The Best Chess Endgame Books, Compared by Level
Updated
There are only about seven endgame books that serious players keep recommending, and they are not alternatives to each other: they are written for different readers. One sorts the material by your rating, one gives you a short list to memorize, one is a reference you look things up in, and one is famous mostly for being harder than the player buying it expected. This page says which is which, so you buy the one that matches where you are.
The reading is the easy half. The ending still has to be played.
The short answer
Beginners, under 1400: read Silman’s Complete Endgame Course, and only the chapters for your rating. Intermediate players, 1400 to 1900: read 100 Endgames You Must Know, and finish it. Advanced players, above 1900: keep Dvoretsky’s Endgame Manual and Fundamental Chess Endings on the shelf as references, and read Endgame Strategy for the part neither of them covers, which is what to do when no theoretical position is on the board yet.
And if you already own one of them: the reason endgame books get bought more often than they get finished is that reading a method and executing it are different skills. You can play the same positions out here against a defense that does not cooperate, for free and without an account.
The books, side by side
| Book | Author | Written for | Best for | Watch out for |
|---|---|---|---|---|
| Silman's Complete Endgame Course | Jeremy Silman | Beginner to master, split by rating band | Telling you what to learn now and what to skip | Thin on rook endings once you pass its middle bands |
| 100 Endgames You Must Know | Jesús de la Villa | Intermediate club players, roughly 1400 and up | The short list of positions worth memorizing | Assumes the basics; not a first endgame book |
| Chess Endings: Essential Knowledge | Yuri Averbakh | Beginners, in one sitting | The smallest honest introduction to the subject | Stops early on purpose, and reads its age |
| Endgame Strategy | Mikhail Shereshevsky | Players who already know the theory | The plans, not the positions: how to steer an ending | Teaches no theoretical position: it is a second book |
| Dvoretsky's Endgame Manual | Mark Dvoretsky | Advanced club players and above | Precision and depth, at a difficulty most readers underestimate | Most buyers under 1800 do not finish it |
| Fundamental Chess Endings | Karsten Müller and Frank Lamprecht | Any level, as a reference rather than a read | Looking one position up and getting a modern verdict | Encyclopedic, so it is a poor book to read through |
| Basic Chess Endings | Reuben Fine, revised by Pal Benko | Readers who want the historical reference | Breadth, with pre-computer analysis to treat carefully | Pre-tablebase analysis: verify anything that decides a game |
How this list was made
Seven books, not ten. The cut is that each one has to answer a question the others do not: where to start, what to memorize, what the minimum is, how to plan, where the exact truth is, what to look a position up in, and what the historical record says. A book that only does one of those better than a book already listed is not a different recommendation, so it is not here.
- Level is the axis, not quality. The list is not ranked, because “best” is meaningless without a rating: the right book at 1200 is the wrong one at 2000.
- Every entry carries a caveat, in the table above. A comparison where every row is a recommendation is a shelf, not a comparison.
- No prices and no edition numbers. Both drift, these books are all in the same $20 to $50 band anyway, and a price that went stale unnoticed costs more trust than it ever bought in usefulness. Check the current one at any chess bookseller.
- No affiliate links, and nothing sponsored. We sell an endgame trainer, we do not sell books, and there is no link on this page that pays us.
- The endgame claims are tablebase-checked. Where this page says something about a position (which endings are solved, why a 1941 reference needs care), that comes from the same tablebase verification every drill on this site runs through, not from another book’s summary.
Reviewed September 2026 against the current editions in print. For the full list of what has been published in this field, including the specialist volumes deliberately left out here, Wikipedia keeps a survey of chess endgame literature.
What each book is actually for
Silman’s Complete Endgame Course: the one that tells you to stop
Jeremy Silman, 2007. The book is organized by rating band rather than by material, and each band ends by telling you not to read further until you have gained the points. That single editorial decision is why it is still the standard first recommendation: the usual way a beginner wastes a year on endgames is by studying rook endings before they can win king and pawn against king. The bands are also honest about how little you need early: the basic mates, the opposition, and the square of the pawn cover most of what decides games under 1400. Those are the first three things we drill too, in the basic checkmates, the opposition and the square of the pawn.
100 Endgames You Must Know: the short list
Jesús de la Villa, New in Chess, English edition 2008. The premise is a claim about frequency: a small set of theoretical positions accounts for most of the endgames a club player will ever have to convert or hold, so learn those by heart and stop. It is the most finishable of the classics and the one most club players actually complete, which matters more than depth at that level. Its core is rook endings, because those are the ones that keep occurring: the Lucena position, the Philidor defense, the short-side defense and the Vancura position. A separate workbook of exercises exists for the same material.
Chess Endings: Essential Knowledge: the minimum
Yuri Averbakh. Short, old, and still the cleanest answer to “what is the least I can know and not be lost”. It is a booklet rather than a course, which is the point: it can be read in an evening and it does not pretend to cover the subject. Use it as a filter before committing to Silman or de la Villa, not as a substitute for either.
