pfren, if would be nice if you addressed the remarks I made to you in post #41. 
Is the Latvian Gambit sound?
Sort:
Nice! Thanks! Just copy/paste the PGN's of the last ones. No need to make fancy diagrams.
I know. I just thought it would have been easier for you. 
By the way, how did you arrive at the numbers you mentioned earlier (25 % and 40 %)?
Nov 17, 2011
0
#53
Morgado managed a draw with Black out of the two correspondence games. Sokolov scored a win with Black, and Hector and Rausis a draw each. 1/2 out of two is 25%, and 2 out of 6 is - woops - 33%. Sorry. I miscalculated.
That makes more sense.
And these numbers doesn't tell anything about the soundness of the Latvian Gambit, since it's not statistically significant. Try flipping a coin 8 times. You might get 8 heads. Does that mean, you'll always get that when flipping a coin. No, of course not.
Nov 17, 2011
0
#55
I don't know about "statistically significant," but the fact that GMs haven't played the Latvian Gambit against other GMs at all for the last 11 years seems to suggest that they don't have much faith in Black's chances.
pfren, on paper it maybe tell something about the Latvian Gambit that 3. Be2 might give White an edge. But in real play, it doesn't matter what this move could do, if no one plays it! Don't you see that?
Did you see my game in post #29? You could say my opponent blundered (with 14. d3), as 14. d4 would have been much better. However, I analysed this game with my coach, and even after 14. d4 (and without blunders), Black could win this game. So, saying that Black can only win with the Latvian Gambit, if White blunders, I do not agree with.
riga..., yes, I agree, it's seems they don't have much faith in the Latvian Gambit. But that's all we can say. We cannot say that the Latvian Gambit is losing at GM level, since we simply do not have enough data (number of games) to support such a claim.
Nov 17, 2011
0
#58
cigoL wrote:
riga..., yes, I agree, it's seems they don't have much faith in the Latvian Gambit. But that's all we can say. We cannot say that the Latvian Gambit is losing at GM level, since we simply do not have enough data (number of games) to support such a claim.
We can say in that in most of the games where GMs have tried it against other GMs, they have ended up losing (sometimes in just 12 or 14 moves).
(I might also mention that 2 GMs playing a game of chess is a bit different from flipping a coin. The strength of the moves they choose probably has an impact on the result of the game).
Yes, we can say that. But it tells absolutely nothing about worth of the opening. This is not my opinion, but has to do with statistics, and what is called the Law of Large Numbers.
Maybe the next 8 Latvian Gambit games between GM's will end with 8 Black wins. If so, the total score (in 16 games) will be approx. 30-70, rather than approx. 70-30 (in the 8 games you have posted).
So only 8 Latvian Gambit games, out of 250,000! Not exactly a popular choice.
By any chance you could copy/paste the PGN's of those 8 games here? Thanks!