d4, c4, e4, and Nf3 are all best. Everything else is markedly worse, albeit playable, except the Grob, which actually loses by force.
What is the best opening move according to engines?
Sort:
E2
FullTiltBunny wrote:
IronSteam1 wrote:
The top four first moves (1.e4, 1.d4, 1.c4, and 1.nf3) are all supposedly a draw with best play.
So it's somewhat pointless to wonder about which one is "best" - as they all theoretically lead to the same destination ...
It's even worse than that. The only move that outright loses is the Grob (1. g4). Everything else has lead to draws in NN correspondence games.
yep, Grob proponents have been on suicide watch ever since. Worse than even f3 - that's just cruel.
Still, just because engines of today say that maybe engines in 100 years will say Nf3 is actually the worst. Until the game gets solved (ie. never) it's all conjecture and trying to get a position that is understood by the player and causes problems for the opponent. I don't believe the grob is actually losing by force, common sense says that it couldn't be so early in the game.
Strong bots have no opinion but an opening encyclopedia which is not really an opinion but what the theory of openings says about of human knowledge based on statistics human vs human.
Afterwards if we make strong bots vs strong bots in rapid time (ex. 10, 15, 20m or 10m+5s per move) with no book, the bots "learn to play" compared to their opponents and it should be on the long term 1) d4