The Improving Annotator

[cool tip]

Article 9288 in
Subject: Re: Best Quick Study??
From: (Kenneth Sloan)
Date: 22 Jun 1996 15:00:58 -0500
Organization: Dept of CIS, Univ. of Al at Birmingham

Annotate your last 20 tournament games.
Kenneth Sloan

In a book by Dan Heisman called The Improving Annotator he attributes his success to self-study of this sort. If Kenneth's advice seems not enough to go on, here are some things to look for to Assess Your Chess {all}.

How good is your annotating?

"Most commentaries in chess magazines and books are superficial and sometimes just awful. Once a certain experienced master explained to me how he worked. You put two fingers to the page with text on it and see that there are only moves under them - in other words, it is time to make a comment. You write something like "The Ruy Lopez always leads to a tense, complicated struggle" - and your fee goes up by a rouble.

  "The ability to distinguish between real feelings and thoughts and this kind of verbal facade will be of use to you, and not just in chess." -- DVORETSKY


Inspired by the popular format of How Good is Your Chess?, where readers are invited to come up with moves which are given points in order to assess their standard of play, I have wondered if the business of annotating could be placed on a similar scientific footing. You can use this system to review books, or to assess how well you make notes to your own games.

  I have long had in mind Phil Crocker's criticism of chess publishing in Kingpin No.22, to which Batsford's, for some reason, felt the need to respond in No.23. As I remember, Phil's main gripes were:

  (a) shallowness: books are full of glib, vague or uncommittal commentary (e.g. "with compensation", no need to mention if it is adequate!) - only general and strategic comments, none cashed out in terms of variations

  (b) padding: complete games are included for no good reason at the expense of extra commentary or additional detail. I do get annoyed by large sections of games being quoted in 'double-column' format without notes viz.

  1. d4 Nf6

  2. c4 e6

  3. Nc3 Bb4

  This uses three lines when only one is required. If you look at older opening books (Nunn's 1982 Benoni, for example) there are many part games and games in block text, squeezing the most chess and instruction on the page. No longer.

  (c) question-begging: books allow important questions about particular variations to be ignored, and only light notes are added to games

  (d) lack of originality/ no value added: analysis and comments by others are quoted freely, games are cited without their significance being explained.

  These were humorously put in the article, but clearly felt keenly - my guess being that Phil has more than once spent the best part of 15 quid on various selections from the Batsogan list.

  Now, with the scheme below, annotators can be classified and titles awarded according to their real merits. Find an annotated game, and award points for the annotator according to the following scheme:

  Dismal padding: -3 points

  Glib observation: -2 points

  Unexplained symbols: -1 point

  Glib explanation: 0 points

  Genuine explanation: 1 point

  Crisp, deep commentary: 2 points

  Arguing for choice with evidence: 3 points

  Secrets of chess revealed: 4 points

  Whimsy or humour: extra 1/2 point

  This might be clearer if we look at a few examples:

Dismal padding: -3 points

Gary Lane commits one of these in his book on the Blackmar-Diemer gambit. Chapter 1 is devoted to the Euwe Defence:

  1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 exf3 5.Nxf3 e6

  Now after 6.Bg5 Be7 7.Bd3 Nbd7 he criticises 8. Qe2 as probably time-wasting (as Qe2-f2 may follow), but still gives three White wins as examples: firstly,

  1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 exf3 5.Nxf3 e6 6.Bg5 Be7 7.Bd3 Nbd7 8.Qe2 b6 9.Ne5 Nxe5 10.dxe5 Nd5 11.Bxe7 Qxe7 12.0-0-0 Nxc3 13.bxc3 Qa3+ 14.Kb1 Qxc3 15.Qe4 Rb8 16.Bb5+ Ke7 17.Rhf1 g5 18.Rd3 Qc5 19.Qf3 1-0 (Diemer Emil - Buis W, Haarlem, 1952)

  and after 8...0-0 9.0-0:

  9...0-0 10.Kh1 a5 11.Rae1 a4 12.Ne5 Ra5?! 13.Rf4 g6 14.Bh6 Nh5 15.Rf3 Ng7 16.Bxg7 Kxg7 17.Rxf7+! Rxf7 18.Nxf7 Kxf7 19.Qxe6+ Kg7 20.Qxe7+ Qxe7 21.Rxe7+ Kh6 22.Ne4 Rd5 23.c3 b5 24.g4 g5 25.Nd6 1-0 (Falkeid - Vallestad, Nordic Team Ch.1965)

  9...Re8 10.Qf2 Nf8 11.Qh4 Ng6 12.Bxg6 fxg6 13.Ne5 Nd5 14.Ne4 Nf6 15.c3 Nxe4 16.Bxe7 Qxe7 17.Qxe4 c6 18.Rf7 Qd8 19.Raf1 Qd5 20.Nxg6! 1-0 (Diemer-Locher, Corr 1950)

  In the main line (where White gains a tempo) he gives a critical game which eventually turned out well for Black.

