Why should we analyze even the blitz games we won?

Nine days ago, on July 20th, I played a blitz tournament in Belgrade. The Serbia Open was taking place at the time, and since it was the final qualification blitz that evening, the field was stacked. I scored 8.5/11—a result I was really satisfied with, especially coming off a tough classical tournament in Zenica where I dropped below 2300 for the first time in six years.

What I want to emphasize in today’s post is that we need to analyze our games even when we win.

It’s easy to argue that if we won, we played well. But that’s like claiming you’ve mastered a university subject just because you passed an exam after answering three specific questions out of forty. You were up to the task that day, and you answered the questions put in front of you—but with a different draw of questions, the result could have looked very different.

In chess, your opponent might not test your blind spots, but the next player will. There is almost always room for improvement, especially in the opening. By analyzing our wins, we spot where our preparation was lacking and make sure we’re ready for the next test.

Below is the game from the tournament that I want to show you.

Opponent started with 1.e4, and I answered with 1...b6. Owen's Defense is my favorite opening when it comes to fast time controls. It gets its name from English priest and chess player John Owen, who lived from 1827 to 1901. He even beat Paul Morphy, the greatest player of his era, using this very defense.

All of this is known theory, more then 200 games in my database, here I played interesting move 9..a5 10. a4 Be7 11. h4 Ba6 12. Bb5+ Bxb5 13. axb5
I knew the theory until this moment, and I knew that I have a plan to attack b5 pawn

And I played 13..Qd7, and after 14.h5 c4 15.0-0 Qb5

I was up a pawn. The game continued 16. Rfb1 h6 17. b3 Qc6 18. bxc4 dxc4 19. d5 $2 Qxd5 20. Rxb6 O-O 21. Nd4 Rc8 22.Rb5 Rc5!

I defended the pawn and later went on to win the game.

I was quite happy with how it went, believing I had played well and deserved the victory. But thanks to at least one good habit left over from my years in competitive chess, as soon as I got home, I tried to reconstruct as much of the blitz games as I could remember, mostly in order to improve the openings!

There were some real surprises waiting for me!
Let's go back to the move 13...Qd7

It turns out that after 13...Qd7? White can replay with 14.dc5! bc5 15.b3 where White keeps his b pawn who is highly restricting my knight and that White is much better, engine gives plus 1.5! I was curious how I played online in that line and I saw that 12 time I actually went for Qd7, and only two times for 13...c4!, which is correct move. Who has discipline he should also analyze his blitz online games!

There was another interesting moment in the game: on move 21...Rc8?, the engine claims this is a mistake because White could play 22.Bh6! or 22.Qg4, initiating a king attack. However, Bh6 is very much a "computer line", a long-term piece sacrifice, while Qg4 is potentially playable, because there is direct threat Bh6 to worry about.
Here is the game as a Lichess study:

I hope you enjoyed the game, that I’ve clearly explained the idea behind the title, and that we will start analyzing our won games with the same level of attention as the ones we lose!