Is the Latvian Gambit sound?

I'm so glad I've happened upon a bit of information to broaden your perspective surrounding this game you play so well, sir. It so happens that Louis Eichborn was the greatest chess player of all time. (Let's again ignore any statisticians hooha to the contrary) One may google at one's leisure. There are about 30 more-or-less riveting games with Adolf Anderssen as evidence.

Gentlemen, we must accept that the Latvian Gambit is unsound. Clearly Mr. Frendzas has proven it to us by word of mouth and 8 grandmaster games alone.

Or else... 

Perhaps you could persuade your friend hotchicken to play the white pieces for the benefit of any who may still have doubts, which would permit your chess.com account to remain unsullied by any victories and protect your friend's precious anonymity. He is, as you say, an e4 player, and so perhaps better suited to the task.

But who will play the part of Louis Eichborn?

What???????

Ah, thanks for broadening my knowledge- never heard of Eichborn before.

Anyway, the guy that played all the games under the "hotchicken" handle in ICC since 1999 is not me, but a stronger player, who hasn't played a FIDE rated game since 2006. He's a grandmaster with double citizenship (British/Greek), currently rated 2456 FIDE, and I can't force him to play if he doesn't wish so.

I would gladly claim that I played the ICC games myself, since I could possibly score worse than him, but the truth is that I was not the one handling the pieces.

As it goes for the only game I played white in the Latvian (that 1.e4 e5 2.Nf3 f5 3.ef5 e4 4.Ng1 weirdo), you can ask the inventor, mr.Buecker, about its vailidity. He is a very regular poster at Chesspub, and the chief at Kaissiber. I have played in the past quite a few of his weirdos, one of them costing me a miniature at the wrong side of the board against a woman GM.

cigoL, please take no offense. My remarks were tongue-in-cheek. And to no avail. It seems pfren will not be trolled into a demonstration, regardless of the odds he gives himself.

Well, well.

well well well look who's back in town.

love it!!

My thoughts exactly.  Well said.

I cannot disagree with this, and I never said otherwise. I only said that the stats themselves doesn't tell us anything. 

The Latvian will be playable again only if Black finds something to counter Leonhardt's 3.Nxe5 Qf6 4.Nc4! fe4 5.Nc3. So far, both 5...Qg6 and 5...Qf7 have a horrible record.

3...Nf6!? is playable, but I guess Black's ambitions are not playing an endgame a solid pawn down (at best!), while 3...Nc6, 3...Qe7 or any other third move are simply losing.

That was actually not my point (what I've highlighted in your comment), although I agree with this too. And no, there is no flaw in my reasoning. My point is simply that we cannot say anything about how well the Latvian Gambit (or any other opening) scores at GM level statistically, if we only have 8 games to back our claim. We might be able to say something about the Latvian Gambit at GM level using other arguments, but a statistical argument (like: the Latvian Gambit isn't a good opening at GM level, because it only scored 30 % in 8 GM games). This is not telling anything about how good the Latvian Gambit is. To say anything about anything using the numbers to back our claim we need more data than 8 games (as I explained previously with the coin flips). 

Well, as we both know, it doesn't really matter for us what GMs do (and don't do). 

I don't know if it's possible, by someone told me that some chess engines can play with a "rating loft". Then, it would be very interesting to have various chess engines play against each other beginning from a specific opening position, with no rating loft for one side, and the "rating loft" of the other side gradually decreased, until the higher rated side wins consistently, or until the lower rated side gets a 50-50 score. Doing this many times with various openings might be a way to tell something about the strength of the opening. Does this makes sense? If not, here's an example. 

Let's say an engine plays the Latvian Gambit against itself a thousand times, with no "rating loft". Let's say the rating (for both sides) is 3300, and the White side wins 60 % of the games. Now, we let it play a thousand games against itself again with a "rating loft" on the White side. We keep doing this until the score is 50-50. Let's say the score is 50-50 in a thousand games with a "rating loft" on the White side of 3000. So far, so good. Now we do the same with the Sicilian Defense. Maybe White wins 55 % of the games, and the score turns out to be 50-50 with a "rating loft" of 2700 (on the White side). This gives us a hint about the strenght of the opening, with the Sicilian Defense being a stronger opening than the Latvian Gambit - if my made-up numbers are what we'll see. 