  1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 exf3 5.Nxf3 e6 6.Bd3 Nbd7 7.0-0 Be7 8.Bg5 0-0 9.Qe1 Re8 10.Qh4 Nf8 11.Ne5 Ng6 12.Bxg6 fxg6 13.Rf3 Nd5 14.Ne4 Bxg5 15.Nxg5 h6 16.Nef7 Qe7 17.Raf1 e5 18.dxe5 Be6 19.Qg3 Rf8 20.Nxe6 Qxe6 21.c4 Ne7 22.Qf2 Nf5 23.g4 Rxf7 24.gxf5 gxf5 25.Rxf5 Raf8 26.Rxf7 Rxf7 27.Qe2 Qg6+ 28.Kh1 Rxf1+ 29.Qxf1 Qe4+ -/+ (Diebert-Fishbein, World Open, 1986: ).

  And that's it! Is this eighth move position some bizarre Zugzwang, so that by triangulating White wins, even if Black adopts the critical plan? Of course not, but the work needed here is all left to you. If 8. Qe2 is time-wasting it should have been noted as such and the three games should have been left out. If these games have important or original attacking ideas then perhaps we should be given them, but also be told what to look for.

  ObSheesh: sheesh.

  [The Blackmar-Diemer Gambit, LANE]

Glib observation: -2 points

The Pride and Sorrow of British Chess, Ray Keene, published one of his own games (against Vella) which opened:

  1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. e3 c5 5. Bd3 Nc6 6. a3 Bxc3 7. bxc3 d6 8. Ne2 e5 9. Ng3.



"A theoretical novelty" says Keene, but the main line is not suggested. He adds that the move offers "a pawn for open lines", which is a shallow comment in the 'with compensation' class.

  Whatever thought may have gone behind this move, we are offered none of it. [Winning with the Nimzo-Indian, KEENE]

Unexplained symbols: -1 point

Keene in his notes to the same game cites a Plaskett encounter with Psakhis which continued instead 7...e5 8. Ne2 e4!?. In fact Keene quotes the whole of Psakhis-Plaskett 1985, without a single real comment. The "!?" mark is actually Keene's, and two further moves during the game are annotated "?!" and "!", but no support is offered for these judgements.

  We can see the same thing at work in the three games with 8. Qe2 in the Blackmar-Diemer above. "12...Ra5?!" doesn't really cut the mustard: what was the standing of this game before ...Ra5? How does Black stand after - is he really busted? What would have been a better move? How would Black stand then? How should Black play in this sort of position - what would be the best defensive plan?

  "...In some places words have been replaced by symbols which, like amulets from a witch's bag, have the power to consume the living spirit of chess...

  "...Oh, those exclamation points! How they erode the innocent soul of the amateur, removing all hope of allowing him to examine another player's ideas critically!"

  -- Tigran PETROSIAN

  [Winning with the Nimzo-Indian, KEENE;

  The Blackmar-Diemer Gambit, LANE;

  How to play the Opening in Chess; ed. EVANS]

Glib explanation: 0 points

(see diagram above)

  9...g6 ("to prevent Nf5...")

  Well, we guessed that. But was it compulsory? Desirable? Unnecessary? Weakening? Solid? Double-edged?

  If you find yourself lapsing into this sort of thing, it's no better than football commentary: "White moves the pieces to the King's-side, Black sets an offside trap, White shoots... and scores! - no, Black's cleared it off the back rank!"

  [Winning with the Nimzo-Indian, KEENE]

Genuine explanation: 1 point

The reason RDK is both the pride and the sorrow is that he can do it when he wants to. His note in the same book to move 8 of Portisch-Fischer (p.113) is as good as anything I've read elsewhere on this line, and claims to improve significantly on Fischer's analysis.

