11

This has been really frustrating for me since I don't get it. When I go through my games I spot errors and simply find better moves but then that's it. I just don't understand how this improves your game, all I did was find mistakes and correct them in a game I've already played. People keep telling me this is the best way to improve but I can't seem to get "it". If anything I suppose going through your games can show you where you might be lacking but wouldn't simply just continue learning and training in all areas of chess be a far more efficient way to improve instead of wasting time analyzing? I would appreciate examples of how analysis helped you become a better player.

  • 5
    Just spotting mistakes and doing nothing more may not be very efficient. Try to come up with the reason WHY you made the mistake. It's one of the few ways to realize your weaknesses in your game, if you're honest with yourself when analyzing. Then it's easier to know what needs to be improved. – Scounged May 16 '16 at 9:19
  • 2
    What if I'm not sure why? I've went through games where I found there were better moves than the one played but I come to a dead end when trying to find out why this happened instead. – user10300 May 16 '16 at 9:36
  • 4
    This is indeed sometimes very difficult to figure out. However, usually it has to do with your thought processes during the game. This is why you should as quickly as possible after the game write down your thoughts during the game, as detailed as you can remember. This way it's easier to determine if your mistakes may come from a simple oversight, or from for example overconfidence. I will give an example of my own analysis of one of my games below after I'm done with school for the day., – Scounged May 16 '16 at 10:34
10

First of all you have to distinguish which kind of error you made.

Opening error:

You should definitely study opening theory to avoid opening traps or get bad positions right out of the opening. On the other side when an opponent plays an unusual/bad move, you need to know how to respond and punish this move to get a better position.

Middlegame error:

Have you overseen some tactical possibilities for you as attacker/defender?

-> Try to solve tactical puzzles to get familiar with certain positions. This will also help you to spot tactics more easily the next time they occur.

Have you had the wrong plan?

-> Study positional topics like pawn formations, weak/strong points, etc. in a position and you will find the right plan straightaway.

Endgame error:

When you have no clue how to play certain theoretical endgames you will easily take a wrong step and lose drawish positions or draw winning positions.

-> Take a good endgame book to study some basics.

Also learning from mistakes you make, means you should not make the same mistake again. Thats especially true for opening errors and endgame errors in easy theoretical endgames.

  • 1
    Excuse me if I misunderstand but it seems that the solution for any errors you find in your own games from analyzing is simply just studying the game more. Doesn't this prove the OP's statement, "wouldn't simply just continue learning and training in all areas of chess be a far more efficient way to improve instead of wasting time analyzing?" Not trying to have go at you but your post doesn't seem to tell me the value of analysis. – Krane May 19 '16 at 9:02
  • 2
    @Krane "instead of wasting time analyzing?" in my eye it's not a waste of time to spot your mistakes, because once you find your mistakes, you will know in which area you have to study more deeply and leave your comfortzone. Otherwise you will study for example the whole opening over and over, but that doesn't help you when you always make a mistake from transposing into the middlegame and choose the wrong plan(which you might not recognize if you don't analyse). In my opinion analyzing one of your own games always leads to improving in all kinds of areas in chess – Don May 19 '16 at 18:03
4

I will give an example of some of the preliminary analysis in one of my most recent tournament games below, and give a concrete example of what I could take from that analysis.

[Date "2016"]
[White "My opponent"]
[Black "Me"]

[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w - - 0 1"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e5 6. Ndb5 d6 7. Bg5 a6 8. Bxf6 gxf6 9. Na3 b5 10. Nd5 f5 11. Bd3 {Here I chose to include some lines illustrating what happens after 11.Bxb5. This information can be found in any reference database, however, and will not be included here} Be6 12. c3 Bg7 13. Qh5 {"A very aggressive move, which really put me on the lookout for possible dangers. I knew that up until this point in the game, we had been following theory, so I was not worried that I had stepped into a trap by mistake."} f4 14. g3 {"I think I did the right thing to challenge the white knight on d5 here. The idea is to let White take on f4, and then use the bishop pair to play the position. Black puts his dark squared bishop on e5 and plays on from there."} Ne7 15. Qf3 Ng6 16. Nc2 Bxd5 {"A decision that had to be made before white can support his d5 knight with another light piece. A light piece on d5 is generally much better for white than a pawn on d5. I did not want to trade away my pair of bishops in the position, but even less did I want to see a white knight safely anchored on d5 for the forseeable future."} 17. exd5 O-O 18. Bf5 Qg5 19. Bd7 e4 20. Qg4 Qxd5 21. Nb4 Qc4 22. O-O-O Ne5 23. Qf5 Nxd7 {"Here I thought I had seen all the way to the end of the sequence that is to come. The move is sound, but black has other ways to win in this position as well."} 24. Qxd7 Bxc3 25. bxc3 Qxc3+  26. Nc2 Rac8 27. Qg4+ Kh8 28. Qe2 f3?? {"A blunder, which came very quickly from me. I thought that the game was completely over after the white queen would be forced away from e2. I completely forgot that a queen on e4 still protects c2. I should have kept my cool throughout the entire combination, so that I wouldn't have moved through the whole line on reflex. With this, black has thrown away the winning edge. Eventually the game ended in a draw."} (28... Qa1+! 29. Kd2 Rxc2+! 30. Kxc2 Rc8+ {"wins on the spot."}) 

