to add Scrabble to this list - it will be around in 45 years for sure, though the number of people playing is orders of magnitude smaller than Chess

Provided you like it, Scrabble can also be nice to acquire new vocabularies in foreign languages.

True, in fact the french scrabble champion recently doesn't even speak french, he just memorized a ton of words: http://www.npr.org/sections/thetwo-way/2015/07/21/424980378/...

This is interesting in and of itself. I think it would be interesting to have a linguistic conversation surrounding it. This is akin to memorizing gibberish (in this case, words without meaning). Which is more difficult than it sounds!

I'm amazed that he also (intuitively) figured out "the French sound" (the linguistic term escapes me) well enough to be able to call out words for not being words. I wouldn't be surprised if many French speakers tried to sneak some high scoring words past him, knowing he can't speak French only to be flabbergasted when he called their bluff.

French has less hybrid linguistic heritage than English, so its orthography is much more regular. Scrabble is a very English game.

One critical skill in chess is the ability to analyze the consequences of your various options, several moves ahead. If you find yourself in a situation you didn't foresee, that's not bad luck, you just need to work on improving this critical skill.

I think high level Street Fighter play has the some factor. But! I'm not arguing one is better than the other. If someone wants to play chess, play chess.

Your definition of "randomness" is interesting (but classifying it as a problem is misleading). Of course, if you don't have profound knowledge/skill of a game and you're not able to evaluate moves, their outcome will seem random just because you're ignorant of them. That's exactly what differentiates good and bad players! If the game were not random for newcomers and they could predict the outcome of every move, they would win or lose solely based on starting conditions (i.e. the game becomes perfect play tic-tac-toe).

Chess isn't random, it's chaotic. Where randomness is unpredictable, chaos is deterministic but so computationally complex that it can't practically be solved. Chess is the art of navigating that chaos.

That definition of randomness is actually fine (randomness as ignorance), it's consistent, and the most often used definition actually. For example, take the Monty Hall problem: it doesn't matter what process Monty actually uses to choose doors, all that matter is you're ignorant of it (so you choose the uniform prior).

In fact I don't think many processes in nature are truly random (quantum phenomena have to have a large effect and you need to choose a no-hidden-variable theory), but it doesn't really matter. It's just a model.

This is very relevant. In the past, Chess was the highbrow game in the Western world--the one played and esteemed by those who believed themselves to be above the median. Becoming good at Chess was therefore a means to impress those people, and in turn everyone impressed by the opinions of those people. If you were good at Chess, you had at least one socially redeeming quality.

There was a brief period where the ultra-intelligent switched to the highbrow game of the Eastern world, Go, likely due to its greater difficulty at grandmaster levels, and its well-established handicapping system that allowed for entertaining games between people of different skill levels.

Now, there are thousands of games played by elites, and none of them have such a clear advantage that anyone would put their leaderboard position on their resume. I already know how to play Chess well enough to let my boss win (without making it obvious that I threw the game), and I have plenty of other games that I can play when I actually want to have fun.

So why would I spend any more time practicing just that one game? I could try to improve my Go ranking, or play Hive, or build a new M:tG deck, or get better at Risk.

Many of the reasons people had in the past to play Chess well are no longer applicable exclusively to Chess.

>esteemed by those who believed themselves to be above the median

An interesting counter point to this idea:


That 1859 article-writer is lucky he lived before computers were invented.

His head would have exploded in moral outrage, from all the people engaged in non-productive amusements, to the general detriment of society.

Those who attempt to make other people ashamed of having fun are just the worst. I catch myself being judgmental and snarky sometimes, but in keeping with the nerd code, I have never intentionally tried to make someone else feel bad about themselves for liking something I don't like. My personal niche obsession might be infinitely superior to yours in all possible ways (in my opinion), but that doesn't make you a bad person. It just means you're from a different nerd tribe.

> or get better at Risk.

Shame. Axis and Allies man. Choose the highbrow games of each category!

I guess DOTA is the "highbrow" game of the video game world, at least as far as prize pools and money. Maybe fighting game players are more elitist (but they tend to back-up their status with extremely in-depth understandings of the game... at the frame-by-frame level)

Pokemon also has a very dedicated following, and is relatively "high-brow" at least within its own community.


I think star craft was, I think DOTA like games are much simpler, but are easier to pick up and understand.

I consider it a worthwhile life skill. There's always a chess set somewhere.

if you have a reason why it's worth pursuing, you will never become really good. You'll become above average at best.

The number one reason why people become good at anything is that they first became obsessed.


