Why is this a blunder?

I played a rapid game today, and I was analyzing the game after I finished. I realized that Stockfish said my opponent made a blunder, 9. exd5?? This exposes the Black king's e-file and gives black an IQP structure, but is that really enough to give the evaluation a +1.5? What are your thoughts on this?


- Marlsen

cuz it opens the king file

@marlsensbudcreame It opens the e-file (and white can easily castle and play Re1 and cause black a ton of trouble due to black's lack of development of kingside pieces).

On the other hand, after 9... Bb4+! 10. Bd2 Bxd2+ 11. Qxd2, 11... exd5 is fine simply because black requires less tempos to castle and the dark square bishop that would go to e3 to attack the queen and then possibly to c5 to attack a pinned piece that has been moved to e7 after Re1+ has been traded away.

But then again 9... exd5 is quite natural of a move, so it's hard to get these positional aspects without careful analysis, possibly even with the computer.

Do not underestimate an open file which the king is on it. It is a hella weakness

I'll have to see the variations first.

Interesting. At move 9, SF says -0.7 (9... Bb4+) and then 9... exd5 gets +1.1 (depends on the params used). If +1.5 as your params gave, that is a drop of 2.2 points.

"A move that changes the evaluation by +1 is considered a mistake on an equal position, but not so much if the position was already won or lost. In this case, the evaluation swing must be more important to be considered worth reviewing."
From: lichess.org/blog/WFvLpiQAACMA8e9D/learn-from-your-mistakes

So the Lichess analysis marks it as a blunder because of the 2.2 point drop. That explains why it is marked as a blunder.

Now as to why it thinks White is +1.5 despite being a pawn down ... well SF can see way ahead and finds a material win. Your statement about the e-file being open is the underlying issue. That together with Black's lack of development.

Which is why I asked the see the variations. So many times people get on here to ask about computer evaluations and they all talk about it in a vague general way which never reaches any conclusions.

"A move that changes the evaluation by +1 is considered a mistake on an equal position, but not so much if the position was already won or lost. In this case, the evaluation swing must be more important to be considered worth reviewing."

lol

In my opinion, either developing move (Bb4+ or Nf3, both getting the king closer to safety) is better than wasting time with pawn takes (which isolates the pawn and opens lines with the king exposed as others have said) since White's pawn on d5 doesn't pose a threat at the moment (if it captures or advances, Black gets to place a bishop on a decent square).

@MrPushwood I think principle of not considering swing of 1 pawn in totally won situation correct. I have seen Magnus C games where he makes clearly inferior move - obviously with intent - to simplify situation. A winning move is a winning move counting centipawn differences is not important on those. Oviously when I make those kinda moves some them are intentional and better part of them are just that Idid not see the move. But there is no way for engine top probe my head.

As for concrete variations: yeah you need those but when weak players discuss with variations I doubt you could draw conclusion from there either :)

And general ideas should lead to decission which variations to analyze. Black should probably have rejected hitting central pawn very quickly on knowledge that having open lines with king in centre is bad and having open lines when behind development is also bad. So start working variationd that might solve those two issues Nf6 ja Bb4 look like ones I would start with. And picking just out of those two: it would totally random regardless amount calculation as the point of evaluation will not anything concrete enought that i coudl firmly judge it. Bb4 probably leads to me trading bishops and not sure I like it. as trading developed pieces does not help in being behind development