lichess.org

How do I improve at the endgame?

"I don't often get to endgames in my games, simply because I'm likely to make a tactical blunder before then."

Then you should not study endgames. You should get better at not hanging pieces/pawns.

Even if you do study endgames and even if you get to the endgame, your extra knowledge will not help you avoid Knight forks and other tactics. And if you hang a pawn in the endgame, it's usually a lot worse than hanging one in the middlegame.

So you should focus on tactics for now.

Focus on tactical endgames.
Mate in one, two, three and four.

If you mastered openings but not the rest, then your opponent has nothing to worry about. I would be more concerned about someone that has a high puzzle rating, then a late bloomer for their openings. It's like the turtle and the hare. The hare has a great start, but runs out of juice to finish the game.

If I want to go somewhere, I most know what I will need over there. If I do not know what I will need over there, I might as well not go anywhere. It goes the same for the endgames.

Imagine seeing a player capturing all the opponent pieces and not knowing how to checkmate with two bishops. That's proof enough that you need to learn endgames first.

Playing out end game techniques should be easier than playing out the tactical middle games or the strategical opening repertoires.

Studying theoretical endgames is important because, in a lot of cases, even though you encounter endgames with superfluous pieces, if you know the theory you can recognize the endgame patterns and ignore the pieces that don't matter or won't make a difference.

I love Jesus de la Villa's "100 Endgames You Must Know," and I think it's a great jumping-off point for endgame study.

A good book is "Endgame Strategy" by Mikhail Shereshevsky.
I also like "Winning Chess Endgames. Just the Facts" by Alburt and Krogius.
And of course Nunn's Understanding Chess Endgames.

@biscuitfiend as @Toscani pointed out, the argument goes the other way - what have I achieved if I can play a game with zero blunders but I have no idea where I'm headed? In fact Magnus Carlsen said that most games between players under 2000 are decided by tactical blunders - does that mean no endgame study until 2000 ELO?

Magnus Carlsen in a tournament game this year blundered a piece for 1 pawn. He still managed to win the endgame…
Endgames up to 6 men can be taught exactly. Middlegames and even more openings are no exact science: there are opinions, different sources give different advice.
Capablanca: " the opening is to prepare the middle game, the middle game is to prepare the endgame."

You should 100% study and remember and recognize all possible endgame motifs. There are very many - King opposition, passed pawns, minor piece endgames, the list is semi-endless.

Compared to learning 10's of thousands of possible Openings in Chess, the list of possible endgame motifs is fairly limited. Learning endgames will also feed back into your general mid-game play too.

How to finesse a pawn move or to achieve a winning position mid-game can often be derived from the afterwards winning endgame. The many endgame themes do repeat themselves. Quite rewarding in general play to know the very many endgame objectives. Many do apply also to mid-game principles just as well.

I would 100% agree with Toscani here. To not be able to apply basic mate with mating material and advantage is kinda naive. To not know that R+N vs R is a draw is also a basic mistake. Knowing about opposition themes and endless winning stuff for White or Black in endgames, really helps you to structure your later mid-game approach

I feel like there are two discussions going on at once here. Let me clarify my position.

When I hear "endgames", I don't think "basic checkmates". Of course you should know how to checkmate against a bare king (or an essentially bare king). In fact, if you actually want to win games, you should know how to checkmate in more than just that scenario: you should practice checkmates as a subject unto themselves, e.g., by looking at all the "Checkmate Patterns" studies on lichess in the "Practice" section.

When I say "focus on tactics", I mean that there's not much point in studying, say, how to evaluate bishop vs knight endings, when you blunder material every game. That information has some value, but it would be much more valuable to you to simply stop blundering pieces - that way, if your opponent blunders a piece, not only will you have the tactical vision required to see it and take it, but you can then be sure you will win the game as you won't blunder it back. Eventually you will find yourself competing mostly against players who do not blunder so much material; perhaps then you can look at some basic endings, since then the information will actually be useful.

Being able to look at the board correctly and see basic tactics is something you can and should train yourself to do; it's not some magic talent that some people have and others don't. And if you aren't able to do this consistently, then it is unlikely that you will be able to correctly apply your endgame knowledge anyway.

tl;dr: You have to learn how to walk before you learn how to run; and you have to learn how to stand before that.

If you google search "Dvoretsky final interview chessbase", then you can read at leisure the very great Russian Chess coach Mark Dvoretsky three-part interview just before dying a couple of year's back. It's quite revealing.

The great Capablanca once-upon-a-time once stated to start your Chess study with the endgame (and never the Opening) and then work backwards from there. Not the actual quote, more the essence. Google the real quote, if you want.

