Hacker News new | past | comments | ask | show | jobs | submit login
How To Think About Chess (taaalk.co)
132 points by FailMore on Mar 31, 2015 | hide | past | web | favorite | 50 comments

Some diagrams and actual moves would have been very useful to help explain the concepts he talked about.

As it is, I am left with no idea what "attacking the king in the midgame" really meant to him. Did it mean sacrificing material for a mate, or a more standard development?

Most midgame analysis use the concepts of force, space, and time[1], and he didn't even mention these terms, so I was left confused.

I am not sure what my level is, but pretty much all I did in my two years in prison was play chess, and in my relatively short stay, I became around the 5th best player "on the yard" and was able to play at the specific table where only the best players could play.

There were over 1500 inmates at the facility I was at, and at least half of them played chess as it was one of the only board games readily available, so playing at that table was no mean feat.

We would play for $1 a game and it became my "hustle" (a hustle in prison is something that allows you to make money doing something of value, like making some else's rack (bed), or sneaking food out of the kitchen etc).

Anyway, I digress...I try to play at coffee shops when I can these days but I find often the existing players seem snobbish and its hard to find a game with the better players.

[1] http://chessimprover.com/force-space-and-time/

As a person of color, I really got this:

"At the very very highest levels winning with black is really quite an impressive achievement, and black will often go into a game playing for a draw. The advantage of the first move is very pronounced, and sets the tone for the entire game, with white generally calling the shots. Attempts by black to generate active counterplay are often very risky. "

"So the aim when black is to survive and make sure you're doing enough sensible things to not get exploited because of the disadvantage of moving second."

That reminds me of the billiards scene from Boomerang:

"The white ball dominates everything. Knocks the shit out of the yellow ball, the red ball, right? And the game's over when the white ball drives the black ball completely off the table."

You'll be glad to know that in GO, the ancient Japanese game which is played with black and white stones on a cross-hatched board, black goes first and has the advantage.

Reminds me of the Bill Cosby outtake


That's a pretty funny double entendre.

I have an expert level rating. I agree with everything in the content. My games are very similar to what he describes. I have not memorized many openings. I just know a few moves in common openings and how to respond to a few quirky responses to them. I want to get to the middle game or out of my opponents memory as quickly as possible. I also sometimes deliberately make some opening weak moves to get out of opponent's memory.

One main thing I found in my experience is, majority of the games in amateur level are decided by blunders. I am not talking about small positional bad moves. These loose pieces are get mated in one move. If you are an amateur first focus on avoiding them by studying tactics before memorizing anything else.

I am a terrible player, but I absolutely love playing. One of my biggest problems is making a move and them immediately seeing the err of my ways. It's beyond bizarre how quickly I'll realize what a shitty move I have just made right after I take my hand off the piece.

I'd say I have a few decent tactics that I sort of lean on, but, like you pointed to, if my opponent does something to break me from that memory, my game is shot and I'm flailing around trying to recover (which rarely works).

Not sure if I'll ever dedicate enough time to reach "decent" status but I definitely enjoy the game.

This. I take my hand off the piece and immediately resign. Allowing a single 'undo-last-move' play would help my game a lot.

I am also an expert player, and agree with most of what was said, but I do think I should point out that this player's style is not the only strategy ("My strategy is usually to throw lots of pieces at their king and hope they die or I can win some material"). Personally, I prefer more closed/quite games, so most of my thinking (even in the middle game) is about how to achieve a favorable endgame, and win from there.

EDIT: I dislike studying openings too, (I started playing 1. g3 to avoid theory while still getting into some fun positions)

I was surprised to see nothing about pressure and the whole pins/forks/discovered attack options that restrict your opponent's choices.

Beyond the basic "stupid mistake" level of play, chess centres around forcing your opponent into a situtation where they have two different risks that they need to defend against and only one move to do it, while not allowing your opponent to do the same to you.

Those dominate much of the thinking of a player in the 1200-1800 range, but become second nature after that point.

I'm a pretty bad player (mostly playing with my 7yo son), and I totally agree about the blunders thing. The most effective way to learn how to avoid them IMO is to play a lot against good algorithms - that might not play hard tactically at the easier setting, but never make stupid moves or fail to detect my own stupid mistakes. You can't win these games by lucking out :)

Not all computer chess programs behave like this, some of them, at the easier levels make stupid mistakes, or don't take advantage of my own oversights. I try to avoid these. Playing against other amateur players that don't always take full advantage of your stupid mistakes doesn't train you well enough for this.

Hi Everyone,

I'm one of the founders of Taaalk and also the person interviewing my knowledgable friend Robert Heaton on this one.

We'd love anyone who might be interested in being talked with or talking to someone else (either someone we'd set you up with, or someone you know) to reach out to us. Please email hi@taaalk.co or hit us up on twitter at https://twitter.com/gotaaalk.

Please don't be too modest in terms of getting in touch! We've found that when someone starts asking you questions you become more interesting than you realise!

Thanks very much for the chess discussion. Great to learn even more.


Besides the awesome content - which is great for starters to grasp some basic concepts of tactics and pay attention to some common patterns on the checkboard -

I love the format too.

