It looks to me as if Black wants to prevent Ng5 following your Bc4, but from database Black playing h6 just increases White's chance of winning (more than 70%). Instead of h6, Black can probably play Nf6, Be7, or Bb5.
Strange line in the Scotch Gambit (4...h6?
Sort:
Yeah i've faced this strange h6 move too. Nxd4 is right and just play it like a scotch with an extra tempo. Easy game ahead.
Still, Ng5 isn't even a good move that soon :D.
Clouseau741 wrote:
I really don't understand what you don't understand.It's a perfectly reasonable move(I am not saying it's good) especially for a beginner or for someone that doesn't want to use databases or read books.It stops the obvious threat of Ng5 that white has.
Other than that, don't assume it's a horrible move.It gives only a slight advantage to white and nothing more.Of course there are better and most ambitious choices but if the player with black doesn't mind to defend, it's quite playable.
The fact that it "scores" awful at databases says nothing but unfortunately very few know how to read a database and most think(like Furtivamente) that because the database says 70% win for white , it's an awful move that gives white almost a winning advantage, far from it.
It's a good move , that avoids heavy analysed lines and there is no way for white to get more than a slight advantage from the opening on this line.
I don't understand it because there are tons of other moves that you can use to sidestep it. I never said I was an understanding person ;).
Actually, I often even get Nxd4, Qxd4 after that.
Your examples are very bad. In the French the center is closed, so tempo's are less important, and those kind of moves are playable because they have an idea behind them.
In the original position however h6 has no good idea behind it, and it is just bad for reasons already given.
In a Scotch Gambit/Italian Game type setup, Black doesn't have the luxury of time to play a prophylactic h6. He's already behind in development and control of central lines. Throwing away a move is just dumb.
But if anyone wants to post more irrelevant diagrams from the French Defense, feel free. There's nothing like a long non-sequitur to aid in understanding.
I agree with Clouseau741. There is nothing wrong with 4... h6. It prevents Ng5, and in the long term, prevents back rank checkmates and a bishop from coming to g5.
It is a prophylactic move against white's bishop and knight, and since the position isn't very tactical, it is okay to make a few quiet moves.
Sure, it doesn't develop any pieces, but it does make developing harder for white, which is almost as good. It also takes white away from the mainlines, forcing him to think for himself.
Quiet moves are perfectly okay in closed or positional games.
You're playing on a free chess site... likely you're facing a few beginners who are afraid of their f7 square being attacked and they reason that h6 prevents Ng5.
Otherwise it's a totally unexplainable move to see over and over 
Well I think that all of this can be summarized (correct me if I'm wrong) as:
- 4...h6 is playable in the sense that it doesn't lose by force
- There are better tries for black then 4...h6 but if you insist on playing it you won't immediately lose.
@Clouseau741 actually I was not saying it is necessarily a weak move, but I am suspicious that those who play it are weaker players, thus the horrible results, like what @ajedrecito said. I am not a strong player myself, but I won't play this move, though I have to admit I don't expect Ng5 coming so soon anw.
In fact I agree with you that with ideas behind it, a seemingly weaker move is playable (one is the surprise effect). I mean, Miles won against Karpov with 1 ... a6 which is seemingly weak. I also learn from tablebase, but I don't simply pick the statistically strongest moves from it.
I think it's too early in the game to judge the move as horrible.
Please do, it has become tedious and is unproductive.
In case the point weren't belabored enough...
I've encountered this:
So far, the most qualified voice on this issue has got to be Estrin and I will take his word for it without bothering to work on it myself or to crank up the BEAST. There is a big risk in trusting data bases too much because the larger databases have too many "trash games " included in them. The funniest example in my own database is in the Spanish opening after black plays 3...a6. There are 2 games in which white plays 4 Bxa6 ?? LOL Whats even funnier is that white wins both games so whites score with 4 Bxa6 is 100 % !! 
There is this line I get in the Scotch Gambit that I get 90% of the time.
Black almost always plays h6 against me. I then transpose into a normal scotch game with a free tempo. Am I doing this right? Or is there some line I am missing.
And why does everyone play this line? There are much better ways to transpose from the scotch gambit then by giving up your pawn and giving away a free tempo.