Dvoretsky's interview is highly revealing. MD was a mathematician and a chess IM, who then trained so many Soviet greats and who wrote so many great Russian books. MD divided endgame material into 2 categories - of theoretical and practical applications. You can know one well, but the other not so well. You must know both. In Rook endgame theory for example, you have the basic Philidor defense, where the defending side exploits a very well-known 3rd rank defense or the well-known Vacuna positions (where the pawn 6th rank or-less, is rook-supported by a white rook guarding the pawn from ahead) or the easy win of the Lucena position thema etc. All must-reads, all must-practise too.

Dvoretsky's instant classic is his very iconic endgame manual, his other (maybe) Classic is "Tragicomedies in the endgame", which shows the other side of the coin. Near brilliance quite often, but ultimately fatally flawed moves. Both are very nice books.

Dvoretsky was brought in to coach Anand vs Kasparov in 1995 (the WTC match in NYC). MD immediately spotted 2 weaknesses in Anand's play. Anand was actually quite weak in endgames, as was later cruelly exposed in WC versus Karpov in 1998 also - in the infamous game of 2 rooks versus Queen endgame there. Seemed nice, but the Karpov's King can march across the board, just eat a pawn (cough) and then the Queen's checks run-out and a simple loss. 2nd weakness was that so wonderfully intuitive was Anand ...that he almost never very deeply and precisely calculated variations.

After game 10 in NYC though, with the match tied 1-1, then Kasparov simply played the Yugoslav Dragon (as Black, with the new ideas of h5 etc) and Anand initially played ultra carefully with Kb1 and Ne2 ...but failed to calculate the fairly easy 28. Nxe7 line, instead pursuing the illusion of the b6 knight fork which Kasparov hit with the stunning Rxc2 move. In the next Yugoslav dragon game, Anand even did not castle at all, which surely amounts to basic chess suicide versus Kasparov. The attacking moves were just too compelling to ever think about simply castling. But ...you must always castle, in these lines! MD abandoned coaching Anand at some point in this era, because player now willing to listen and/or learn from basic flaws.

MD also gives some great insight into Kasparov too. Kasparov also fairly so-so in endgames and ultra-reliant on his incredible Opening preparation (unmatched in that pre-computer era, and backed by dozens of top GM's working night-and-day for him). The very young Kasparov being highly innovative in unknown positions, then ever-more slowly becoming highly distrustful of facing opponent innovations. Mostly Karpov hitting him with newer ideas, then Kasparov would do everything to just simplify and quickly look to draw and the return with even better home analysis next game. This stuff was a big feature of the infamous and monumental Karpov-Kasparov battles.

MD also identifies Toplaov, Yusupov and many other top GM's who were quite weak in endgame stuff. Better after they had received his infamous trainings. MD so methodical here. 100x typical positions (or ideas or themes) in Chess, all sub-divided and card-indexed. All themes then further quantified by 10x Chess positions on a physical A4 piece of paper. Each player maybe weak in 10 or-so areas, thus first identify and then very specifically target these areas. Could be a general weakness or could just be psychological too ...like Anand cold-refusing to deeply analyze positions at any point.

MD quite glowing in his praise of Fischer, Carlsen, Kramnik, also Karpov too - as very well-rounded players. Although MD also states that Carlsen is perhaps not the very best in (rook) endgames or even in general play ...but his chess intuition is by far the best atm.

MD also states that the #1 learning tool is mostly self-analysis and knowing and identifying your own weaknesses and then learning from them. Don't make the same mistake again. Karpov concurred here too, in advice. Analyze with friends, not alone and also identify your own weaknesses. One Dvoretsky lesson in endgames is very simple -rooks are very special then. You can often even sacrifice a pawn or else destroy your structure (with doubled pawns or iso-pawns or else with a cut-off King) just to mobilize your Rook(s) fully then. Good advice, I would like to think, as endgames and Rooks just go hand-in-hand so very often.

PS: Very many Russian Chess books have been translated to English, but countless others have not been (yet). Not too difficult to simply read all Russian Chess literature though, like Fischer who taught himself the basics of "Chess Russian". Just self-learn the Cyrillic alphabet (about maybe 2 weeks work for the 33 characters) and then learn about 75x chess-specific Russian terms -like ferz', clon, peshka etc or exchange/sacrifice/en prise etc and about another 75x general-purpose Russian words (all very Chess related) like take, threaten, promote etc. You can 100% forget about the 30+ possible endings for the adjective "Black" in proper spoken or written Russian, as they are all irrelevant here. Just the root of the word Black is all that matters (which is chyorn). Your mind will just fill in all the blanks then, gradually anyways. Russian chess books are just wonderful, are mostly very simply written (the positions are why you buy them, not the prose) and Russian books are very highly instructive, by nature.

This topic has been archived and can no longer be replied to.

Reconnecting