It's really a conversation, a chat, but with solid content and serious people. So following is really easy, and satisfying. (unlike audio/videos of interviews here you can go back and reread lots of times a passage you lost - which is good.)

So Kudos also to the Taaalk team for the great design and format.

I like the format and the content as well, but the domain name and company name seem pretty terrible, if I'm to be honest.

Imagine explaining it to a friend or prospective user.

"Yeah it's called Taaalk, just go to taaalk.co"

"What, talk.co?"

"No, you need 3 As."

user types 3 more As

"Taaaalk.co is giving me a 404."

"Sorry, you needed 2 more As, let's start over."


"Just google 'talk'. Actually, wait, don't do that..."

And for that matter, what is the official pronunciation? "Talk"? Or do you have to elongate the vowel? How long should you elongate it for?

Seems like they should decide on a new name before they invest too much into this one.

Yeah, but who shares URLs with their friends by speech these days? In fact, who really speaks?

Filthy peasants.

Or "Yeah it's called Taaalk.co with 3 As"

Even if you say that, a not insignificant number of users are still probably going to type either taalk.co or taaaalk.co to start with.

A .co domain costs that much?

I spent AGES trying to choose an easily pronounceable, non ambiguously spelled name for last project as I expected to have to tell people it a lot. A typical conversation goes:

"Just go to playlistful.com and..."

"Urgh, can't you just send me a link?"

Speaking of chess, I strongly recommend this youtube channel:


I have no affiliation with him, but he is a brilliant trainer and he is a very lovely game narrator. I feel it is appropriate to talk about this channel, because I have not played serious chess in decades but just watching this guy go through these famous games, gave me such insight that reading this article (the on from OP) sounded rather naive.

As much as I enjoy and watch almost every video, chessexplained is not very educational unless you're pretty high rated, it's very entertaining nonetheless.

The article's good. However, the comment about experts not getting to the endgame as often as non-experts is likely wrong. I don't have the data either, but I'm fairly experienced at chess, and in my experience high level games get to an endgame moderately often, but amateur games almost always end before an endgame. Amateurs typically don't even know how to play an endgame, because they rarely get to it so they don't realize why they should learn. I know a lot of amateurs focus a lot on the opening because they see themselves winning or losing in the opening, when that's more due to tactical blunders than anything essential to the opening.

It'd be interesting to see some statistics for this.

Agreed: the higher level the players, the more endgames. Think of two people fighting on a balance beam. The weaker the fighters, the greater the chance that one of them is going to slip off the beam (make a blunder that is punishable) earlier in the fight.

I like this. I've been getting back into chess as of lately and I keep a board on my desk while I do work on my laptop. Here's my strategy for playing 'solitaire chess' as I call it - I make a move, then I move a poker chip to the side whose turn it is, I go back to doing whatever I was doing on my laptop for a while (until I forget the game entirely), then I return and make another move. Then I rinse and repeat this process until the game is over - it's amazing how rapidly this improves your strategy compared to playing against another person or a computer. Because as they say - you are your own worst enemy... Or most brutal competitor as I like to think ;)

You'd want to be careful with that, Stefan Zweig has a novella about a guy playing chess against himself in his head who loses his mind! http://en.wikipedia.org/wiki/The_Royal_Game

I have to agree there is a slight bias towards White winning, but it really isn't a landslide [1]. And it's probably a cliche by now, but I always liked the saying (apparently by Tarkatower [2]):

    The winner of the game is the player who makes the next-to-last mistake.
[1] https://en.wikipedia.org/wiki/First-move_advantage_in_chess

[2] https://en.wikipedia.org/wiki/Savielly_Tartakower

The first-move-advantage becomes increasingly relevant the higher the level of play, of course.

I got really into chess for a bit, but started to find it sort of boring after a while. Midgames can be interesting but the early games are simply displays of memorization. Really incredibly good players end in draws more often than not.

Games like Starcraft are pretty much the exact same - there's a defined early game, a midgame and endgame.

I play chess "seriously" (expert level) and am about a gold/platinum league Starcraft II player, and the games do show a surprising amount of similarity. Both show the most variance in the midgame, begin in identical formations (where memorizing openings is common), and converge to endgames that have somewhat less variance than midgames.

Huge opening theory and high draw rate are primarily the reasons why I switched to another mind game called Gomoku. Gomoku, when played at a high level, requires as much depth as Chess. Go is a also good choice but I find Gomoku much easier to keep in my life.

>> there's a defined early game, a midgame and endgame.

all things with defined start and finish criteria can be conceptualized this way.

careers: education is early game setting you up for success, job/career is midgame, retirement is end game.

relationships: dating is early game, engagement/marriage is midgame, death/divorce is end game.

software development: envision project/write requirements is early game, coding/developing/testing is midgame, shipping and bug patching/customer support is end game.

