A Three Blunder Game

I played this 2/1 game (and won) but from a quick computer analysis afterwards as I do with all my games in which no immediate game-ending mistake occurred, I was informed I had made three blunders.   I thought I had played a decent game so I was a bit taken back. I ran a second, longer analysis and this one said I made 2 Blunders (?). Puzzled, I thought to keep that result, shown below:


Here's the game with the Blunders computer-explained.  I'm still not sure they are Blunders. The one mistake I regret most was missing the mating sequence at the end.

I'm obviously not good enough to spot them myself - they show on the website but not on the phone app? Weird.

They are in the game annotations.

I just noticed - in the entire game I played no good moves.

I think computers have a bias towards defending material and see an attacking style game as having more mistakes in it. I think computers do not see that style of play as being good.

If I remember correctly, the computer told me I made a mistake in a game and then suggested a mate in 22 as if I could calculate that well.

 You would have to be Mrs. Gilbert for that.

https://www.chess.com/article/view/ellen-gilbert-queen-of-chess

 Good game

Not with 2, possibly 3 blunders.

Just a better game than that of my opponent.

It was a 2 1 game. The c3 blunder wasn't really a blunder, since they didn't capitalize on it by castling. Instead they tried to save the bishop, and you pretty much had a better advantage after 16 ..ba5 17 Qb5+ you have a better advantage than if you hadn't played c5.. Am a rookie so I think it was a good game given the time control.

Liked your article on Mrs Gilbert

I'm not accustomed to being confused by/not agreeing with any moves of mine that the analysis computer dubs as 'blunders,' as they all are usually pretty darn blunderific, so i envy you your dubious blunders.

Regardless of how obvious it might be that they've been made, it's never fun to see any number other than zero next to that stat...but what really twists the knife (and jiggles it around with a mocking laugh) is when i see the following annotation tied to it:

Even worse than the "MATE IN 2" is the "From winning to losing."  

Oh, the horror...the horror.

She was so amazing, the article wrote itself.

Just when we think we're being clever, the computer pulls the rug out from under us.

after 16. c3, I was expecting either 16...Bd6 or 16....Be7

Dear <Batgirl>

Ng5 wins the game on the spot. One threat is Nxh3, simply winning a piece; the other is Bh5, and f7 falls with resulting disaster for black.

It's all in the algorithm. When the absolute (not relative) difference between moves is great enough, the computer will label them as blunders. 

Your first move, Bg3, probably gives you a fine position, and in fact probably you're probably winning, since black is just wasting time. But notice that Ng5 completely crushes your opponent, with threats like Bh5 (and of course Nxh3), which turns out to be a massive advantage, the difference compared to your move makes it a blunder. 

This is very common, btw, when one side has a massive attack, and often the computer, if given enough time, will show that both moves are actually winning. 

That's what I mean about being clever.  I anticipated the weak threat ...Bh3 and saw I could not only defend against it but win back a pawn with Bxf4/Bg3... but so much for being clever. Even though it was a fast game, I never even bothered to look for a better move.

I wouldn't be bothered by it in a 2+1 game. 

When you have a massive lead in development, the computer will come up with all sorts of powerful attacking moves. It's enough for a human to find something good and playable. 

That's true - and in addition, it isn't so easy to see a move like Ng5 when you are facing a checkmate and have to respond within seconds...

(Pretty much what <Cherub_Enjel> said).