Maybe because there is no rational reason.

Take chess (I have 2000+ fide elo, so I have some experience here). The practical value of being good at chess is close to zero, unless you can make money of it. Now, there are some 1500 GMs in the world, and most of them can't even make enough money from chess to become a professional. If you approach this with a rational mind, you will never even start putting in time. The same reasoning works for almost everything (sports, arts,...).

I've played chess casually for years, and my main observation is that no one actually wins if both players are reasonably well matched. The balance tips when a player makes a mistake that leads to them losing. Consequently, the only thing you really need to do to not lose is to make sure you don't make daft blunders, and to recognise when your opponent has. I'm pretty sure it's different at higher levels where people don't make those errors nearly so much, but if you're just starting out, rather than trying to win just remember to not screw up.

1200-1300 is about the rating where players stop making blunders to such a grand scale. Also, at that rating, blunders start to get punished much more heavily. Before 1200, it really is all about learning to stop blundering. At GM level though, they're gaining advantages at minuscule levels, until it leads to some "big" breakthrough, which they can continue to exploit.

At some points you'll move from mistake vs correct play to finding imbalances. Most things aren't really mistakes but contain advantages and disadvantages. Matching all these imbalances together to a strategy and at the same time disturbing the opponent from putting his imbalances into synergies, that's what expert games are about (or at least higher advanced level, I can't say I really understand expert level).

"The winner of the game is the player who makes the next-to-last mistake."


Just like Connect 4.

Accidental downvote from fat fingering the mobile site, sorry.

Upvoted your post to restore cosmic balance.

I started playing chess at a fairly young age (5-6) and moved into competitive chess vs adults at age 8 and playing near national level at around age 12. In all that time I never read a single book about how to play chess, or try memorize openings. What I did do is play a lot of competitive chess and discuss strategy / tactics at every opportunity with strong players, both in terms of my own games and also in general.

The original article pretty much nails it as a guide, in my opinion. Highly recommended.

The massive discussion from two years ago is interesting:


He only mentions it indirectly but two things are crucial to your game as well, and you should spend significant amount on that:

1) Get good ressources. He mentions a few books, but they may not be best suited for you. There are also a lot of other tools, like training websites, programs, video guides, classes. Reason is: Instead of experimenting out the best path through some ideas, you can read about them and get that realisation much faster. Having better ressources and knowing how to use them may cut 80% of your time to achieve a certain goal.

2) Find the right better players. Not everyone that plays better can also transfer some of that to you. And some people are simply just not right for you.

Where books etc are the way to learning the right thing (instead of everything), the right teacher is about giving you better feedback. You can only see your own lacks and mistakes so far. A lot of your stupidest mistakes you simply don't even see as a mistake. For instance, one of my major problems in hitting (kung fu training) was that I slightly lean back while stepping forward. How could you even realise that without a teacher/experienced friend?

Find good resources, find good teacher, learn from the mistakes/experiments of others what is the right path, then do it a lot, then get feedback from the right person, then continue exercise and feedback sessions for a few years. Recipe to learn anything.

"Chess and Programming - You have to be smart enough to understand it and dumb enough to think it matters."

If you want to revive Chess you should play Chess960.

It was such a bummer for me to explain to my son that you don't really have a choice of moves in the opening...

A little study of openings definitely improves your games, but in ordinary Chess the opening is about as interesting as bowling or watching golf on TV.

1) Chess 960 is cool, but really misses a lot of the interesting parts of chess. Also completely ridiculous to talk about "reviving" chess since chess is at near peak popularity. Unclear what goal you're suggesting could be achieved. Chess on TV? Chess a mainstream activity? I think there are lots of reasons why this is really difficult starting with chess is very very complex due to the availability of perfect information and several hundred years of people studying it. It's still more popular to watch videos of people playing chess than it would be of people doing biomedical research.

2) Openings are not as dead as you think. There are certain lines that are more well studied than others, but until you're 2650 FIDE (don't worry, 99.99% of players won't get there) you can deviate and have just fine results.

3) I think your thoughts about the opening are probably mis-informed. Look at the games from the recent london chess classic here:


In round 3 Vachier-Lagrave already spent 9 minutes as white on a single move at move 9 in a open sicilian (maybe not as few choices as you're suggesting).

Again in round 4 he spent 5 minutes on move 9 in a sicilian and then 12 minutes on move 11.

I have a feeling that the kid's frustration has less to do with deviating on move 11 in an open Sicilian than with not getting to play 1. a4, 2. Ra3, for example.

