I was just wondering how others go about analysing their own games and how much time they spend.
I spend 2 or 3 hours per game and find that each one generates further questions which take more time to answer. I tend to go through with a chess board, database software and pen/paper to start with. I look at the opening first and figure out where I or my opponent could improve. Then I tend to look at what I recall as being the turning points in the game to see where I went wrong. Next I go through the whole game and concentrate on the tactics. Finally it's the 'Fritz' tactics review. At the end of the process I have a list of questions for further work.
One new thing I've started doing is making positional sketches of my own mistakes. The personal element makes them more memorable and hopefully I won't make the same mistake again.

This sounds very thorough to me - I assume you also go through the game with your opponent immediately afterwards?

? This sounds very thorough to me - I assume you also go through the game
? with your opponent immediately afterwards?

Definitely. I think that this is the most helpful part of the analysis. It is amazing how much you miss and also fascinating to hear the reasons behind some of your opponent's moves. Unfortunately, sometimes the interesting games are the longest and that means that there is not much time to recover before the next round starts!

Still, I'd be interested how long it takes other people to review their games.

Thinking about the number of games played in a year, this season I am aiming for between 35-45 games. I'm sure it is very dependent on how much time you have, but, if you are working full time and have a family, it can be difficult to devote as much time as you would like, and I think the analysis is even more important than the playing. I have restarted playing for one year so far after quite a break and I am hoping to see a noticeable difference in my play next season.

?
? Still, I'd be interested how long it takes other people to review their
? games.
?
I put my games into chessbase as soon after I played them as I can, and try and record what I thought at the time and what variations I had actually seen during the game, and what we discussed afterwards. I have also started recording how much time is left at various points in the game and I note this into chessbase also. This takes about an hour. Then I go through and think about what I should have done in retrospect, then I check the opening book, and finally put Fritz on it. So thats at least another hour.

If I conclude that I had no idea what to do in a position then that could lead to further study of other games in the same opening. The hardest thing is sometimes knowing exactly why you went wrong and how to avoid it in future. You could potentially study them for hours. But after a while you want to move on to the next game .....

I rarely go over games with my opponents - there just doesn't seem to be time (it may be time for the next round, or time to eat, or...) - or they just pack up and go away right after the game.

When analyzing, I first try to study the game on my own without a computer. I take notes, especially writing down variations that I want to check with my silicon assistant. I also identify points in the game where I didn't have a good sense of what was going on.

Then I enter the game into my database, and let the engine do a full analysis. When it is finished, I go through it again with the engine running. At this point I add annotations. I might add opening improvements, or add the main line where the game went out of book. I note any tactical oversights. I add descriptive prose to turn analysis into a meaningful narrative I can take away.

One thing I really like is to spend a fair amount of time fiddling with positions where my intuition did not square with the "truth" of the position. That means playing out as many side variations as I need to in order to "feel" the right move in that position. I think that is where most of the improvement takes place.

Using that above approach, I have discovered that lately I have overlooked some winning lines (for either side) because they seemed naively "anti-positional" in my mind's eye. Now trying to fix that, somehow.

40 games average 2 hrs per game = 80 hours plus 3 hrs analysis per game (120hrs) = 200 hours, add 3 hours a week chess on the internet (with 2 weeks holiday) (150hrs) = 350 hrs add some time traveling to club nights and tournaments (50 hrs) we get to a (conservatively estimated) 400 hours! That's the equivalent of 10 weeks full time work in a year. My family would disown me, my business would collapse, I'd end up living in a caravan. Got to admire your commitment though!

And, more seriously, analysing my own games is something I've done to little of except when I did the personal coaching through this site - where you have to analyse your own games in order to take part. With a game on tonight I will follow your method over the weekend and see how I get on.

Recently I read the new book by Josh Waitzkin where he gives some very readable exposition on spending hours and hours on in depth post game anaylsis (no game details though as it's aimed at a general readership.)

I notice everyone seems to do there own analysis first without an engine and I suspect this is a key to gaining more benefit from the time. The problem I've had with running the engine through a game first is that it's all over too quickly and without much effort and so not much benefit.