How to review your games?

Sort:
Oldest
pdve

I find it impossible to review my own games. Even though stronger players strongly suggest to do this.

If I review a game, I would play the same move again, even if it was a losing move.

How do I do this?

kikvors

Pick one position from the game, a position where you was in doubt about what to do, that you're unsure of afterwards. First, try to write down what your thoughts about it were during the game, why you played what you played. Put those notes away.

Next day, set it up on a board. First think about for ~10 minutes, without moving the pieces. If you think of a concrete line (a reason why some move won't work, threats, possible ideas of what to do), write it down. Also look for candidate moves, write all of them down. Try to put in words what is going on in the position -- what's each side trying to do? Is that necessarily the only thing they could do, or are there other things they could try as well?

After the 10 minutes, start moving the pieces. Check your written down lines. Look for new ideas! After 15 more minutes you should have a good idea of what's going on, what works, what doesn't, and whether the move you played was good or not.

Take 5 minutes to check the most tactical lines on a computer, there may have been hidden tricks.

Now take your notes and compare.

Repeat after your next serious game.

If you get good at this, you can start doing it for two positions per game. Then three. There will also be positions that don't need half an hour. Eventually you'll be able to go through a whole game.

I found that I only got better at this after working through a lot of Yusupov positions -- first I had to learn how to look for the best move in a position I had on a board at home, then I could apply it to positions from my own games.

pdve

hmm .. interesting. i will try this.

baddogno

I'm lazy so I usually just throw an engine at it to catch the tactical shots we both missed.  Then maybe I go back over it and see where my plans were flawed.  That BTW is not the recommended way to do it; the pros all say throwing an engine at the game is the last step but I have the same problem you do: OK, I'm looking at my game, now what?  At least if I know where the missed tactical shots are by each side I have a place to start.  I'm not encouraging you to do it my way, grinding it out yourself is undoubtedly a better approach if you're young and ambitious but I'm old and not.

Kikvor's advice seems quite sound compared to mine. Laughing

kikvors

I'm trying to get better at running, so I need to go out and run every couple of days, always going a little further than I could last time, and so on.

Letting an engine analyze your games is like trying to get better at running by driving those practice miles in a car. You get there, and quite quickly as well, but don't be surprised if your running doesn't actually improve a lot.

Rsava

kikvors wrote:

trying to get better at running by driving those practice miles in a car. You get there, and quite quickly as well, but don't be surprised if your running doesn't actually improve a lot.

**********************

So THAT'S why my running isn't improving. ;)

Great advice for analyzing/reviewing games. I think I will give this a shot, it is something I have always struggled with myself.

baddogno

In my own defense I never said use an engine exclusively.  I just use it to find the tactics I missed and then go from there.  Absolutely agree that just using an engine is probably actively counterproductive.  You end up following computer lines that are unnatural and often fly in the face of accepted strategy.  Your approach is certainly better, I was just offering a short cut that works for the old and lazy.  

IpswichMatt
kikvors wrote:

I found that I only got better at this after working through a lot of Yusupov positions -- first I had to learn how to look for the best move in a position I had on a board at home, then I could apply it to positions from my own games.

Hi Kikvors, what do you mean by "Yusupov positions" ?

Scottrf

I presume he means positions from Yusupovs 'Build, Boost, Evolve' series.

IpswichMatt

Thanks Scott!

So are these books really that good then?

Scottrf

It depends what you're looking for. It's very good as a structured review of strengths and weaknesses, and forcing you to think hard about a chess position (which has to be better for learning that just reading). But in terms of actual explanation, it's minimal. There is a bit of text and a few examples of the subject area (forks, centralisation etc), then a number of test positions.

I don't really know your level but it's fairly tough. I would say the tactics positions in there I mostly found around the 1900-2200 level here. Which I didn't particularly struggle with, but the positional chapters I found very tough.

IpswichMatt

When I played I think my last rating was 137 BCF - so it sounds like it's way over my head. Cry

Thanks for the response though

Scottrf

That's about 1700 FIDE? I don't think so at all, I believe they are aimed at 1500. The TT numbers here are a lot higher.

molokombo

the advice yusupov himself gives i relation to reviewing your own games is that you should first focus on finding the turning points of the game - decide when mistakes were made, where the evaluation of the position changed or an opportunity for sharply changing the situation on the board was not exploited.

next he says you must seek the reasons for your mistakes, and then look for new possibilities which you did not notice during the game.

kikvors

Yes I meant positions from those books, but of course that's not necessary.