Endgame Strategy: the plans rather than the positions
Mikhail Shereshevsky. Every other book on this page is organized by material. This one is organized by idea, and it covers the phase the others skip: the ending that is not yet a theoretical position, where you have to form a plan rather than recall one. Do not hurry, create a second weakness, activate the king, and think in target positions rather than in moves. Those four ideas do more for a 1700 than another twenty memorized rook endings, and they are the ones we wrote guides on: do not hurry, the principle of two weaknesses, king activity and schematic thinking.
Dvoretsky’s Endgame Manual: the hard reference
Mark Dvoretsky, first edition 2003, revised repeatedly since and edited by Karsten Müller after Dvoretsky’s death in 2016. It is the most precise of the general endgame books and the most frequently abandoned, for the same reason: the analysis is exact, the exercises are hard, and a lot of it assumes you can hold long forcing lines in your head. The book marks its essential material separately from the deep analysis, and reading only the marked part is a legitimate way to own it. If you are below 1800 and it is your first endgame book, it is the wrong one. The deep material is mostly rook endings and the exact pawn endings built on corresponding squares and mined squares, which is also the part of rook endgame theory that repays exact study.
Fundamental Chess Endings: the one to look things up in
Karsten Müller and Frank Lamprecht, Gambit, 2001. A single-volume encyclopedia covering every material distribution in order, and one of the first major endgame books written with computer checking as part of the process. It is not a book you read; it is the one you open when a specific ending has just cost you a game. If you only ever want one endgame reference on the shelf, this is the most complete choice per centimetre.
Basic Chess Endings: the historical one
Reuben Fine, 1941, revised by Pal Benko in 2003. It was the standard English-language endgame reference for half a century and it is still worth owning for its breadth, but it was written before computers and parts of its analysis have since been shown wrong. Read it as history and as coverage, and verify anything that matters against a modern source or a tablebase.
What tablebases changed
Every endgame with seven pieces or fewer is now solved. Not estimated: solved, with the exact result and the exact distance to it known for every legal position. That has settled arguments the older books could only reason about, and it is the reason a 1941 reference needs treating carefully while a 2001 one mostly does not.
What it did not do is make the books obsolete, and the reason is worth being precise about. A tablebase returns the move with the shortest forced win. That is frequently not the move a human can find, remember, or repeat, and it comes with no explanation attached. The knight’s move box for king and queen against king is slower than the fastest mate, and it is still the right thing to teach, because a player can execute it every time under a clock. Books survive tablebases because a method you can carry beats a move you had to be told. You can read more on how tablebases work and what DTZ means.
Reading an endgame is not the same as playing one
The common failure with all of these books is not comprehension. It is that you follow the winning method on the page, agree with it, and then fail to produce it three months later with a clock running and an opponent who declines to play the book’s losing move. The gap is not knowledge, it is execution, and it closes only by playing the position out.
That is what this site is: the same classical theory, played rather than read. You take a position, the opponent defends with maximum practical resistance instead of resigning early, every mistake is explained the move you make it, and a position you missed comes back mirrored and colour-swapped so you keep the method rather than the squares. Every goal is verified against the tablebase, so what the drill tells you is winnable is winnable.
Common questions
What is the best chess endgame book for a beginner?
Silman's Complete Endgame Course, because it is the only one of the classics that tells you where to stop. It is split into rating bands and asks you to read only your own, which removes the usual beginner mistake of studying rook endings before knowing king and pawn versus king. Averbakh's Chess Endings: Essential Knowledge is the shorter alternative if you want the minimum in one sitting.
Is Dvoretsky's Endgame Manual too hard?
For most players below roughly 1800, yes, as a book to read front to back. It is dense, the analysis is exact, and the exercises assume you can calculate long forcing lines without a board. The book itself marks its essential material separately from the rest, and reading only that is a legitimate way to use it. Below that level, Silman or de la Villa will teach you more per hour.
100 Endgames You Must Know or Dvoretsky's Endgame Manual?
They answer different questions. De la Villa's book is a short list of positions you are expected to know by heart, chosen for how often they decide real games. Dvoretsky's is a reference that covers far more and goes far deeper, at a much higher difficulty. Most club players get more from finishing the first than from starting the second.
Do endgame books still hold up now that tablebases exist?
The methods do, the individual verdicts do not always. Endgames with seven pieces or fewer are now solved exactly, and that has corrected specific evaluations in the older literature, Fine's 1941 book most visibly. What tablebases have not replaced is the human method: a tablebase gives you the move, not the plan you can remember and repeat.
Can you learn endgames without a book?
You can learn the positions without a book, and most players learn them badly that way, because reading a technique and executing it under pressure are different skills. The practical route is to take the method from one book at your level and then play the same position out against a defense that does not cooperate, until the method survives contact.
Disclosure: ChessEndings is an endgame trainer, so we are an interested party in a page about endgame study. We are not affiliated with any of these authors or publishers, none of this is sponsored, and there are no affiliate links on this page. Our own curriculum is built on classical endgame theory from public sources that predate all of these books, including Philidor’s Analyse du jeu des Échecs of 1777, and every position in it is verified against the tablebase rather than copied from a book.