How do I self-analyze my games? Advice needed.

Greetings all. I'm sort of newish to chess (played as a teen, stopped for 49 years, just started again a couple of months ago (thanks Queen's Gambit) and have been reading a lot and playing some games). My daily rating is in the 1300's (rapid is in the low 1100's -- I just can't think that fast). OK, with that intro, here's my question.

Stjepien (aka Hanging Pawns) makes the following suggestion for analyzing one's own games.

1) manual dissection and write up of your mistakes
2) look up Grandmaster games of the same positions to see what moves were played there and try to understand why. Write everything down
3) Run the computer analysis to see what you missed. Write it all down.

I note that many people have said similar things, I'm just using his formulation to explain why I have such problems doing this.

1) manual dissection and write up of your mistakes

that's great, but -- particularly in a game that I won -- I rarely see my mistakes! And even when the engine says "move 19 was an error" (the key moments thing is pretty cool), I most often *still* can't see a better move! This is particularly true regarding a tactic missed or a positional error. If I'm going over a 45-move game, am I supposed to look for a tactic I missed for every move? (After all I missed it the first time!) Granted, some mistakes are easy to spot, because I got into trouble. But in games where I win but missed tactics, or made a positional error (such as pushing a pawn) I just don't see my mistakes unless someone says: you messed up on move 21, find a better move.

2) look up GM games of the same positions to see what moves were played there and try to understand why. Write everything down

At my level, often, by move 6 or 7, it's a unique game. No GM has ever played it. So where do I go from here? (Granted, this is useful for the move where I went "off book" and to try to understand why. But by move 10, this advice is useless).

3) Run the computer analysis to see what you missed. Write it all down.

and here is where I discover a ton of mistakes I made.  Tactics missed, bad pawn pushes, putting a piece on a sub-optimal square, etc.

Thoughts?  Advice?  I know I'm not supposed to rely on the engine.  I'm trying not to rely on the engine.  But -- again, particularly for tactics missed and for positional errors in a game where I have won -- I'm just unable to see my mistakes.

Keep it simple.  Write down your thoughts, ideas, plans, what you think your opponent was trying to do, etc.  ONLY use an engine to check for blunders, and missed tactics.  FORGET what or how GM's analyze. 

When doing analysis in the opening phase of the game,  use this:

Opening Principles:

1. Control the center squares – d4-e4-d5-e5
2. Develop your minor pieces toward the center – piece activity is the key
3. Castle
4. Connect your rooks

Tactics...tactics...tactics...

The objective of development is about improving the value of your pieces by increasing the importance of their roles. Well-developed pieces have more fire-power than undeveloped pieces and they do more in helping you gain control.

Now we will look at 5 practical things you can do to help you achieve your development objective.

They are:

1. Give priority to your least active pieces.

• Which piece needs to be developed (which piece is the least active)
• Where should it go (where can its role be maximized)

2. Exchange your least active pieces for your opponent’s active pieces.
3. Restrict the development of your opponent’s pieces.
4. Neutralize your opponent’s best piece.
5. Secure strong squares for your pieces.

Don’t help your opponent develop.

There are 2 common mistakes whereby you will simply be helping your opponent to develop:

1. Making a weak threat that can easily be blocked
2. Making an exchange that helps your opponent to develop a piece

Before making a move, do the following:

Pre Move Checklist:

1. Make sure all your pieces are safe.
2. Look for forcing moves: Checks, captures, threats. You want to look at ALL forcing moves (even the bad ones) as this will force you look at, and see the entire board.
3. If there are no forcing moves, you then want to remove any of your opponent’s pieces from your side of the board.
4. If your opponent doesn’t have any of his pieces on your side of the board, then you want to improve the position of your least active piece.
5. After each move by your opponent, ask yourself: "What is my opponent trying to do?"

As for anything past the opening?  That is when the game begins to get difficult.  After doing all this legwork yourself.  Now run the game through an engine.  DO NOT get caught up in all this accuracy, caps score, brilliant moves crap.  It does nothing to improve your game. 

Just some good old fashioned hard work will do the trick.