  [KEENE, Winning with the Nimzo-Indian]

Crisp, deep commentary: 2 points

It should be no secret to readers that John Nunn is one of the world's finest chess authors. He is noted for a sharp style and deep analysis of complications, but he can also be illuminating about strategy:

Nunn - Gelfand, Munich, 1991


17. Ng3

  "As a young player I was puzzled by games in which White played the manoeuvre Nc2-e2-g3 against the Pirc. It seemed to me that the Knight was not very well placed on g3, because Black's g6 pawn prevented the Knight advancing. Indeed its one and only duty seemed to be to defend the pawn on e4. Then, in 1984, I lost a game with Black against Murray Chandler, in which he used precisely this manoeuvre. The crucial distinction is whether White is attacking or defending. If White doesn't hold the initiative then the Knight on g3 is truly inactive, but if White holds the initiative and has pressure in the centre then the Knight can be very useful. The Rooks and Queen operating on the open files are so dangerous for Black that he cannot counter them directly; instead, Black must somehow aim for counterplay. The only weakness in White's position is the vulnerable Pawn on e4. If this is secure, then White has plenty of time to improve his position. The function of the g3 Knight is precisely to support the e4 Pawn and give White the freedom of action he needs to step up the pressure.

  "In the current position, Black is attacking e4 twice, but at the moment he cannot consider taking the Pawn because of the exposed Knight on d7."

  [John Nunn's Best Games, NUNN]

Arguing for a given choice with evidence: 3 points

For club players, reading what a GM says can often be persuasive without it really standing up to proper standards for argument. Some annotators like Larsen and Timman like to tease readers with their opinions: "I still don't regard [the King's Indian] as a fully-fledged opening" says Timman, annotating a game where he played the system as Black!. This is entertaining and may prompt the reader to think more deeply about the opening, but a properly based conclusion should be based on reasons. What alternatives are there, and do they stand scrutiny? Perhaps only when GMs disagree are we in with a chance of seeing this proper level of commentary. I remember being engaged by a comment in a Hodgson/Day book on the Grand Prix attck, where they said "Hodgson says White is better, because of X, Day thinks Black is better because of Y". It gave me the idea that here is a situation where these factors X and Y might be in rough balance, and might help me judge them better next time.


For these reasons I have long been fascinated by a book in which two annotators wrote separate notes to the games of the 1972 Spassky-Fischer match. Here's an example:



15...P-QN4 "?"

  "Dr.Euwe thought that this was a deliberate sacrifice, since Black does get some compensation for the Exchange. Gligoric disagreed, contending that only White has any winning chances now.

  "The main question is how well Black stands after 15...R-B2, recommendation by Euwe and Najdorf. 16 B-R7 R-R1, 17 Q-K3! N-Q2, 18 N-Q4 improves the White game, delaying ...P-QN4 and aiming to get rid of the fianchettoed bishop. 17...N-N5, 18 Q-N6 QxQ, 19 RxB QxR, 20 QxR gives White such an enormous advantage that it has to be classed as a forced win (20...RxB, 21 Q-N8ch wins).

  "So, in response to the immortal chess question of the late Oscar Tenner -- "Did he fall, or was he pushed?" -- I tend to favour the latter alternative. The Maroczy Bind is not without teeth."

  Everything one might want: a proper consideration of the alternatives, a focus on the critical issue, and a reasoned conclusion. Perhaps not every position is as amenable to such a definitive analysis, but it is a goal to be aimed for.

  [Timman's Selected Games, TIMMAN;

  Both Sides of the Chessboard, BYRNE and NEI]

Secrets of chess revealed: 4 points

Every so often someone will write a comment that enables me to think about not just that position, but actually improves the way I can think about all positions.

Najdorf-Petrosian, Zurich 1953


Now Black tried 12...c5; Bronstein comments:

  "A serious positional mistake, whose significance the reader can realise from the following considerations: Black has an open file on the Queen's side and can provoke b2-b3, whereupon his plan must be to attack White's b3 by advancing his a-pawn. This attack had prospects of succeeding if Black can support his pawn when it gets to a4. What is he to support it with? He has no white-squared bishop, and the square c5 has beeen taken away from his knight by his last move. Moreover, it is clear that Black will not be able to maintain his knight on d4 forever. Hence he is now left without any plan that has prospects of being successful."

