I was reviewing some of the Speed Chess Championship games and came across a really interesting novelty from Nakamura.

[Event "Speed Chess Championship 2020"]
[Date "2020.12.09"]
[White "So, Wesley"]
[Black "Nakamura, Hikaru"]
[Result "0-1"]
[ECO "C65"]
[WhiteElo "2816"]
[BlackElo "2736"]
[EventDate "2020.10.31"]
[FEN ""]

1. e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 4. d3 Bc5 5. O-O Nd4 6. Nxd4 Bxd4 7. c3 Bb6 8. Na3 c6 9. Ba4 d6 10. Bb3 a5 $146

I found it hard to believe that Nakamura would play this over some more obvious lines like h6, Bc7 Qe7, and castling kingside even. As a result of this, I ran some computer analysis and it turns out that yes, at a depth of 34, 10. a5 is the best move for Black (using a private development engine at the moment so results may vary however several other engines picked up around a depth of 36). This leads me to believe that this was a pre-thought out position and had been studied by Nakamura and Chris Littlejohn.

My question is: Considering the depth required to analyze this position, and for 10. a5 to determined as the strongest move, the likelihood of the line playing out as studied is improbable to almost an impossibility. Is it really worth the effort?

  • It is. Just because it needs more depth on an engine it doesn't mean it's harder to play for a human as Black. Humans and machines don't think alike.
    – David
    Dec 10, 2020 at 7:19
  • That wasn't the intention regarding computer being harder etc, was just more that the position was obviously studied for some time,
    – Merenix
    Dec 10, 2020 at 7:38
  • Talking about depth without naming the engine is completely pointless, as depth means wildly different things in different engines. However with current Stockfish, a5 is shown on my computer after only a second, depth 23. Regardless, even if you needed depth 34 with current Stockfish, that is not high depth. Depth 50 is high depth, depth 34 especially with a very fast computer which most world class players have, doesn't take a whole lot of time.
    – koedem
    Dec 10, 2020 at 12:21
  • 1
    And in Caissabase, I was able to find 3 games where this a5 had been played previously (Carlsen-Aronian 2015 European Teams, Karjakin-Caruana 2015 Zurich CC Classical, and Karjakin-Aronian Superbet Bucharest Rapid & Blitz 2019 [in the playoff]). Thus, it wasn't exactly a novelty as far as I can tell.
    – pulsar512b
    Dec 10, 2020 at 22:27


Browse other questions tagged or ask your own question.