Short Answer
> Quick answer: King and pawn endings first, and specifically the opposition and the square of the pawn. Then basic rook endings, above all the Lucena and Philidor positions. Those two groups decide more club games than everything else combined. Minor piece endings come third, and everything else can wait years without costing you anything.
The order matters because king and pawn knowledge is what tells you which trades are winning. Study it first and you start making better decisions in the middlegame, long before an endgame appears on the board.
Why the Usual Order Is Backwards
Most endgame books are organised by piece type, starting with material you rarely hold and finishing with the pawn endings you reach constantly.
That structure suits a reference book and fails as a study plan. Opening a five hundred page volume at chapter one gets you queen versus pawn positions, which are genuinely fascinating and will decide roughly none of your games this year. People work through thirty pages, lose interest, and conclude endgames aren't for them.
The order that works is frequency weighted. Learn the positions you'll actually reach, in the order you'll reach them, and stop when the return drops. That's a much shorter list than any book implies and it's finishable in a few weeks.
There's a compounding effect too. Knowing pawn endings changes how you play the middlegame, because every trade is a question about what endgame you're heading into. Knowing queen versus pawn changes nothing until you get there.
I spent a winter on rook and bishop versus rook and cannot recall a single game where it mattered. The week I spent on the opposition paid for itself in about four games.
King and Pawn Endings Come First
This is the foundation and it isn't close.
The opposition is the single highest value idea in the endgame, because it turns a position that looks like a coin flip into one with a known answer. Once you can see who has it, king and pawn positions stop being guesswork and start being calculable.
The square of the pawn is the second piece, and it takes about ten minutes to learn properly. Being able to tell at a glance whether a king catches a passed pawn saves you from both losing races you could have won and entering ones you can't.
Add key squares and the basics of triangulation and you have most of what club play demands. Our guide on king and pawn endings covers these positions with the concrete patterns rather than the theory language.
The payoff arrives earlier than you'd expect. Knowing that a particular queen trade leads to a won pawn ending is middlegame knowledge dressed as endgame study, and it's the reason this group goes first rather than because pawn endings are common.
Then Rook Endings, and Only Two of Them
Rook endings appear more often than any other piece ending, which makes the small amount you need here unusually efficient.
The Lucena position is the winning technique with an extra pawn, and the bridge building idea behind it is one specific manoeuvre you can learn in an afternoon. Without it, a completely winning position gets drawn regularly.
The Philidor position is the drawing technique when you're the one defending, and it saves half points constantly. Between them, these two cover the majority of practical rook endings that arise with a small material imbalance.
Beyond those two, the useful additions are the principle of keeping your rook active and the habit of cutting the enemy king off along a file or rank. Both are ideas rather than positions, which makes them easier to carry into a real game. Our rook endgame guide goes through both positions move by move.
What you can skip for now is the enormous body of theory on multi pawn rook endings. Those get genuinely hard, and they're hard for strong players too, which means your opponents are unlikely to punish you for not knowing them.
Minor Pieces Come Third
Bishop and knight endings matter, and they matter later than most study plans suggest.
The concepts worth having early are simple and comparative. Bishops prefer open positions and pawns on both wings, knights prefer closed positions and a single side, and a bishop that's blocked by its own pawns is worth considerably less than it looks. That comparison alone guides most of the trade decisions you'll face.
Opposite coloured bishops deserve a specific mention because they invert the normal rules. Being a pawn or two up is often not enough to win, and being a pawn down is often enough to draw comfortably. Knowing that changes whether you head into the ending at all.
Bishop and knight mate is the famous one and it's the clearest example of study time badly spent. It's genuinely difficult, it takes real effort to learn, and most players go entire careers without reaching it once.
Practise the trade decisions rather than the mating patterns. Deciding correctly which minor piece to keep is a skill you use in most games, and the puzzle trainer puts those decisions in front of you far more often than they come up in casual play.
What You Can Safely Ignore
A surprising amount, and being explicit about it makes the rest feel achievable.
Queen versus pawn, rook and bishop versus rook, and the more exotic material imbalances are all real endgames with real theory, and they're all rare enough that learning them now is a poor trade. They'll still be there when you're strong enough to reach them.
Deep theoretical rook endings fall in the same category. Once there are three or four pawns a side, the theory becomes long and the practical value drops, because both players will be playing on general principles under a clock anyway.
Fortress positions are worth recognising rather than studying. Knowing that fortresses exist, and that some material advantages simply cannot be converted, prevents you from grinding a dead position for two hours. That's a concept, not a curriculum.
Skipping these isn't laziness, it's sequencing. Every hour spent on a rare ending is an hour not spent on the pawn endings that decide your games this month, and the ordering is the entire point of having a plan.
How to Actually Practise Them
Reading endgame positions and playing them are different skills, and only one of them transfers.
Set the position up and play it out against an engine or a bot from both sides. Winning the Lucena as the stronger side teaches you the technique, and defending it teaches you what your opponent is trying to do, which is what lets you recognise it early in a real game.
If you would rather not set positions up by hand, Lichess keeps a free set of endgame practice studies that drill exactly these positions against the engine, and working through the pawn and rook sections there covers most of this plan without any preparation on your side.
Do it repeatedly rather than once. These are motor patterns more than calculations, and the goal is reaching a point where the bridge building idea appears without you having to derive it. The practice bot is fine for this and you can reset a position endlessly without an opponent losing patience.
Then look for these endings in your own games, including the ones you won. Games that ended in a queenside pawn race you happened to win are exactly where the opposition would have told you the answer in advance rather than by luck.
Keep the whole plan to a few weeks. Opposition and the square of the pawn, then Lucena and Philidor, then minor piece trade decisions. That's a finishable list, and finishing it puts you ahead of most opponents at club level who have read chapter one of an endgame book three separate times.