@IMBacon -- thank you so much for putting together such an awesome document.  I hope many people read it.  

That said, much (not all) of the above I was aware of (seeing it all in one place, however, is great for me). 

What you said makes total sense -- including the part that your advice was mostly about the opening.  And, indeed, I'm referring to the middle game.

Perhaps an example of what I'm trying to articulate might be worthwhile here.  (This is a game-analysis forum, after all).  Here's a very simple game I played and won.

I hope the above (not the best example, but I hope you catch my drift) helps.  All the opening guidelines were followed.  I developed pieces, accumulated marginal advantages, calculated to win a pawn twice, etc.  Self-analysis for this -- before I got the engine involved -- added pretty much nothing to what I previously thought of the game.

Am I making sense here?

https://www.chess.com/forum/view/game-analysis/how-do-you-convert-this-3-5-position-i-ended-up-in-a-losing-position-61271761?page=4#comment-61316455

This is a good example of what you want to achieve

Hanging Pawns on YouTube has a few (maybe 1 or 2) more videos on the subject.

https://youtube.com/playlist?list=PLssNbVBYrGcD2mB7JrHbpP5qyT_ncxCRj

I get it, you don't know what you're looking for.

The engines don't explain it, they give moves used in the calculations.

Step 1- most important is to capture exactly what you think during the game. OTB, you have to wait until after the game. Online, you voice record and talk out loud, or play daily and keep notes...lots of notes.

Do your thing to find mistakes, etc after you annotate your game with notes.

Then you can look to the engines and more importantly other people for help.

Thank you both for enlightening words.

@PerpetuallyPinned -- you summed that up perfectly "I don't know what I'm looking for" !!  Thanks for articulating.

@IMBacon -- thanks for responding so quickly and looking into my game (picking a game that I won and didn't make serious errors is not the best example).  Today's a tight schedule for me, so I didn't find the best game to explain what I'm feeling -- but I think you and Pinned are getting it. 

FWIW, the engine does like 20 Na5 (what?!).  As for 21 Nxf7, you suggested Be4.  The engine likes that, and also likes Be2, and likes 21 Rc1 the best.  I inserted those lines into the game and repost it here.  But . . . alas . . . if I were doing any self-analyzing of the game, since I wouldn't know what to look for, I would not have seen it, except with the help of the engine.

Show me, don't tell me...that the engine likes it

Ha!  Yes, I've been told that more than once.

Thanks!

"But . . . alas . . . if I were doing any self-analyzing of the game, since I wouldn't know what to look for, I would not have seen it, except with the help of the engine."

And unless you understand what the engine is showing you?  Engines are worthless for anything except finding blunders and missed tactics. 

Think of it as cheating on a test in school.  Sure you have the answers, and you will get the questions right.  But you are not going to understand "why".

Take this game you lost.

Keep your analysis simple.  NO one is expecting GM level analysis. 

For each move ask yourself the following:

Does my move following opening principles?

Am i leaving any material undefended?

Has my opponent left any material undefended?

Have i double checked each move?

What is my least active piece, and ho can i make it more active?

Do I gain any type of advantage with this trade?

What is my opponent trying to do?

What am i trying to do?

If you dont know how to answer any of these questions?  Welcome to the wonderfully complicated world of chess.    But the best part is that you have a few million people willing to help.

Aww -- gee -- take *any* other game of mine!  I was tired, plus I stink a rapid chess (I can't think that fast -- that's why my daily rating is 250+ points higher than my rapid!)

Your first lesson. 

Stop with the excuses.  I didn't choose the game to pick on you.  I chose it because it was short and quick.  Every time i pick someones game, its always the same excuses.  Every game is a learning experience...especially the losses.

Thank you.  I want to comment on your first and last comments:

1.  When I read annotated games (which we are encouraged to do) we get GM-level analysis.  And I think that colored my view of what *I'm* supposed to be doing.  Both of you are, essentially, telling me to "chill out" about that.

