Analyzing without an engine

For the past several years I've analyzed all my serious, slow games after completion and it's been extremely helpful. It took a long time before this habit started to pay dividends in terms of results, but I'm convinced that it made me a better player. It's less that I discover specific errors and learn not to repeat them than I remember what thought process led me to make certain moves and analysis showed that my entire line of thought was wrong. But that's not what I wanted to write about.

Up until recently I had almost always done my postgame analysis with the help of an engine. The first thing I want to know after I finish a game (especially if I didn't win) is what mistakes I made and what I should have played instead. The computer identifies tactical mistakes instantly and most of the time will tell me if I made an error in evaluation. It's not as good at finding errors in planning, but no one said using a computer will identify every flaw in your play. It's great having such a helpful tool, and it saves a lot of time having the computer do the heavy lifting of finding poor moves and showing you better ones.

Given that my online rating has not increased in the past year, I thought I would take the advice of many instructors of not using an engine to help me until I've already gone over the entire game without computer help. So far it's been a difficult slog. The problems are:

1. Without using a computer it takes forever to complete the analysis--literally hours, usually split over several days--and it means that I play fewer games.

2. Analysis is difficult work even with engine assistance; without it it's even worse. I'm doing this because I hope it will ultimately make me better at chess, but it is not fun.

3. Using an engine to check my work after I've analyzed a game without one usually reveals horrible errors in my understanding of the game even when I've had as much time as I needed to think about it and am able to move the pieces around. It can be very humbling and occasionally humiliating when you've spent half an hour analyzing an unplayed line that you've concluded would have been better than what you actually played, only to have the computer point out something you missed on the second move of the variation that refutes all the work you did.

I had many of the same difficulties when I started analyzing my games with engine help, so it's possible I'm just going through the same learning process. At first I wasn't convinced that that was helping me get better either but I kept at it because I felt like I was learning something, even if the results did not follow. Eventually after a year or more, they did. I can't be certain that analyzing my own games was responsible for the improvement--I was doing a lot of other things too including tactics exercises and playing a lot more slow games--but I believe it had a lot to do with it. I'm hoping that going over my games without an engine the first time through will eventually lead to similar gains.

In the meantime, it's just a lot of extra work. I can only hope that it's like eating your spinach: it may not taste very good but it's good for you.