You Won. That Doesn’t Mean You Played Well.

Winning doesn’t always mean you played well. Sometimes the most valuable lesson is hidden in the mistake your opponent failed to punish.

Winning a game is a very simple way to stop asking questions.

Your rating goes up. Maybe the finish was nice, or maybe there was a mating sequence you found. You look at the final position, feel satisfied, and just move on to the next game.

Losses are very different.

When we lose, it is obvious that something has gone wrong in that game. We explore where. We open the analysis board, look for the blunders, check the opening, and just try to understand how in the world it led to our loss.

After a win, the result seems to have already answered the important question:
I played better.

So many people draw the conclusion that if they won, it means their tactics worked and that they played well.
But that is not always true.
Sometimes your opponent might have simply failed to punish you, and the lesson you can gain from it may be more valuable than it seems.

A game I almost learned the wrong lesson from

I played this game recently.

The position became kind of messy very quickly.
After 7...Qh4+ 8.g3? Qxe4+, Black’s queen entered with tempo.

A few moves later we reached:
9.Qe2 Qxh110.dxc6
Black had already taken my rook.

Then came a very important decision by Black:
10...dxc6?

Replay the original game on Lichess

The board above starts immediately after 10.dxc6, so you can look at the position before Black makes the decision that changes the game.

The game later continued:
11.Nc3 Qxg112.Be3 Qh1?
Those moves gave me time — precious time.

My pieces started coming out. Black’s king remained exposed. The queen ended up locked on h1, and the actual game finished pretty quickly.
After 13.O-O-O, my attack gained momentum. The knight came into d5 and c7 with tempo, Black’s king was forced into the open, and the game ended with 19.Qg4#.

Mate.

If I stopped there without analysing, I could tell myself a very pleasing story.
Black got greedy.
I developed quickly.
I attacked their exposed king.
I found the mating sequence and won.
All true.
But also incomplete.
What if Black didn’t make those mistakes?

A win can hide the mistakes you needed

This is what makes wins dangerous from a learning perspective.

When you lose after making a bad decision, the punishment is obvious.
When you make a bad decision and still win, the feedback can be completely wrong.

You may remember the game as proof that your idea worked.
But maybe your opponent missed a tactic.
Maybe they chose the wrong plan.
Maybe they failed to convert a material advantage.
Maybe your attack only existed because they gave you enough time.

A move does not become good because somebody failed to punish it.
Your next opponent does not care that the previous one missed the chance.
And sometimes a win hides not only your mistakes, but the mistakes you needed your opponent to make.

So I changed the mistake and played on

This is where I used Zen.

After analysing the game, I could see that Black had a much stronger option instead of 10...dxc6?
Rather than only clicking through a long engine line, I went back to the critical position in Zen’s Explore mode and changed the move to the stronger engine suggestion:
10...Qxg1

Then I stopped looking at the lines.

I switched to Simulation mode and continued playing from there.

That was the interesting part.
I already knew that Stockfish considered Black better. What I wanted to know was what that actually meant if I had to keep playing the position myself.
And naturally, I tried to do what had worked for me in the real game.
Develop the pieces.
Get the king out of the centre.
Create threats against Black’s king.
Try to turn the material disadvantage into an attack.

Here is the simulated continuation:

Replay the simulated continuation on Lichess

The simulated game continued:
10...Qxg111.Be3 Qh112.Nc3 Bb413.O-O-O Qxc6

I had developed the bishop and knight and castled queenside, just as I had in the real game.
The queen was even back on h1.

In the original game, that setup became the beginning of my attack.
But this time Black had not played 10...dxc6? first.
That one difference changed the position.

Black was able to take the pawn on c6, continue developing, and deal with my activity without giving back the material advantage.
The game continued, and by 16...O-O the difference from the real game was becoming hard to ignore.

Black had survived the immediate danger, castled, and kept the extra material.
I was still trying to create activity, but there was no exposed king to chase anymore.
I kept trying to make something happen, but Black had already consolidated.

The simulation eventually ended with 33...Qg3#.
Black mates.

Obviously, this does not prove that every move was forced or that this is the only way the position could have developed.
That wasn’t what I was trying to test. I wanted to see what happened if I corrected the critical mistake that had allowed me back into the real game and then tried to play the position the way I had played it before.

And this was probably the most useful part of the simulation.

I tried the same ideas.
I developed my pieces.
I castled queenside.
I tried to attack.
The queen even went back to h1.

But the attack that looked so convincing in the real game simply wasn’t there anymore.
In the actual game, I mated Black on move 19.

In the simulation, I kept trying to recreate that attack and eventually Black mated me.
My original 1-0 suddenly looked rather less reassuring.

Sometimes the most useful move was never played

We naturally analyse our own moves.

What should I have played?
Where did I go wrong?
Could I have found something stronger?

Those are good questions, but they only examine half of the game.

Sometimes the most useful discovery is a move your opponent could have played against you.
That can reveal a tactic you never saw.
A weakness you did not understand.
A pawn break you were not ready for.
Or, as in this game, a position where your opponent had already won material and simply needed to avoid giving you enough activity to compensate.

There is a question I increasingly like when reviewing my games:
If I were playing against myself, how would I punish this position?

That question often makes a win look much less comfortable.
And that is good.
That is usually where the useful lesson is.

Sometimes you should finish the game your opponent didn’t

This is one of the reasons I’ve been building Zen around more than a normal post-game engine review.

Stockfish can find the critical moments, and Cassia can help explain what is happening in the position.
Explore lets me change a move and branch away from the original game.

But Simulation is the part I found particularly useful here.
Change the mistake.
Start from the new position.
Then play.
Instead of asking only:
“What was the engine line?”
you can ask something more practical:
“If my opponent had actually found this, could I handle what came next?”

In this case, I even tried using many of the same ideas that had won me the original game.
They simply did not work the second time.

For this game, the answer was not especially flattering.
Which is exactly why analysing it was worthwhile.