Endgame Training

Endgame Training

May 24, 2015, 5:46 AM |

A few months ago I began working seriously to improve my endgame play. I had been meaning to do this for a long time, but what finally convinced me was losing from a completely won position in a game against my invariably superior but charitably helpful older brother. Instead of getting my first ever victory over him, I faced the ignominy of admitting to myself that I had been using every kind of avoidance strategy to avoid doing what I needed most. I’m sure many people recognise the symptoms: too much time devoted to tactics, online blitz, learning new ideas in the opening, watching videos, playing through recent games, even re-examining middlegame imbalances. All helpful; all good; all serving my deeper, unacknowledged goal of avoiding the endgame. The truth is, studying the endgame is hard work. And it takes a lot of time. Different time from the opening, which can be studied by glancing at lines and ideas, adding them into your database. The endgame requires calculation, which means uninterrupted attention. It’s not so easy to find this kind of spare time in the midst of a busy life.


Now, let’s be clear, it’s not as though I knew nothing about how to play once queens come off the board. I had sporadically examined various books and videos about technical endings. You know the sort of thing – the basics of king-and-pawn endings, how to mate with various pieces, and so forth. I had dabbled with various lists of important techniques in books by Jeremy Silman, Axel Smith, and others. I had watched several excellent video series by Daniel Rensch. I felt I had done my duty, covered my bases, lit fires at the altars of Caissa. So why was I still playing terribly in the final stage of the game?


The answer, it turns out, was quite simple: all of my study had been theoretical, none of it practical.


Okay, but before I got around to realising this I decided to fix the more obvious problem: my endgame training had been too sporadic, unsystematic, intermittent. So first of all I set myself a course of study. I began reading Jonathan Hawkins’s book From Amateur to IM. It’s as good as the reviews suggest; it’s focused on the endgame; and it’s much more practical in its orientation than many endgame manuals. At this point I began to realise that my training had hitherto lacked a practical dimension. But I still felt I needed to be a bit more systematic in my study of technical positions, so I went back to a book I had only browsed through before – Jeremy Silman’s Complete Endgame Course. I made myself work through even the simpler, more familiar material. For the last chapters, which are aimed at higher-rated players than me, I glanced through to pick up the ideas rather than doing all the exercises in detail. Where did all this study leave me? Well, I’d recommend both Hawkins and Silman – I learned a lot from both books – but I now felt I needed to start studying puzzles and playing more endgames.  



As it happens, at this point I got lucky. I stumbled upon Edmar Mednis’s Rate Your Endgame (revised by Colin Crouch), a fantastic book that made everything else start to fall into place. What’s special about this book is that you don’t just do individual puzzles designed to illustrate particular themes, you play through entire endgames and get marks not just for finding the right move, but also for finding second- and third-best moves, and for getting your analysis correct too. The games are organised by theme (knight endings, rook endings, etc), but the introductory sections do not give hints as to tactical motifs that you should be looking for. You really have to do the analysis yourself from scratch, and it is your positional understanding, your plans, and your knowledge of how to manage the transition to technical endings that gets you rating points, as well, of course, as hard calculation. The rating system may or may not be realistic – I can’t say. But what is great about it is that you can assess your progress and engage your competitive side to keep you motivated. This was the kind of endgame study I could relate to! I spent hours on each chapter (too long to be realistic), playing endgames that tested whether I could put theory into practice, whether I really understood what rook activity meant and how to weigh it against other factors. I really cannot recommend this book enough. I wish it were reprinted and that more like it were produced. The other thing that I found really useful in it was the way in which the introduction explained what it meant by an endgame – not so much a technical measure of what kind of material was on the board, but a judgment about what kind of play was required. Mednis argues that many players lose endgames by trying to play them like middlegames, and he gives examples. All very helpful. When I’d finished reading the book I wanted more like it. I haven’t found anything quite as good. But what I needed, of course, was to play endgames of my own in the way that I had played those in Mednis’s book.



I won’t take you through examples of my games – I’m still a learner, still in need of more over-the-board play rather than online games. But, believe me, though my endgame play is still patchy and a world away from exemplary, it has improved immeasurably. My study has changed too. I now consult endgame manuals like Dvoretsky or Mueller and Lamprecht not to read them systematically, but to learn more about positions that I’ve just played or misplayed. When I’ve finished an interesting game, I take notes on my plans and calculations, then try to re-examine them and see how I can improve. Every so often I try to think back over recent games and identify things I’ve learned, patterns and trends, or areas that I still need to work on or reconsider. Out of all of this I might draw the following general points:

  1. Knowledge of technical endings (such as rook v bishop) is most useful if you know how to build it into forming plans in more complex positions.
  2. Calculation and tactical awareness are critical in almost any endgame. Spotting a zwischenzug or zugzwang is often what makes or breaks a player's analysis.
  3. In a practical endgame with some pieces still on the board, seizing the initiative and creating threats is often as valuable as it is in the middlegame.
  4. Knowing whether king safety or king centralization is a bigger factor in any given position is often the key to understanding how to approach the entire endgame.
  5. Computer analysis of the endgame is very often close to useless.


None of this is rocket science, but I think it has helped me, and I imagine many others would benefit from the same process I’ve gone through and am still going through. Maybe some day, if I ever again get a winning advantage against my brother, I might just manage to convert it!