... I could go on and on, but I'll leave that as an exercise for the reader

I think you've got the parallels for "early game" right, but the parallels for "endgame" wrong. The endgame is the phase directly before the end of the game, where two players who have each managed to avoid screwing themselves over in the early game, and who have each managed to match their opponent's multiple-front challenges and feints in the middlegame, finally are forced (usually by the game's increasing resource constraint) to "put all their cards on the table" and do the interesting and tricky things that make their playstyle unique.

right. the end game is before the end of the games. we're on the same page on that.

disconnect on our understandings is that I didn't define the game ending criteria for each example. in chess, it's either a checkmate or a draw that ends the game.

for my examples, my criteria are death for the career, the relationship ending, and the software product reaching the end if it's supported life.

The point was that checkmate is not "endgame," it's literally when the game is over.

Endgame is a part of the game, the late stages. It comes after the midgame.

It doesn't go "early game, midgame, checkmate."

Same for all of your examples, where you note the endpoint, not the endgame.

look, I get it.

everything has a starting point. (you decide what career you want) then an early game phase. (you go to school to get a degree) than a midgame phase. (you start work and collect paychecks during your career) than an end game phase. (uou live the retired life using the resources you earned during your career ) than the game actually ends when predefined conditions are met. (you die or whatever)

got it.

we are literally arguing in circles because we are not conceptualizing the problem the same way and not communicating clearly.

I think the thing trying to be pointed out here is that "retirement" is not inside what most people consider to be the game of a career. When you quit your last job, your career is over; that is the checkmate moment.

Retirement, on the other hand, is the equivalent of the time between games (or after you quit playing games) when you have a static ranking from your previous play (maybe you're a retired world champion, say), that other people treat as a milestone in their own development—something to reach or to pass.

Now, if you said life, then sure, the endgame phase of that is retirement. But the early-game phase of that is when your parents are providing for you, and only extreme incompetence (or genetic bad luck) can cause you to "lose."

again, we're arguing in circles because we are conceptualizing the analogies differently. our definitions of "career" are different.

oh well.

ca·reer kəˈrir/ noun 1. an occupation undertaken for a significant period of a person's life and with opportunities for progress.

There really aren't different definitions. It's a "work life", it doesn't include retirement.

When I played I basically refused to memorize openings and instead worked on developing my own. It was weak, and the highest level players crushed me, but players below them were often so thrown off by the opening that I was able to gain an advantage and win against people who would have soundly trounced me if I had used a standard opening. I ended up getting out of chess when I reached the point where I was primarily playing against those who were unfazed by non-standard openings.

My 2 cents, and somewhat related: everyone should read Josh Waitzkin's The Art Of Learning [1].

[1] http://www.amazon.co.uk/The-Art-Learning-Journey-Performance...

Good conversation at about my level (read: plays a bit, but not necessarily well). I've noticed that many new players seem to obsess on their Openings, while my skills are very much the Endgame (I'm great at holding multiple logic trees in my head, which becomes an asset late in a game when you can pursue these to a checkmate). Basically, if I can survive 15 moves against comparably skilled players, I will win.

But that's an "if". I thrashed a mate in about a dozen games in a row, including one where I called 'Checkmate in 8' which drove him mad. And then the following game he 'pantsed' me, drove through to a checkmate without me even taking a single of his pieces.

Is there a point here? I guess you'll rarely win a match in the Opening, but you can lose it. And maybe more people should practice end game positions instead of just playing lots of games from the start and researching Openings because of the literature volume dedicated to those.

Is there a playstyle of chess where both players agree to skip the opening, just setting up the board as if the opening were played "by the book" by both players and going from there? It seems like that'd be much better for getting amateur-level players practice in their middlegames, since they wouldn't need to worry about perfecting the opening before they could even attempt it.

This type of training game is more often used to practice a certain structure that arises from a specific opening, and thus is a type of practice used more towards the higher levels of play (International Master+).

The thing is, if two amateurs are playing, neither of whom deeply know openings, then the starting position (or 4-5 moves in) is essentially new territory, like what might arise for a Grandmaster 15-30 moves into the game. Does this make sense?

The slight variations in the opening tend to be important to the middlegame that follows. You can set up the pieces 8 moves into mainline Ruy Lopez, for example, but if you start from exactly that position every time then you just end up with a "second opening" after that point. Randomizing the start position is possible but undermines the whole "pure battle of intelligence, no chance involved" shtick that is part of chess' appeal.

Fundamentally with the exponential fanout of the moves but a relatively small set of choices initially, there's always going to be a point where the game transitions from "likely to be an exact position I've seen before" to "likely to be something new, but similar to previous games".

I don't know of human games that do this (except for practice), but this is done in computer games. See the Thoresen Chess Engine Competition (TCEC) where the first 8 moves for both sides are played by the book and then the engines take over. This is good as it ensures the engines are tested over a wide variety of positions.

I play ridiculous amounts of blitz and bullet chess and I can tell you that openings dominate faster games. The more bizarre the better.

But you had better know the main openings and a lot of sidelines fairly deep or you're toast.

Of course, my memory is next to useless, so for me, to really "know" the openings means I understand them. When my opponents plays weak moves "to get me out of my memory" that's when I really dominate.

Honestly, sometimes chess strategies are like rock-paper-scissors. Guys with great memory can sometimes trounce me, but people who use this "play weak moves to confuse my opponent" oh yeah, I eat that for breakfast.

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