This is a preliminary analysis made of the game, without the use of anything else than a reference database in the opening. Note how few variations I have included in my analysis at this stage. What is there is basically just my impressions of the game soon after the game was finished. Instead I have focused on trying to remember my thought process during the game, and write it down in words.

So, when doing my analysis in this fashion, what can I learn from my game and my own mistake?

In this example it becomes very clear: I realized that I played too quickly in a critical stage of the game, when I thought I had seen all the way to the end. I became too excited. Thus, it will be easier for me to recognize this happening in future games, and I will have a chance to make myself slow down a bit.

This particular lesson will not be available for me to read in any book. I will only be able to learn it through experience if I try to remember as accurately as possible what happened during a game where something went wrong.

This is the true reason why analysis of your own games is so good for self-improvement. Well-done analysis will reveal things about your game that you cannot find very easily elsewhere. It is important to analyze your games yourself, without the help of computers in the preliminary stages. This is to ensure that you think back on your experience during the game, without being distracted by the computer lines.

Computers should be used as a final touch, when you feel like you cannot think of anything else that was noteworhty in the game, to check how sound your analysis is. Do not remove your original analysis! Keep it, and note where the computer suggests improvements.

Edit: The reference database is used in the opening to quickly reveal opening mistakes. It is not a bad idea to include references to master games where the master played a better move than the move played in the game. It has the primary purpose of finding out how to play the opening properly, if you don't already know that.

  • It ended in a draw? C'mon black... – Mr Pie Jul 13 '18 at 9:46
  • @user477343 Yes, it ended in a draw some twenty moves later. I realized that I had made a mistake immediately when my opponent took on e4 at move 29, and despite trying with all my might I found no win. In hindsight I know this is because the win is already gone, but during the game it became very frustrating. In the game I took some serious risks in order to still win, and I was lucky to escape with a draw. – Scounged Jul 13 '18 at 11:36
  • What I am saying is that a better move for black as opposed to 28. f3 was Queen going to a1 check. King moves to d2 (because knight is pinned), then rook takes knight on c2 check, being a sacrifice. King takes back (otherwise free queen for black). Then wham! Qxa2+ and black wins white's queen. That is what black shoulda done. You were playing black, right? – Mr Pie Jul 13 '18 at 12:10
  • 1
    @user477343 Yes I was playing black. Did you by any chance notice the subvariation that I included in this post? – Scounged Jul 13 '18 at 13:05
  • 1
    hahahah, no, but that is definitely hilarious... I mean, great minds think alike. – Mr Pie Jul 13 '18 at 13:06
2

In a way your question seems to be about learning skills in a very general sense, not limited to chess.

You have to find a way to generalise at least a bit from the improvements you find and then work on making those improvements become habits. As you say it's no good just working out "oh in this exact position it would be better to take with the pawn".

For example from my own experience (disclaimer: I'm a weak player and don't analyse enough!) I have found that I:

  • don't see mating patterns unless they're pretty basic (eg corridor mate)
  • lose too often by blundering pieces when otherwise in a winning position

So now I practise mates and try to do an extra "safety check" for hanging pieces in any given position, and I'm improving.

In short you're right, the analysis of a game is just the start of a process that goes something like:

Find error > Identify most common errors > Make a plan to address those errors in future.

Hope this isn't too basic!

2

To be able to improve, you need to isolate:

  • The move that was a mistake
  • Why the move was a mistake
  • Why you made that move in particular
  • Why you didn't make (an)other, better move(s)

To get better at analyzing, you want to be able to spot the weak move by yourself. Initially, you may not be able to do this very well. But try it: Go backward through the game from the end, and try to isolate the last position where you felt you were either equal or better. Then, retrace your steps and try to identify the offending bad move that happened after that.

Looking at that move, figure out what your opponent did that exploited that move (this assumes you're working on a move that your opponent punished, and not one that he missed). Then, try to find an alternative move you could have made that foils his idea. It can do this in several ways:

  • Avoid the structure / situation he exploited
  • Make a more compelling threat, so he doesn't have time to exploit it
  • Prepare the move with another move first, so the weakness is addressed before you play it

Now, try to figure out why you played that bad move in particular:

  1. Did you play the move reflexively, almost without thinking?
  2. Did you expect a different outcome?
  3. Did you fail to anticipate his reply?
  4. Did you anticipate the reply, but underestimate its impact?