A ran into some advice from Dan Heisman in the last hour, and he said essentially the same thing.  He said: "look at the engine evaluation changes.  If you opponent makes a move and it greatly favors you, that means  you probably have a tactic there.  If it goes back down after that move, that means you probably missed a tactic."  And etc.

This makes a lot of sense and is a sense of relief.  What you all are telling me is that when I read in the books that some random pawn move is great because it shores up the pawn structure (or something like that) -- that I shouldn't expect to find that (at my level) when I'm analyzing my own games.   But the above is at least a way to tell me *where* I may have missed a tactic.

(BTW, Heisman also has -- as his first suggestion -- to go over the game with your opponent.  Because he knows some stuff about the game that nobody else in the world knows)

As for your last comment -- I am thrilled to be part of a community where so many people are wiling to help others.  

Thank you both very much.  I really appreciate the time and effort you put into answering -- and you should both know that I will endeavor to put your advice into practice.  (In other words, your time and effort will have helped at least one player improve!  I hope others are reading this and are helped by it, too).

Thanks so much!

Well, most of those excuses are true!

But that is irrelevant to your main point: every game is indeed a learning experience.  I certainly learned from that game!

As for specifics: yes, you're correct, I have an unhealthy fear of doubled pawns -- too much.  

And, as it turns out I just read yesterday this is related to one of "Lasker's rules", which you (intentionally or not) elaborated on.  Which is that if you're going to do the Bishop-takes-pinned knight thing to double the pawns of your opponent -- don't do it before they castle!  Because -- as you pointed out: you get double pawns here on a strong file (c) and open up the b-file for the rook -- and you can just castle on the other side if you need.

So, yes, I did seriously mess up that game late at night and I stink at rapid, etc etc etc -- but you're right that there was still a lot to learn from it.  (But I do invite you to check out any other game of mine! )

Here's a game where I was offense, and I initiated just such a trade -- and those who are a lot smarter than me (including a coach) told me that BxN (move six) was a bad move.  And, indeed, I almost got nailed, in part, because he aimed his rook at my king on the b-file, and he simply castled long.  (I won because he messed up later -- I was kind of lucky)  Specifically, I am talking about move 9 Bxf3 -- not a good move.  I'm not finished going through the game, but here are my notes on it.  (I'm not posting it because it's a win -- I was lucky! -- I'm posting it because it supports your point of when BxN is a poor choice, and how it got me into trouble)

Thanks!

1. Absolutely forget about GM analysis.  Its way above our heads, especially modern GM's who's games are so engine driven.  Do what you are capable of at your level.  Learn ad grow from that.  Chess like life is baby steps. 

2.  The ONLY thing you should be using an engine for is missed tactics, and blunders.  getting caught up in all this accuracy scores, why wasn't my move brilliant?, percentages, and how many moves deep you need to know openings is nothing but a waste of your valuable study time. 

3.  When you look for tactics.  You want to look for Forcing Moves: Checks, Captures, Threats.  tactics will always be forcing lines, which means they are easier to calculate.  But as with anything chess.  Start at the beginning.  Start with 1 move mates, and work your way up.

4. Don't concentrate on pawn moves.  As lowly as the pawn is.  They decide the flow of the game.  With every pawn move, you gain something and you give up something.  Their is a reason entire books are written about nothing but the pawn.

5.  Anytime you can, yes always go over the game with your opponent.  Bounce ideas off of each other, and learn from each other.  Unfortunately here unlike in OTB play.  People rarely want to go over games.  But if you would like to play an unrated game of daily chess?  I would be happy to, and we can go over it after. 

Bacon's doing all the work, no need to thank me.

I'd like to expand on checks/captures/threats...

Threats- Checks & Captures (Pressure & Tension)

We know a check 

Captures-

Pressure- a one way threat to capture (a bishop threatens a knight or pawn from a distance, but they can't capture it) en passent is a 1 way capture

Tension- a mutual threat to capture (2 pawns/bishops same color/etc)

This will help when looking for threats (helped me anyway)

Agree with much of what is posted here. 

However disagree with going over your game with your opponent.  At your level he will not know enough to help you and probably will hurt you.  Always have a much stronger player than you look at your games.