5

I played a game starting with

  1. d4 d5 2. c4 dxc4 3. Nc3 a6 4. a4 Nc6 5. Nf3

Black's next move is 5... Bg4 and it was considered by chess.com as

Inaccuracy (-0.28) A better move was 5... e5 (5... e5 6. d5 a5 7. e4 f6 8. e2 b4 9. g5 O-O 10. O-O)

enter image description here

But according to 365chess.com, the move 5... e5 is seldom played while 5... Bg4 is more popular with a high winning percentage for black. Why does chess.com consider 5... Bg4 as an inaccuracy?

enter image description here enter image description here

Zuriel
  • 8,749
  • 6
  • 39
  • 93
  • 2
    In general, you can't trust computer analysis of less then 1 point. Additionally computer analysis in the opening is very flawed. There are many lines the engine will say +0.80 and two moves into its main line the assessment changes to +0.00 – Ywapom Jan 11 '19 at 22:32

2 Answers2

6

Chess.com listed this as an inaccuracy because the engine is running at a limited depth, only looking at each move for a few seconds.

When I put the position after 5...Bg4 into Stockfish, it gave an evaluation of +0.12 at 16 ply. But at 18 ply, the evaluation dropped to -0.38. Most likely if you had run the analysis at "maximum" (which I realize is locked for non-premium members) the engine would have analyzed enough to know that 5...Bg4 was not an inaccuracy.

D M
  • 18,051
  • 39
  • 87
  • Thanks! How about the move 5... e5 then? Is it dubious? After all, it is almost never played. – Zuriel Jan 12 '19 at 05:24
  • I wouldn't call a move "dubious" just because it hasn't been played much. It's at least good enough that two masters out of 64 in your database decided to actually play it. It might depend on things like whether you prefer to trade queens, too. – D M Jan 12 '19 at 17:29
0

5...Bg4 isn't an inaccuracy, the chess.com engine just considers 5...e5 to be better. Deep Fritz, running at depth 23 on my computer, considers 5...Bg4 to be a slightly better move (giving a -0.31 evaluation). On 5...e5, it says that the endgame after 6.Nxe5 Nxe5 7.dxe5 Qxd1 8.Nxd1 is just equal.

Maybe chess.com's engine is better than my Deep Fritz, but that's not really the point. Engines consider one move in the opening to be a bit better than another all the time. It's important to take such differences with a grain of salt: if both moves are playable, choose the one that's easier/more practical to play.

In your scenario, 5...Bg4 clearly looks more comfortable than 5...e5. You're developing a piece and putting pressure on White's knight (which defends the d4-pawn). Meanwhile with ...e5, you're inviting White to push your knight offside.

Inertial Ignorance
  • 19,788
  • 22
  • 68