Learning from Game Analysis
Bad habits, critical moments, and running engines
First off let me apologize. It’s been too long since I last posted. Last time was my April Chess (aka accountability) report. Today we’re going to see what we can learn from analysis of our games.
Secondly, please note a small change to my Substack. I changed the name to “Chess in Small Doses” as a pun and a mission statement all at once. There’s just too much written about chess to ever take it all in. My own journey has been to focus on small steps to improve my thinking or knowledge in chess. A long slow game per week, some Chess Steps Homework, and a weekly lesson is how I got here. So for this Substack, I intent on keeping it tight and focused on one thing per post…otherwise “small doses.”
Lastly, I’m sure I’m not the only one but having my kid’s school end for the Summer has not helped Daddy’s free time to write. I’m sure you can empathize.
That being said, today we’re going to explore what we can learn from game analysis. In May I played against a much higher rated opponent and lost as Black. Afterwards we did a post mortem and later I did my analysis.
Analysis comes in at least 3 different flavors.
You can look at the opening book and see where you deviated from theory.
You can look over the moves yourself and try to identify the critical moments where the game was either won or lost.
Look at the engine evaluation to look for what you missed, got wrong, or got right.
Of course there’a MUCH more to analysis than those three steps but let’s have a moment of honesty. Analysis is by far the best and most assured way to improve your game. Analysis also takes time, lots of time. Many people skip right to the computer analysis and fail to learn from their mistakes. It’s like having ChatGPT write your emails or papers. Sure you get it done, but it’s not your voice and you’re not really learning how to write.
Improvement comes only when you add knowledge or eliminate a mistake. After you learn how the pieces move and basic tactics, everything after that is small improvements. From about 1200 on up my sense is people jump from latest thing to latest thing only to get frustrated with the lack of improvement. Do these three steps and you’ll see gains.
Step 1 - Look at the Book.
Both Lichess and Chess.com have opening book resources so you can see what is the most common moves. It’s useful to look at the game and see exactly where you left book. It’s more useful to see where you left the book and WHY.
In my prep I chose to play 5…Bc5 because I wanted an active bishop with pressure on the king. However that’s about as far as I got. After 6.Nd2 the best move for Black is either 6…Be7 or 6…e4 (more on that little move later). The point is that 5…Be7 or 6…Be7 is just fine and Black gets to enjoy the space for now.
Instead I chose to play 6…Be6 which isn’t best. Why did I play that? I thought development was important. I played this move on “general principles” and not based on knowledge of the opening or analysis. I was aware at some level that I was out of book, but didn’t stop to see what was now on the board.
This turns out to be a common issue for me. The tendency I have is to just keep moving instead of scanning for a better move. We can’t calculate forever, but just looking at Checks (none), Captures (Bxf2+), Threats (Nb4?), and pawn breaks (e4, d4) would have maybe started the process. That’s something I can work on making a habit.
Step 2 - Identify Critical Moments
In each game there are moments where the game turns. When going back over the game, it’s incredibly useful to try and identify where YOU think those moments were. Take this position from my game:
It’s Black to move and played 7…Qd7. My plan was to play Bh3 and try and trade the strong LSB bishop. What I missed was the loss of time I’d face after 8.b4! My bad habits bit me again!
After7…Qd7 8.b4 Bd6 9.b5 Ne7 10.c4 I lost a ton of time and space on the queenside. The game was almost decided from there.
White is clearly better. The center is going to be opened and my king is still in the center. In the future I now know to watch out for that c3…b4 move against a bishop on c5. Maybe that was also the reason why Be7 was such a popular move, removing the bishop as a target.
Step 3 - Run the Engine
One advantage Lichess.org and DecodeChess.com have over Chess.com is they will show you the Evaluation graph. It’s just a bar that shows you which side is better. I find it incredibly helpful to see if I missed anything or if I got anything right. Going back over the game just using the graph can really help reinforce what you learned in the analysis. It will also reveal what you may have missed. Here’s the graph from my game.
Pretty much a smooth win for white… but it seems I missed something early in the game.
The engine says this was -0.9 which I’m not sure what it means other than Black has an opportunity. You may recall I played 6…Be6 here. Using the simple Checks, Captures, Threats, and Pawn Breaks idea I could have started looking for a better move. Turns out the pawn break 6…e4! is very strong.
After 6…e4 7.dxe4 dxe4 Black is doing well. (I try to calculate at least 1 1/2 moves and not much more that 3.) Black is even threatening e3 here!
The engine eval graph told me I missed my chance to make it a favorable game. Instead I allowed White to gain Queenside space and have a comfortable game after that.
Conclusion
Answering at least three questions during game analysis will help you hone in on the weaknesses you have in your game.
When did you leave book (and why)?
What were the critical moments?
When did the engine evaluation change?
Thank you for reading and as always if you like it, please share my Substack with your friends. Best of luck!
Nick Vasquez