Sicilian 3.Nd4??

I was looking at the Sicilian in the openings explorer, and found a line that has me thoroughly confused: 1.e4 c5 2.Nf3 d6 3.Nd4??

Even more baffling is that in the eleven times it was played (not by GMs, mind you, but 2000-2700 ranked players in actual tournaments) every single opponent replied to this hanging piece with Nf6?.

Here is a game where white left the knight hanging for three more moves before finally moving it out of the way to push d4:

What am I missing? 

The moves are probably just recorded wrongly. 

And?

I'll admit that was my first thought, but if you look at the games, the pgn would have to be completely made up for all the moves following Nd4. I don't have over the board experience myself, so maybe these sort of errors are common and I just don't know about it?

I think that the cause is because a player was moving fast and got ahead of himself.  The usual line is 1. e4 c5 2.Nf3 d6 3.d4 cd 4.Nxd4.  I made a similar mistake once:  1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 and here the correct sequence is 4...d5 5.exd5 Na5, which I've played dozens of times.  But I was on auto-pilot and played 4...Na5, a move too soon, which of course is disastrous.

It is also possible that after the slip, the two players agreed to transpose back into a main line rather than just throw away the game.

you got me laughed xd

It is also possible that it was a blindfold game and both players thought they were following book, i.e., that after white played 3.Nd4?? black assumed it was 4.Nxd4 and they were in the main line.  Maybe the ship got righted around the time white played d4... he tried to move his bishop and the engine said it was illegal, and he looked up at the list of moves and realized what had happened.  Pretty funny if that actually happened to both players.

Thanks for the responses. Whatever happened, I'll chalk it up as a curiosity and be encouraged with the prospect that I'm not the only one capable of blundering a piece on move three.