Learning

How to Study Chess Games with the Computer

How to Study Chess Games with the Computer

In this post I will show to you a common method to study chess games with the help of the computer. Studying games played by strong players is a well known training strategy that helps learning new tactical patterns as well as strategic plans.

By the end of this article you will know what are the steps to follow in order to optimize the time you dedicate to study chess. A clever usage of chess software (engines and database) will lead you to dramatically reduce the time wasted!

So, how to study a chess game with the computer? Here are the steps that you should follow:

  1. Go through the entire game once, quite quickly, without stopping to analyze at any particular moment and without using the engine, but taking note of the moments that seem the most interesting.
  2. Go through the game again, this time with the engine on. After each move, give to the engine about 30 second to analyze the position and take note of the moments where the evaluation changes significantly.
  3. Now, without playing the moves in order, set up on the chessboard each of the positions that you noted at Steps 1 and 2. Spend as much time as you need to figure out what is going on. Why was the next move played? And why did the engine change evaluation (it should mean that there has been a mistake)?
  4. Opening study. Go once more through the first 15-20 moves (the exact number depends on each game). Do you actually know this opening? Compare it with your Opening Repertoire database. Is there any new move that you didn’t know about?
  5. Middle-game study. Set up the game’s position where you think that the opening is over. Are you familiar with the structure? Try to guess the plans for both players and then compare it with the plans actually played.
  6. Tactical training. Now that you have better understanding of what has happened in the game, take again all positions that you noted at Step 2, and analyze again the mistakes that players have made under the light of the strategic plans you are now aware of. Was a mistake purely tactical? Or was it a strategic misunderstanding of the position?

This is pretty much everything you need to know. However, if the above steps sound a bit too vague still, then keep reading because I am going to elaborate each of them much more in detail. I will also explain, for each step, what feature of your play is being trained.

Improve the Chess Board Visualization (Step I)

The first step in my recipe is to “scan” the game rather quickly without a chess engine.

You can do this very easily, simply by loading the game (most typically in PGN format) into your favorite chess software, for example LiChess or ChessBase, and use your keyboard arrows to replicate it move after move.

Do this with the engine turned OFF. For example, in LiChess you can toggle off Stockfish analysis just with a click (on a toggle-like button, indeed).

The tricky, but also most important part, is that you must keep your brain AWAKE while doing this. Weak chess players often have the tendency to shut off their brain when they look at the engine’s evaluation. This makes your training process passive, and therefore much less effective.

Instead, you must keep your spider-sense well awake, in particular to spot any moment in the game that rings a bell, or that just seems interesting for some reason.

You can, and should, also keep trace of these moments, for example by writing down on a piece of paper the move number, or more simply by adding a comment in the file, like “Interesting moment spotted!”.

What is the purpose of this first, very quick, step in the process of studying a chess game? The idea is to work constantly on the board visualization and on your intuition. Strong chess players “feel” the position, which is why they are very strong even at extremely fast time controls.

By simply taking a quick look and making conclusions such as “something is happening right here right now”, you will train your board visualization. Of course, lots of error will happen at the beginning, but don’t worry — keep doing it anyway. Improvement will arrive by repeating the process over and over again.

Error checking in a Chess Game (Step II)

The second step in this chess training plan is to scan again the entire game, and again rather quickly, but this time with a chess engine active.

Your task in this phase will be to take note of the moves that create a strong change in the engine’s evaluation. Watch out to avoid misinterpreting the word “strong change” though!

Amateurs, and in general inexperienced chess players, often make the mistake to consider any change in the evaluation like a blunder of the player. Things can’t be more far from the truth!

In my opinion if your level of play is anywhere below 2300 you can discard every change in the evaluation that is less than 1.00 point. So, for instance, if the evaluation at some point is -0.2 and after the black’s move it becomes +0.2 you might be tempted to conclude that the black players has made a mistake.

This could even be technically true, but it’s irrelevant from all practical points of view. A change in the evaluation of 0.40 is something that you would never spot at the chessboard in a real game.

Chess engines and chess software in general are great tools, but they must be used with grain of salt. Calling “mistake” a move that loses less than half of a pawn if frankly non-sense.

Based on my experience, jumps in evaluation of 1.00 and above are signals that something wrong has actually happened. Strictly speaking about numbers, loosing 1.00 is kind of blundering a pawn, which is something you should definitely watch out for.

So in short, you must take note of all moves that cause a change in evaluation of 1.00 point or more. Again, you can do this on a piece of paper, or simply by adding a comment in the file, like “Stockfish passes from -0.6 to +0.8”.

What is the purpose of this phase, that is scanning the game with the chess engine? The idea is that your eyes will be trained in checking errors made on the board. Mainly classical mistakes such as hanging pieces, or 2-moves combinations that loose material, but chess engines can also detect strategical mistakes, like when the fact of choosing a wrong plan changes the evaluation drastically.

If you iterate this step in the training when studying (a lot of) chess games then your natural instinct will be strengthen in avoiding mistakes and blunders… before they happen!

Training Chess Tactical patterns (Step III)

The first two steps were somewhat “easy”: you just kept pressing the right-arrow on your keyboard and made the additional effort to keep your brain awake. Which shouldn’t really be called “effort”!

By doing that, you now have a list of positions to analyze. The ones coming from Step 1 are positions that you think are interesting, basically because something caught your attention. The one coming from Step 2, instead, are interesting because the chess engine tells you something is not quite right.

Now the “hard” work begins, and it will also be the more productive work.

First of all, you have to forget about the game. Yes, just forget about it. Your task now is to focus on each position like if it was a puzzle to solve.

One position after the other in your list, set it up on your chessboard (in the computer is also fine), and study it. Analyzing a chess position is a tough task, but you can start by asking yourself the following questions:

  1. What are the main features of White’s position? What about Black’s?
  2. What are the main weaknesses of White’s position? And in Black’s camp?
  3. Who has the initiative?
  4. Is there any counter-play target for the side without the initiative?
  5. What plans are available for the two sides?
  6. What tactical themes are in the position?

Then, something that I warmly recommend is trying to guess the next move (or the next few moves). As always, the choice of a move must be based on a mix of precise calculation, positional and strategical thinking, and an overall objective. In this sense the above questions should be very useful.

The effort you make to choose a move is like a gym for the brain. You should also set a time limit, for example, a maximum of 5 minutes allowed to choose the move, but within this limit try to focus at the highest peak you can.

When you are ready with your chosen move, compare it with the move chosen in the game. If the two are different then the questions you should ask yourself are the following: Are the two moves pursuing the same objective? If yes, what is the difference in playing one versus the other? If not, what is the other move’s objective? After seeing it, do you think it is any better than your choice?

All of the above must be done with the engine OFF. After you have analyzed the position and made some conclusions with your own brain, then it’s time to turn the engine ON again.

Let the chess engine analyze the position for at the least 60 seconds. Then, new questions to think about arise:

  • Look at the top-3 suggestions from the chess engine. Do they have similar evaluations? If yes, then it means several moves are possible in the position. If not, then it means the player has a almost obliged choice, and missing it might bring big troubles.
  • Is your move among the top-3 engine’s suggestions? If yes, is it the best among those three? If it’s not the best, why so? What is the difference between the first engine’s choice and your choice? Play a few more moves in both variations (with your move and with the engine’s move) in order to better understand the differences.
  • At the previous bullet point you went deeper in two variations. For sure, doing that has pointed out many details that you previously missed, both tactical and strategical. What did you miss? Was it something that you more or less knew in advance, or a complete new light on the position?

This part is probably the toughest of all steps in this recipe. Making your own conclusions about a chess position, BEFORE looking at the chess engine evaluation, requires a lot of effort.

Yet, it’s the most important part of the training because it simulates very closely what you are required to do during a real game. In a tournament game, over a real chessboard, there is no engine that suggests you moves or evaluations, so you have to make your own conclusions. And these conclusions will guide you through the choice of a move.

Training your decisions process is a incredibly valuable part of your overall chess training. In my opinion, the secret is to keep doing it over and over, again and again. Study as many chess games as you can. Maybe group them by the opening. And when you arrive to this Step III then give everything you have to make your own analysis of the position, and then discuss your choices by comparing them with the chess software that you are using.

Improve your Opening Repertoire (Step IV)

In a previous post of mines I talked about how to build a opening repertoire, therefore here I will assume that you have your own set and ready to use!

The objective of this step, Step IV, is to compare the game you are currently analyzing with your prior knowledge about the opening that was played.

And the final target can be either to update your personal files with new ideas in this opening, as well as refreshing in your memory the theory that you already know.

You should start by quickly “scanning” the opening stage of the game. Although there’s no universal knowledge of when a opening ends, in most cases it should happen around move 15-20. If it’s a opening that you have already studied, then you should be perfectly capable to understand where that precise moment lies.

This initial rapid “scan” through the moves is, once again, in order to familiarize your eyes with the overall opening. Your brain will work in the background and will make statements such as “I know this very well”, or “what the heck is that?”, even if you don’t pronounce those words.

When you are finished with this first step, I recommend you to open a new window on your screen with the repertoire file for this opening. If you can have both the game chessboard and your file board opened at the same time then it’s perfect.

If you are using LiChess then you can simply open two tabs in your internet browser with the two boards.

Your task now is to carefully check the actual game move by move and see if a move is or isn’t in your file.

When a move is not therein, then you should ask yourself: is it an important move for this opening? How many other times has it been played (you will have to use a chess database to answer this)? Does it pursue a similar plan/objective as the other moves that I already know, or is it a completely new plan that I was unaware of?

You must think about the above questions and eventually answer them. Again, making conclusions is the single most important thing in a chess game. Finally, you should decide whether including that move into your file, with a few explanatory variations, or not.

When a move already is in your file, then your task is to compare the line played in the game with the variations that you have in your file. Are they similar? If not, what kind of plan did the player choose in the game, and why don’t you have it in your file? The reason might simply be that the chosen plan is wrong and therefore you discarded it when you studied the opening in first place.

You can use the chess engine to check if the plan is completely wrong. However, even if it is, I would still suggest you to add it into your file along with a comment such as “This plan is wrong because … (add the reason)”.

Once again, repeating this process over and over will boost your memorization skills, and help you remember lines and plans when you play over the board.

Practice Strategic Plans (Step V)

After you’re done with studying the opening (in Step IV), it’s time to go again into the middle game.

This time is going to be more straightforward though. Set up on your chessboard the position where you think the opening stage is over (you must have decided this during the previous step).

Now turn off any chess engine, and close any chess database that you may have opened. Give yourself a new, slightly larger, time limit. I suggest 10 minutes.

Spend the next 10 minutes like if you were in a tournament. Put away any distraction that you might have, like cellphones, social apps, etc. and decide your next move.

Even if you remember what move was actually played in the game, due to the previous steps in this training, it doesn’t matter. Still, spend 10 minutes in full-focus mode, calculate as much as possible and evaluate as many options as you can.

At the end, you must decide on ONE move. Just one. In a tournament game, you must always choose one single move, and you have to be quite fast at taking this decision.

Write down your choice, and, if you want to, also the main points of your analysis that lead you to choose such a move.

With that done, it’s now time to work again with the chess engine. Turn it on and let it analyze the same position for a couple of minutes or more.

Your task now is to understand its choice, that will likely be different from yours! And also, you should analyze ALL variations that it considers, and compare them with yours. You must understand what key elements in the position you missed, such as tactical shots or strategic weaknesses.

One question you might have at this point is: What’s the purpose of doing this as (almost) the last step?

You’re right. You’ve already analyzed the game quite in some depth before, in Steps I to IV, therefore your training in Step V can be “biased” by the knowledge that you’ve already got.

I have a question for you too. Why would that be bad? In fact, I claim that it’s not.

The concept that I am trying to stress here is always the same: Repeated things help (literally translated from the Latin Repetita Iuvant). My advice is to keep repeating these important concepts and analysis over and over, until the analytical process becomes natural for your brain.

Practice Chess Calculation (Step VI)

Down to the last step! This time you are going to train pure calculation, which is arguably the most important skill in critical moments of the game.

You will start with the list of positions that you got at the end of Step II, those where a move causes a change equal to or greater than 1.00 point in the chess engine’s evaluation.

For each of these positions, your task is to set it up on the chessboard and start calculating variations! Spend a reasonable amount of time performing your calculation (I suggest between 5 and 10 minutes), while trying to stay as focused as you can.

I warmly recommend to try to put yourself into the tournament mindset, because the ability to focus and concentrate is a skill that you should train too. That means no distractions, no music, no chats.

When you’re done with calculation, you might want to spend 1-2 minute to write down the variations that you calculated, to make sure you don’t mix them up in the upcoming step.

Now you can activate the chess engine for the last time, and your objective is to check each one of the variations that you calculated with the engine. You should really check EACH move in EVERY variation.

The engine will tell you what you missed, and what tactical shots you overlooked. Your ultimate goal should be to have a sort of photographic memory that allows you to quickly spot these patterns and therefore to make less errors in your calculations.

Of course, this is a very difficult objective to achieve. Nevertheless, there is no way other than training, training, and … training! Good luck!

Conclusions and references

In this post I showed a 6-steps recipe to study every chess game.

I am aware that it’s not a easy training. Actually, it takes a lot of effort, and if you have a full-time job and lots of other commitments (like I do), it may seems impossible at first.

However, I saw that in my case it is possible. In fact, at the beginning it seemed really really hard, but after the first couple of games the process becomes somewhat “easier” (not really :D). As with everything, one simply has to get used to it.

Also, if you keep doing it you will learn a incredible number of things by looking at other people’s games in real depth. You will learn how to spot tactical opportunities and how to quickly recognize if a strategic plan is sound or not, in a given position.

I will leave here links to previous posts of mines that are chess training related, and can be useful too.

Have fun!!