The point is that the only exercises I used to do were tactics exercises (books, Tactics Trainer, etc). I tried to study the rest (positional play, endgames, openings, etc) as well, but not with exercises. As a result, when I set down to try to analyze some position, I'm extremely focused on tactics and confused when there aren't any.

The Yusupov books treat all those subjects, with lots of exercises. So as a result I have a slightly broader look when I try to analyze. And slowly, it starts to feel useful to just setup some position from some game and analyze it. I really couldn't do it before.

I think starting with half an hour on one position as I wrote is probably too tough, start with less.

MCBeaker

In my OTB games I put a small dot on my scoresheet after any moves that were hard for me to decide on. The intention is that I'll pick these positions to review later Innocent

eridpnc

How do you actually find your moves?

orodz
kikvors wrote:

Pick one position from the game, a position where you was in doubt about what to do, that you're unsure of afterwards. First, try to write down what your thoughts about it were during the game, why you played what you played. Put those notes away.

Next day, set it up on a board. First think about for ~10 minutes, without moving the pieces. If you think of a concrete line (a reason why some move won't work, threats, possible ideas of what to do), write it down. Also look for candidate moves, write all of them down. Try to put in words what is going on in the position -- what's each side trying to do? Is that necessarily the only thing they could do, or are there other things they could try as well?

After the 10 minutes, start moving the pieces. Check your written down lines. Look for new ideas! After 15 more minutes you should have a good idea of what's going on, what works, what doesn't, and whether the move you played was good or not.

Take 5 minutes to check the most tactical lines on a computer, there may have been hidden tricks.

Now take your notes and compare.

Repeat after your next serious game.

If you get good at this, you can start doing it for two positions per game. Then three. There will also be positions that don't need half an hour. Eventually you'll be able to go through a whole game.

I found that I only got better at this after working through a lot of Yusupov positions -- first I had to learn how to look for the best move in a position I had on a board at home, then I could apply it to positions from my own games.

 

Hello. As someone who is just really wanting to seriously learn this game, I just want to thank you for this. Because I was literally just about to sit down and start reviewing my past games, but had absolutely no idea where or how I should start. I am a beginner with all this. I have played before, lots of times, but I only really knew how to push the pieces around. No real tactics or all that stuff I am learning. I'm barely just learning opening moves, and all that beginners stuff. Probably obvious by my score. But again, thank you for this. I really do want to grow in this game and you just gave me a good place to start. Looking forward to improving my decision making. 

kindaspongey

Possibly helpful:
Simple Attacking Plans by Fred Wilson (2012)
https://web.archive.org/web/20140708090402/http://www.chesscafe.com/text/review874.pdf
https://www.newinchess.com/Shop/Images/Pdfs/7192.pdf
Logical Chess: Move by Move by Irving Chernev (1957)
https://web.archive.org/web/20140708104437/http://www.chesscafe.com/text/logichess.pdf
The Most Instructive Games of Chess Ever Played by Irving Chernev (1965)
https://chessbookreviews.wordpress.com/tag/most-instructive-games-of-chess-ever-played/
Winning Chess by Irving Chernev and Fred Reinfeld (1949)
https://web.archive.org/web/20140708093415/http://www.chesscafe.com/text/review919.pdf
Openings for Amateurs by Pete Tamburro (2014)
http://kenilworthian.blogspot.com/2014/05/review-of-pete-tamburros-openings-for.html
https://www.mongoosepress.com/excerpts/OpeningsForAmateurs%20sample.pdf
Discovering Chess Openings by GM John Emms (2006)
https://web.archive.org/web/20140627114655/http://www.chesscafe.com/text/hansen91.pdf
Chess Endgames for Kids by Karsten Müller (2015)
https://chessbookreviews.wordpress.com/tag/chess-endgames-for-kids/
http://www.gambitbooks.com/pdfs/Chess_Endgames_for_Kids.pdf
A Guide to Chess Improvement by Dan Heisman (2010)
https://web.archive.org/web/20140708105628/http://www.chesscafe.com/text/review781.pdf

mocl125

Here is my favorite method of doing it: http://chesswinning.com/how-to-analyze-chess-games/

Forums
Forum Legend
Following
New Comments
Locked Topic
Pinned Topic