Chess 960 isn't going to change that. Plus I bet a player would have only a minor effect on his rating (the improvement might be slower) from playing 1. a4 any except 1.. e5 or 1.. e6, 2. Ra3 every game.

I'm rather mediocre (1700ish on Lichess), but I feel one can't really improve without playing long games. Blitz is addictive and doesn't require cutting a committing chunk of time out of the schedule - but it remains a game of chance, I believe it even develops some bad habits such as guessing, or gambling moves (bet he won't notice that...). What do you think?

Started playing at a young age. Got really obsessed again for a few years in college when online play was first coming out (Yahoo! Games, FTW), but it's not something I've played for awhile. The advice is sound, especially the "just study one set of openings and know it really well".

However, Chess is less fun than alot of other options because it does take more practice to get really good and it's less social than board games (e.g. Settlers of Catan, Pandemic, etc.), card games (Poker, Hearts, Bridge), or online video games.

The plus side is the logical planning and longer term strategy boost from study. The downside is that it's not going to help social interactions as much and alot of what it teaches is contrary to alot of what you need to pick up now, which is how to navigate among many different groups some of whom may be working for or against you. In many ways, running a raid in an MMO teaches you more useful skills than what it takes to get a really good ELO rating.

What would be a right way if I just wanted to play very aggressively, you know, to impress or scare friends? :-) Lots and lots of tactics/puzzles? (SFMBE)

This wouldn't make you "good at chess", but you can learn a bunch of opening traps and mate patterns like the Scholar's Mate and the Légal Trap. Some traps are wicked simple (ie, e4 e5, nf3 nf6, nxe5 nxe4, qe2. If black moves their knight to safety, nc6+ sacrifice wins the queen.).

Does anyone have an estimate for how much time "50 puzzles a day" takes? Surely if the puzzles are actually stretching you, they'll take a few minutes each? Is 50 realistic for an amateur player?

Tangential, but:

1) best app for Mac (once upon a time I'd care if it was pretty, but not so much now) for computer chess? 2) best communities for 'beginner-enthusiast' and online games?

I would like to know how to get from 1800 to 2100, fast. I'm stuck at 1800 level for many years(I'm not an active player though).

From the article:

    To improve quickly you
    need to play often.

Finetuning is only reasonable if you are actively playing and are still stuck. It's unlikely, though, if you are an attentively playing, watching and thinking about your games.

I have no idea why this article is so popular. It claims to show the way to get better at chess by the doing the following: 1. Study the opening, 2. Study the middlegame, 3. Study the endgame, 4. Study your own games, 5. Study master games, 6. Play OTB, 7. Study tactics, 8. Use "psychology," meaning, don't be afraid and never give up. Along the way a few books and other recommendations are thrown in.

This is all rather trivial stuff.

While, I don't disagree with everything that the author says, I think that it should be acknowledged that his claim is pretty much unfounded. There are those who have not done any (or all, or much) of these things and their rating is above 1800, and there are those who have done all of these things and their rating never comes close to 1800.

Beyond that the author does not seem to be aware of the well-known debate on this subject.

There are at least, broadly, two views on chess improvement. One side, as represented by Silman and Aagaard, argues that chess "meta-knowledge" is key. A player must first look at the characteristics of the positions (e.g., understand Silman's imbalances), and with this understanding, and only then, can a suitable move be found. This group usually advocates a "thinking process" as well. The other side, as represented by Watson and Hendriks, argues that it is only the moves themselves that are important, and the correct move in many cases contradicts the "rules" of strategic analysis. For this group, only "concrete analysis" of a position (i.e., looking at the moves without prejudice) has any possibility of leading to a good move. In this "concrete analysis" group, at least Hendriks (if I understand him correctly) argues that in a tournament situation where there are significant time restrictions, and experimentally moving the pieces is not allowed, only knowing the position itself, or similar types of positions, can help the player find the correct move.

So, we have the "meta-knowledge" group advocating the learning of strategic and tactical ideas, and then applying that knowledge to a given position with proper thinking technique. Accordingly, this group believes that if you want to improve your chess, you need to learn more strategic and tactical ideas, applied with an improved thinking technique.

And we have the "concrete analysis" group advocating the learning of the correct move in specific positions. Accordingly, this group believes that if you want to get better at chess, then you must learn many hundreds, even thousands, of positions.

The truth is probably somewhere between the two extremes. My own experience is that as far as tournament OTB play is concerned, I have benefited more from the concrete analysis approach than anything else.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact