The most simple way to get a fairly equal game without any trouble is to just go back to the 4-knights with 4. ... Nxe5 5. d4 Nc6 6. d5 Bb4 7. dxc6 bxc6 (back to basics now. This is the Scotch 4-Knights mainline with one more move played by both sides)
There should also exists some "refutations". But these are quite tricky.
Objectively it's not good and there are a couple of supposed refutations. It's not that simple however, this line is much trickyer to refute than it looks. The only time i faced this as black in online chess I was lucky to swindle my opponent from a losing position. I would not suggest spending much time studying it, since if you keep improving one day it will become insufficient for your purposes.
If i remember correctly you can download a nice article on this variation for free from the wikipedia page for this gambit.
The article:http://www.chessville.com/UCO/Halloween_Gambit/UON13HalloweenGambit.pdf
A funny point is that according to some authors the same gambit played a tempo down by black might possibly be sounder than the original version. After 1.e4 e5 2.Nf3 Nc6 3.Nc3 Nf6 4.g3 Nxe4!? 5.Nxe4 d5 white can not retreat his knight to g3, which is the most solid defence in the original version.
I've watched a good video on Youtube (thre's probably more than one) that explains some of the main lines of the halloween gambit. The line that Turm_Breuberg suggests is one of the best refutations that I've come across too. It leads to some pretty devestating attacks (mostly discovered attacks) but requires black to give up the dreaded fork on e7! Here is some stuff from the video:
I have heard of this Halloween Attack which is
1. e4, e5
2. Nf3, Nc6 (or Nf6)
3. Nc3, Nf6 (or Nc6)
4. Nxe5
I have played this in some of my games and figured a way to break even or gain positional advantage. However, I also heard that there is a way to counter it. If you know, please tell me the counter and tell me if this opening is good, thanks.