Each of these behaviors has multiple possible explanations, and you need to isolate the ones that were in effect.

  1. If you didn't think hard enough:

    • You need to re-examine the position after every reply. You may see something you didn't notice before, even if you anticipated his reply. You should follow the discipline of, at a minimum, examining every check, capture or threat (CCT) that your opponent might make on his next move if you did nothing, and every check, capture or threat you can make immediately. This is particularly important if the reply was unexpected.

    • You must also avoid reacting. If the opponent's reply was unexpected and poses an immediate threat, there's a natural tendency to react quickly, out of fear. You must avoid this. If he captured one of your pieces unexpectedly, don't immediately recapture. Look at all the possibly useful replies.

If you still make this mistake, try setting a minimum think time. For example, in a 30-minute game, make sure you spend at least 30 seconds selecting your move every time, unless it's rote opening theory.

  • Or, you may need to curb your appetite. Some players make speculative sacrifices that are easily refuted. This is typical of many players under 1600, but some persist even when their ratings somehow get higher. If you notice that you are consistently capturing pieces with negative results, try to understand your motivation, and moderate it. Once you force yourself to recognize that winning is more exciting than sacrificing and then losing, you'll kick the habit.

  • Conversely, you may need to man up. Some players play defensively, even with White, as a matter of preference. It is 4-10 times harder to defend than it is to attack, because the attacker (unless he sacrifices) has multiple options, each of which may offer substantial benefits and incur only minor costs, while the defender frequently has only one move that will work, while all others lose. So, the defender has to think harder. He will therefore get tired first. He also is likely to have the poorer morale, and as a result, the weaker fighting spirit.

If you want to avoid these disadvantages, you must follow Steinitz' dictum that you must (not may) launch an initiative if and only if you have the advantage.

  1. If you anticipated his reply, but expected a different outcome, then you either misevaluated the resulting position, or failed to find or anticipate a strong move later in the game that your move made possible.

An example of the first case might be trading your better bishop for his worse knight, leaving yourse;f with a drawn rook endgame you thought was winning. If so, you need to learn more about minor piece endgames, and about rook endgames.

An example of the second might be your not noticing that 3 moves later, a recapture by your opponent puts you in check, halting your combination abruptly. If that's the case, you need to do tactical exercises of at least 3 moves' length, and get them right at least 75% of the time.

  1. If you failed to anticipate his reply, you're either not doing the CCT part, or you're not also examining seemingly unlikely or purposeless moves like quiet moves that leave a piece hanging, moves that close a mating net around your king without actually giving check, and moves that prepare a tactic, such as masking a queen, rook, or bishop with another piece that can be moved away with tempo, or aligning a rook with your queen or king.

To address this problem, do exercises where you take a position from a game (it doesn't have to be one of yours), and catalog all of the possible moves with reasons why they are useful or not. Don't reveal the actual move played until you're finished. Then, figure out whether you selected the best move, and for the correct reason. You might find Bruce Pandolfini's "Your Move" column in USCF's Chess Life useful for this kind of training.

  1. If you anticipated your opponent's move but underestimated its impact, try to figure out if you made one of the following mistakes:

    • Did you find at least one justification for his move? If not, get into the habit of doing this. Every move has at least one justification.

    • Did you come up with the wrong reason for his move? If so, why? Was it because you'd never seen the idea before, and didn't recognize it? Or did you just fail to notice it? Or, did you see one justification, when in fact there were several justifications for the move? Gaining more experience will address the first two, but to fix the last one, you have to get into the habit of asking yourself "What else?". Make sure you discover all of the possible explanations for his move, not just one of them.

    • If you found the right reason for his move, but your response was ineffective, was it because:

      • His idea was something you considered impractical, or easy to refute? If so, you need to improve your calculating ability, including finding all of the reasonable candidates, using CCT and possible surprise moves.

      • By the time he made his move, there was no longer any effective defense? In this case, you needed to notice the idea sooner; maybe while analyzing the preceding positions, you didn't spot the final move, because you didn't recognize the threat. Two causes are common for this:

      • In the earlier position, you stopped analyzing before there were no more threats. This is fixed by working harder during analysis, and cataloging all the reasonable moves when visualizing the final position, along with their threat potential, and only halting your analysis when you have concluded that none are significant.

      • You were unfamiliar with this kind of threat, and so couldn't recognize it. The only remedy for this is experience. Based on the type of threat (tactics, endgame, forcing a strategic weakness, etc), find games or exercises with that kind of element and play through them.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.