This reminds me of a quip by Frank Teuton (NM from Canada) when he fell into a big hole in ECO's analysis: 
Quote:

I remember the critical line being 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.d4 exd4 5.Nxd4 Nxe4 6.Nxe4 Qe7 7.f3 (instead of 7.Bd3) 7...d5 8.Bb5. Looking in the database now, this doesn't score so well for white, and indeed the strongest players are playing 7.Bd3. But I didn't look at any games in detail, so of course it's just statistics, caveat emptor, YMMV, etc.

Actually, this idea might be more relevant for black. How so? Well, after 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.a3 (Gunsberg Variation) 4...d5 5.exd5 Nxd5 6.Nxe5!? Nxe5 7.Qe2 was supposed to be tricky for black, because the "strongest" counter 7...f6 8.d4 Bb4 is disallowed by the pawn on a3. But it looks equal after 7...Bd6 8.Nxd5 O-O 9.d4 Bc5, because a2-a3 doesn't seem to make any difference here.

https://www.modern-chess.com/play-the-four-knights-practical-repertoire-for-whit...

^From the link above: We are happy to introduce another fascinating opening project by GM Pavel Eljanov - Play the Four Knights - Practical Repertoire for White. When sending the draft version of this product, Eljanov wrote, "I think it's my the most sound and practically valuable database so far".

I've liked the Scotch Four Knights, at least as a simple backup option, for a while now, but I was put off when this ...Bg4 line was discovered. This product seems interesting then...