When done with this experiment, we can try to set a "rating loft" on both sides, letting Black play at e.g. 1500, while trying various ratings for White, until we find the value where the score is 50-50. At this lower level, I'm guessing the difference in rating between the two sides will be less. If so, what does it tell us? That the opening choice means less at a 1500 level, than at GM level. Or at least, this is my hypothesis. If I had an engine, I would try it out. Unfortunately, I don't. If anyone does have one, I would be very curious to hear the result. 

I hope this makes sense. 

Ah yes... I see the reasoning. We don't know if 1.e4 f5 2.ef5 Kf7 is bad, as there are no GM games on it.

It might be a strong opening, since two 2300+ players have employed it OTB. And in 15 games registered in Chessbase Mega Database, Black has squeezed out six draws- not bad, eh?

And the Kadas Defence (1.e4 h5) has a slightly better score than the Latvian in the database. It must be a terrific opening, right?

And the Kadas Attack (1.h4) has scored over 75% in the hands of his inventor, mr. Gabor Kadas himself- the opponents including 4 GM's. I think we have found the new nuclear weapon for our white repertoire...

pfren, you seemingly don't get it. So, I repeat: statistically we cannot say anything about the Latvian Gambit at GM level. If you do not agree with this, then there is only one thing for you to do: take a beginner course in statistics. 

Oh, and this experiment would certainly tell something about the openings.

No.

I am new to chess.com and this Forum.  I recently ran into the following which I haven't seen elsewhere. 1 e4 e5 2.Nf3 f5 3.Nc3 fxe4 4.Nxe5 Nf6 5.Bc4 d5 6.Nxd5 Nxd5 7.Qh5+ g6 8.Nxg6 Nf6 9.Qe5+Be7 10.Nxh8 Nc6 11.Qf4 Qd4 12.Bb3 Bg4 13.Qxc7 Rc8 14.Qxb7 Nh5 15.Bf7+Kf8 16.Bxh5 Bxh5 17.Qxc8+.  It's the fifth move Bc4 that was new to me instead of the "book" Ng4.  PLaying this in Lichess, and getting Black's moves 5 to 7 endorsed by Stockfish, not sure how to improve.  For example, what would Black play at move 5, OTHER than ....d5? At move 8, Stockfish advised ...hg but then the game still went very quickly south on Stockfish's follow-up analysis.  SO, the question is, what does Black do against 5Bc4?  I haven't seen this analysed anywhere yet.  And if there is nothing better after % Bc4, what does Black do prior to that, to avoid being forced into that position anyway?  Hmmm...has anyone else come across this alternative 5th move by White?  If yes, what is the answer?

5.Bc4 is practically less good than 5.Ng4! which wins a pawn for nothing, but still a dangerous move which puts Black on the defensive.

The only wannabe alternative to 5.Bc4 d5 is 5...Qe7, but Black is lost after 6.d4, so there are no real options here.

The critical position arises after 5...d5 6.Nxd5 Nxd5 7.Qh5+ g6 8.Nxg6 hxg6 (8...Nf6? just loses quickly) 9.Qxg6+! Kd7 10.Bxd5, which was played 4 times in a recent ICCF thematic tournament (thematic tournaments are unrated, so correspondence player may well play any sort oif nonsense opening without paying in rating points).

Black is worse, but he may hold after tough defense.

https://www.iccf.com/event?id=66550

Thanks for link to ICCF games.  So, on basis that we are at very start of game - there has been no point, really, to veer off earlier, since we are only at move 5 - it makes White's Nc3 option challenging.  I will have a good look at the ICCF games that use it.

I once played the White side  in 2004 exibition game,

Taylor  Reinhart