Related: How to improve on the early endgame?
How exactly do I study or learn endgame theory? Is it just memorizing certain positions that are quite common and knowing how to attack and counter?
Chess Stack Exchange is a question and answer site for serious players and enthusiasts of chess. It only takes a minute to sign up.Sign up to join this community
The endgame, as said by many top-level players, is the most difficult part of the game. It decides whether you will win, lose or end up drawing the game. This is the only reason why Magnus Carlsen ends up grinding the win in almost every endgame (except those which are drawish).
Endgames require a lot of practice or I should say concept-building practice. There are tons of endgame theories out there, like Lucena, King Square, Rook vs knight or bishop, equal pawns, etc. One must understand the concept behind that theory and not memorize that particular position. Instead, one should try to memorize the pattern, because you will not get the same position every time (just like tactics). The best way to study endgames is by practicing the endgame positions. This way people will build the habit and required intuition for the endgame.
One cannot become Magnus Carlsen in the endgame in a month or two, not even a year. It requires you several years of daily practice and an intuition-building process to master it. After lots of experience in practicing and playing endgames, you will find out that the moves will automatically come to you, and you won't have to think long to find the best move.
It looks like you want to know how to play endgames.
The reason that Magnus Carlsen is a superb endgame player is not because he has memorized a lot of standard positions or that he has memorized "the patterns behind them". It is that he has a wonderful imagination and nobody can tell you how to acquire that. Like a sculptor, he sees the statue inside the stone and knows how to release it. I am not completely sure that you do want to "Learn endgame theory". You may want to "learn how to play endgames" which is a bit different.
There a great many posts related to this. As far as I can see, they mostly contain good advice that I will not try to replicate, but there are a few rules to follow if you are to avoid common mistakes.
Realize that the King is a strong piece. If there is no risk of being mated, his fighting power is roughly that of a minor piece, but unlike either a Bishop or a Knight, there are no gaps in his armor. From an early stage, ask yourself how your King will be used. As the endgame approaches, whenever you have a spare move, consider nudging him a square closer to the action.
Avoid passivity, unless there is really no choice. This applies especially to Rooks, which can get bogged down in a maze of Pawns. Rooks need open lines to display their power. If a Rook is the only piece that can defend a weak Pawn, and if doing so would reduce its mobility, you should always try to find a counterattack instead.
Remember the "Rule of Two Weaknesses". One weakness (isolated Pawn, exposed King) can often be defended, but if there are two, the attacker can often switch between them, and the defender may not have have enough manoeuvrability to keep up. Good endgame players know how to do this. Great endgame players know how to create the situation.
Never make an unmotivated Piece exchange. Weak players often exchange a piece for no better reason than that it is possible and will result in there being "less to think about". Strong players usually prefer to leave all options open, at least until the strategic objectives have clarified.
Having said this, there is some indispensable theory, available in many books. When can you win with your last Pawn, under a variety of conditions? What is, or is not, mating material? But above all, teach yourself to enjoy endgames.
"Checkmate ends the game". So, the first step in learning endgame theory is learning the basic checkmates - KRR vs K, KQ vs K, KR vs K, KBB vs K and KBN vs K.
There is no point in reaching a simple theoretical won endgame if you can't finish it off. A cruel opponent will refuse to resign and you will be embarrassed by the 50 move rule. This first step is vital and fortunately it is also the easiest.
The second phase involves learning what kinds of positions you can turn into one of these basic mating combinations and how to get there.
There are three key components:
Taking the simplest combination of K+P vs K the result depends on where the pieces are and whose turn it is. For example white king on e1, pawn on e2 and black king on e8. Is it a white win or a draw? The correct answer is that we don't know until we know whose turn it is. White to move is a win. Black to move is a draw.
There are key squares in relation to the position of the pawn which the white king must be able to reach to force a win. If black can prevent that then the result will be a draw. Like the simple mates you just need to learn this stuff.
KRB vs KR is a theoretical draw but there is a difficult way to win if black goes wrong. It is important to know the ideas involved so that you can win with the RB when your opponent goes wrong and similarly when you just have the R so you don't throw away half a point. With the extra bishop there are positions from which you can win. You need to know what they are and how to win from there. That way you know what you are aiming for if you have the extra bishop and you know what to avoid if you just have the rook.
In my opinion the best endgame resource is the two DVD volume set which GM Nick Pert produced for Ginger GM. I thikn it has all the endgame knowledge you will ever need in audiovisual format which many of us find easier to absorb than books.
Start with the most fundamental endings such as king and pawn vs. king, king rook and pawn vs. king and rook.
Then learn how more complex endings are made of combinations of simpler endings. For example, you may see how to exchange a complex ending into a winning king and pawn ending.
Finally learn how to transition from a complex middle game to a complex winning ending.
practice -> study -> practice -> study -> practice -> study
For a beginner, probably working backwards is most productive:
If you have K and Q and the opponent just has their K, you should know how to checkmate in a short amount of time (and avoid stalemate).
1b. This means if you have K and P against K, and can queen the P (without stalemate), you will win.
If you have K and R against K, you should win in a small number of steps, but it requires more thought than K and Q. But it should be known to you like the back of your hand.
Everything beyond this is more complicated... K and one or two minor pieces against another K and minor piece is complicated.
But/and if you can see how to simplify to one of the easy cases, then you have won. (And try to avoid allowing your opponent to reach that: e.g., if your opponent is ahead by one pawn, don't allow them to trade all pieces to leave the situation at K+P against your K!)
Get a good endgame book. I would reccomend Silman's but any book that teaches the basics is probably fine. Even Reuben Fine's book is probably Fine...sorry for that. Fine's book is what I started with but it's pretty dense and more of a reference work than instructive.
Also, learn to analyze the position.
Look at the pawn structure. You should be mostly concerned with passed pawns and potential passed pawns. The pawns will dictate how the endgame goes.
Bring your king to the center as soon as the endgame starts
Advance the unopposed pawn
Create a passed pawn
Principle of creating a secondary weakness
This is a very instructive endgame I saw in a book once. It will teach you everything I said above. In this case the unopposed pawn is the b pawn. You want to advance it and create a passed pawn on the queenside. Black will be forced to stop that pawn but you will then move over to the king side and raid his pawns creating a secondary weakness. It's impossible for black to cover both weaknesses.
Practice against a strong engine until you get it. This is a forced win for white so you should be able to beat any engine of any strength. Once you get it you'll be better at the endgame than 90% of the players out there.
6k1/p4ppp/8/8/8/8/PP3PPP/6K1 w - - 0 1
End-game is the easiest part for learning and everyone should start learning chess from end-game and then can migrate to middle and start-game, you can solve some puzzles, read books(Lasker method is recommended) and chess magazine, watching videos and also chess.com and lichess websites has some good training materials for learning end-games.