On Learning Chess Backwards

Strong players often talk about "learning endgames first".

For those who don't understand what this means, here's a basic introduction to it:

The idea is to learn chess by starting simple, like this:

The player (with the white pieces) learns how the king can help "escort" a pawn toward promotion.

They also learn the King & Rook versus King checkmate (or the King & Queen versus King checkmate ), which are some of the most useful, and instructive, patterns in chess.

Once they've mastered this, the instructor (or the player) can begin adding more pawns, for each color, to the endgames, making them a little more complex, and a little more difficult, according to the player's comfort level.

Something like this:

Eventually, some minor pieces can be added to the endings, making them even more complex and challenging. Something like this, for example:

Little by little, the player is mastering the art of playing chess, starting from the endgame, first.

(It's also important that white's position be superior to black's, so that the learning player won't become discouraged by positions that are too difficult.)

Eventually, (likely after a few months) the player will be ready for all the pawns and pieces to be on the board.

This will be: the Opening Position.

At this point, the player should be fairly skilled, with a grasp of basic tactics, material advantages, and basic strategic maneuvering.

Now they're ready to start at the point where, ironically, a lot of beginners usually start: with opening principles, and basic opening theory.

Thus, when learning backward, openings are actually the final stage of learning ... not the first.

And now you understand what strong players mean when they advise players to: "learn endgames, first".

What do you think of this approach? Have you had any experience in a system like this?

I think its a really interesting try, because it makes sense for the player to learn the basics (How the pieces move, basic checkmates, etc.) first, all the while they are learning endgame theory they are supposed to learn later anyway. In the early stages of chess learning, opening theory is useless, because even if they get pass the opening, they will not survive the middlegame, much less the endgame that follows

And now I realise I have learnt chess wrong.

What do you think of the concept of starting over everything you know? Like, I am mid-beginner level now, but what if I start over using the method you described?

I think it would improve your chess a lot because it will deepen your understanding of basic and complex endgames, and that will, in turn, improve your middlegame technique, because you will know if you are entering a winning, drawn, or losing endgame by trading. Some endgame tactics and techniques apply to the middlegame and even the opening sometimes.

I believe it would make you a stronger player, for sure.

Though you can you still do whatever you're currently doing, as well, for additional fun and enjoyment. No reason that you can't do both.

I agree completely.

Learning openings, for beginners, usually leads to players who get progressively weaker, as the game moves toward the ending ... which isn't ideal, in my opinion.

Interesting! I agree. 

An interesting idea, but a player working in reverse would perhaps understand less of the complex middlegames and lose before they can get to a "winning" endgame position. I like the idea of learning the basic checkmates and endgames, first, just don't go too far in reverse- A lot of ultra beginners I have seen play chess without knowing basic checkmates. If you don't know how to win, why play? 

Very fascinating concept, though. 

Ah! Great! Thank you for the advice. 

One problem with this advice is some beginners try to follow it by getting a beginner level endgame book... which AFAIK all teach technical positions... which aren't useful for what you're suggesting.

Learning endgames first by focusing on endgame strategy is very useful because you learn how to coordinate your pieces, make plans, that sort of thing... unfortunately I don't know of any endgame resources that teach this at a beginner level.

Good point.

Yes, some technical endgame positions can be quite difficult ... certainly not at all appropriate for learning beginners.

This is why it's better (IMO, anyway) to learn from endgame positions where white has a clear material advantage ... so the player can learn to convert winning positions, without getting too discouraged.

They can, perhaps, "graduate" to harder endgame positions later on, where things like tempo and opposition begin to play a role ...

Same here. I found Reinfeld and Horowitz's "First Book of Chess" in my grandparents' basement, and it was what started my love of chess.

A great first book.

But then I dove into Nimzowitsch's "My System" and Fine's "Ideas Behind the Openings", which were very instructive, but miles above my comprehension.

I thought I was learning great things, but for the most part, I was simply confused.

By the time I found a coach, I had developed a lot of poor habits, and thought of myself as an Opening Theorist ... who would blunder instantly, the moment things went out of book!

Partially, I made this thread because this approach is how I'm teaching my wife ... working from the fewest amount of pieces, and adding more, little by little. It seems to be the only method she'll tolerate.

Along the way, I've found myself thinking the same as you - that it'd be nice to start over from scratch, and avoid the things that were incorrectly learned the first time ...

It is also the way Capablanca structured his "Chess Fundamentals"

Your idea is interesting.  I think the missing piece is to infrastructure to allow people to follow it.

For example, I have tried the rook endgame challenge multiple times on chess.com.  I am good at Lucena, have won all the challenges at least once and I am 1/3 of the way through 100 Endgames you should know.  But I cannot solve them all 5 in a row yet.  Looking at the leaderboard, only 31 people have completed it in the past 90 days.

I think that problem set is too tough at that difficulty level.  To implement your ideas, there should be a number of setups of increasing difficulty (read: closing material gap) and leaderboards to gamify completing the challenges.

I would also be quite interested to see tourneys where the initial positions came from endgame theory or famous books like the Shereshevsky book.  In each round, the pairings would play both black and white - like in bridge.  You score would be the sum of the two results / 2.

"Gamifying" this approach is a fascinating idea.

Currently, I'm using this approach in an intuitive way ... simply setting up positions that fit with the student's skill level, then gradually increasing the number of pawns/pieces, per their comfort level.

The tricky part with making this a "challenge" sort of approach is that such challenges usually depend on the player finding the "best" solution. Though I believe, when it comes to learning, a player need only find "good" continuations, and not necessarily "the best" ones.

Comfort and understanding should be the primary goal (in my opinion). Accuracy and precision should come later.

But I do like your suggestion. Very thought-provoking ...

Learning the end games step by step is instructive from a perspective of the power of the pieces  and looking at the whole board, which is something beginners can miss - that bishop on b2 or g2 that suddenly sees the whole diagonal with just the simple move of a knight or a bishop that was in blocking it; when you just start with a clear board, it is more innate to take in that control from 2 squares away, or right in the corner.

The slight caveat is that not that many beginners games get to the end game - generally a few pieces have been blundered and someone has an overwhelming advantage.

Like anything, why just learn in one set path? In learning a language you can be learning the grammar (verb tables, declensions - don't try to do prepositions until you've been learning a language for at least 28 years though) for half of your lesson, and spend the other half looking at real life conversations. In chess you can do end game study AND play a few games and analyse how you open, and your middle game.

I'm thinking of actively trying to experience this by doing the endgame situations highlighted in James Stripes' book "Essential Tactics: Building a Foundation for Chess Skill". (In a lot of the tactics there, Stripes talks about how the post-sequence position provides winning opportunities.)

Wonder if I'll be able to combine that with the endgame scenarios in his book called "Five Days to Better Chess: Essential Tools"?