  Looking back, I can now recognise how sometimes a plausible move leaves my position without a way forward, and I don't think I'd be as alert to the problem as if I hadn't read that comment. Perhaps it's an insight that can be obtained elsewhere, but many people say they have had that "aha!" experience reading Bronstein.

  [The Chess Struggle in Practice, BRONSTEIN]


Whimsy or humour: extra 1/2 point

My introduction to chess, as so many, was through Irving Chernev: as much as the magic of the game I was attracted to Chernev's delightful whimsy. Another author I like is Stephan Gerzadowicz: who could prefer the pedestrian "White takes yet another Pawn" to Stephan's "They're like peanuts, you know"?

  [Logical Chess, CHERNEV;

  Thinker's Chess, GERZADOWICZ]


  Having got the points system we can start to see a few patterns emerge. For a game of average length in a book or magazine, we can classify annotators as follows:

0-10 points: hack work - club standard annotator

Your lower-graded club colleagues may be entertained or even assisted by your efforts, but there are enough chess books in the world without you adding to the pile. Improve or give up.

10-20 points: good effort - candidate master annotator

It is clear from your score that you can understand chess and can communicate your understanding, and this is the heart of annotating.

  There is probably another level to be reached: have you considered all important alternatives? have you tried to explain how the general strategical tensions are reflected in the key variations? have you explained how the perceptions of this type of game have changed over time?

20-30 points: excellent - master annotator

You may be justly proud of your efforts in annotation. It is likely that there is little more to be said about a game when you have written about it. You have noted all the important choices and explained the players' motivations at each point in the game.

30-40 points: grandmaster annotator

Here we have a standard of annotation which even masters will find illuminating. Of course, not all moves of all games will be able to support the weight of commentary you are able to offer, but among the notes you offer will be gems which will stay with the reader for the rest of their life. Tsar Nicholas nominated the first five "Grandmasters of chess"; may I indulge myself by nominating my first five "Grandmasters of annotation". Take a bow, Mikhail Botvinnik, David Bronstein, Mark Dvoretsky, Bent Larsen, and John Nunn. Honourable mention: Peter Clarke.

In a class of its own: Secrets of Grandmaster Play by Nunn and Griffiths.

My points scheme fell apart here. This is the gold standard of annotation: the games are analysed in all their tactical and strategical complexity (some positions being given more than two pages of notes), there is due regard given to the realities of tournament play (for example, how much did Nunn really see in the game?), the background to the game is given consideration, and there is a genuine attempt to educate. New revised edition published 1997. A must-read, but take a deep breath!



1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5

  "This is White's only chance of gaining any advantage against the Grunfeld. Much weaker, for example, is 4.Nf3 Bg7 5.e3 O-O 6.Be2? (a genuine beginner's move).  So far, Petrosian-Botvinnik, Game 5, 1963. Now Botvinnik played, 6...cxd4? 7.Bxc4 c5? (Better is 7...Nc6 and then 8...e5) 8.d5! e6? (Better is 8...Ne8) 9.dxe6 Qxd1+ 10.Kxd1 Bxe6 11.Bxe6 fxe6 and, despite all of Black's lemons, the game is still only slightly better for White, which proves again the weakness of White's first move, 1.d4. (...)" Bobby Fischer, 1963.

  Purdy commented:

  "We like especially the opening paragraph of Fischer's annotation to 4.cxd5.  It is matchless in the literature of chess."

Automatic Game Annotations

[A service to the internet chess community provided by [Cool
Blue Cat] Cool Blue Cat Marketing]

  Do you have a puzzling game? Submit the moves one by one to the Oracle who will annotate the game for you!


Select the options describing the move.

Then, activate the SUBMIT RESPONSE at the end.

  • Colour: White | Black
  • Piece: K | Q | R | B | N | P
  • Move or capture?: Move | Capture
  • File: a | b | c | d | e | f | g | h
  • Rank: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8
  • Captured piece: (if any) [None] | Q | R | B | N | P
  • Check or mate?: [Neither] | Check | Mate

      Time spent thinking?:

  • Minutes or seconds? Mins. | Secs.
  • 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 10-20 | 21 or more
  • Your assessment of the situation before the move? = | += | +/- | +- | & | =+ | -/+ | -+
To submit your responses, click on this button:

 To clear the form, press this button: