Why Grandmasters Almost Never See Ghosts
Grandmasters almost never see ghosts. You do. I do. Every player below master level does. And no, ghosts in chess have nothing to do with spooky bishops or the rook your opponent forgot to put back on the board. A ghost is a threat that doesn’t exist, an illusion that appears only inside your calculation. You “see” a piece covering a square it actually isn’t. You avoid a move because of a tactic that isn’t real. You panic over something your opponent cannot possibly do.
Everyone has felt it: you calculate a line, think your queen is hanging, abandon the idea, then later realize the threat was imaginary. You didn’t lose to your opponent. You lost to a hallucination. That has happened to me so many times, especially in classical.
But here’s the real problem: most players don’t know why ghosts appear. They think the issue is nerves, or lack of talent, or calculation, “just being hard.” In reality, ghosts appear because you calculate in the wrong order. Your brain jumps back and forth between lines, mixes positions, skips steps, forgets what was captured, and suddenly you’re evaluating a position that exists only in your imagination.
TABLE OF CONTENTS
2. How Most Players Calculate And Why It Breaks You
3. Kotov’s Tree of Calculation: The System That Kills Ghosts
4. Why Kotov’s System Eliminates 90% of Ghosts
In this post, my aim is not to make you avoid "seeing ghosts", even grandmasters do it sometimes. My real objective is to help you realize why that happens, how to detect when it happens to you, and what you can do to prevent it from happening regularly.
Masters in general avoid this, not because they’re fearless, but because their calculation is orderly. They learned a system. And the most accessible version of that system goes back to Alexander Kotov.
Before we talk about ghosts in detail, we need to talk about the method that prevents them.
2. How Most Players Calculate And Why It Breaks You
Take any player below 1500 and ask them to calculate a complex forcing sequence. Now, let's get a position and analyze a possible response.
Their inner monologue usually looks like this:
Do you see where I am going? Instead of focusing on the quality of moves, people often enhance quantity. That is almost always wrong, also outside of chess.
Have you ever tried multitasking? It is a very interesting concept, and I encourage you to look it up. In some way, calculating as seen above is the mental equivalent of trying to read five books at once by flipping between pages. Every time you jump to a new branch of the position, pieces shift around, threats reappear, moves you already rejected suddenly seem playable again, and the whole thing turns into a fog. In that fog, ghosts appear. Think of it this way. Changing duties requires mental effort. This effort is precisely what helps you "see more" when correctly used.
You don’t evaluate positions incorrectly because you’re weak. You evaluate incorrectly because you’re evaluating positions that don’t exist. Remember those five books I talked about? Now imagine you read the first one, then the second one, then the third, and so on. Undoubtedly, your brain will understand more.
Calculation happens just like that.
Kotov’s method is a cure for that. Well, sort of.
3. Kotov’s Tree of Calculation: The System That Kills Ghosts
Alexander Kotov, in his book Think Like a Grandmaster, proposed what remains one of the most influential calculation systems ever taught. His “Tree of Analysis” is simple enough that you can start using it today, although it will take a lifetime to master.
“Having chosen your candidate moves, you must examine them one by one, calculating each variation to its end. Only then can you correctly compare them and make your choice.”
That single sentence is the core of the method. The brilliance is in the structure it forces on your thinking. Let’s break it down simply.
Step 1: Choose 3-5 candidate moves
Candidate moves
Candidate moves are the small set of promising moves a player seriously considers in a given position, rather than calculating every legal move. By narrowing the options, calculation becomes deeper, clearer, and more efficient.
This is the key. You find yourself in a certain position, and time is limited. Thinking about all legal moves would be slow and completely inefficient. That is why you must identify the most promising moves (remember the rule: checks, captures, and attacks). Depending on your level (and time control), you should choose a concrete number. If you're below 1000, consider 2-3 moves in complex positions. The higher the level, the higher the number of options, and the higher the depth is needed.
The point is that you choose the moves before calculating. This stops the brain from randomly wandering into new branches halfway through analysis.
Why don't we take a position and try to apply this rule?
White to move.
This position comes from the eighth game of the World Chess Championship of 2018 between Carlsen and Caruana. With the position above, I want you to identify 3 candidate moves. Remember: checks, captures, and attacks. Got it?
Scroll down when you are ready.
7
6
5
4
3
2
1
Time's up! Ok, the numbers needed to be added so the options can't be seen. Let's take a look!
How did you do? I didn't expect you to find those variations, just that you understand the key ideas of candidate moves.
Step 2: Analyze one by one
This step is slow, and you should take it seriously. After you list your possible choices, start analyzing the one you think is best. Analyze possible hidden branches and do not look at the other candidate moves until you finish.
This is important because some lines are "prettier" than others, if you get me.
Positional maneuvering is normally hard to calculate, can lead to errors easily, and, overall, is boring. However, calculating fun and short tactics requires less mental effort, as there are few branches (as the sequences are usually forced). That is why, when we calculate boring lines and our brain sees another option, it tends to go that way.
In this step, I want to teach you how to educate your brain, in a way.
The three main ideas to do this are:
1. Be methodical
2. Be methodical
3. Be methodical
Got it?
Yeah, but, Chesser, care to elaborate?
Of course.
This, according to Kotov, is where amateurs break down. They start analyzing one line, see a problem, panic, and jump to another move before finishing the first. Later, they return to the first line with half-remembered details and end up mixing two branches of the tree. That mix produces ghosts.
Kotov’s rule:
Pick a move. Analyze it until the position becomes obviously static (end of forcing). Evaluate. Only then move to the next candidate.
This is the moment where ghost-busting happens.
Step 3: Compare and choose
As simple as that. Well, sort of.
Only after finishing the entire tree do you compare lines. This final evaluation suddenly becomes much easier because the branches are clean and separated. You are not choosing between “three blurry positions.” You are choosing between three clear ones.
The quality of your evaluation improves immediately. The ghosts fade.
4. Why Kotov’s System Eliminates 90% of Ghosts
Ghosts appear because of cross-contamination between lines. Kotov’s separate-branch system eliminates that contamination. You never mix positions. You never imagine pieces where they can’t be. You don’t hallucinate threats because every threat is tied to a specific branch.
Your brain becomes a filing cabinet instead of a washing machine.
Kotov’s system doesn’t make you a genius. It makes your thinking clean.
And clean calculation is ghostproof.
However, reading a post about it won't help. The way to apply this is simple. First of all, puzzles are great exercises to start using this method. But I believe there is one better. I came up with this myself, and I feel it is very interesting.
You enter chessgames, and you find a random game. When you do so, advance the first 15-20 moves and think about what you would play. This way, you do not know if there is a tactic, if you must find a key positional idea, or if you just need to develop. After going through the steps of the method. Enter the position in Stockfish and analyze those branches. Please tell me in the comments if you tried it and how it went. I am so excited to see results!
Seeing ghosts in chess is not a sign of weakness. It’s a sign of disorder.
Most players don’t lose games because they can’t calculate. They lose because their calculation has no structure. Grandmasters don’t calculate more variations than you (though they do). But they calculate them cleanly.
Kotov’s Tree of Calculation is not magic, and it won’t turn you into a machine overnight. But it does something far more valuable: it forces your mind to slow down and stay honest. One branch at a time. One position at a time. No mixing realities.
At first, the method will feel artificial. Your instinct will fight it. That’s normal. The paradox is this: once you learn to calculate with order, calculation becomes more intuitive, not less. You won’t stop making mistakes. But you’ll stop losing to things that were never there.
And that alone is worth a lot of rating points!
Thanks for reading 
.
