Tests are too weak: 3 fixed test, 1 quine test, and no random tests.

This kata is impossible to perform "correctly" short of writing a full blown JS parser. This should not be the intention.

Pretty awesome...

It seems your code doesn't take into account chained method calls. I tried optimising my code to f.toString().replace(...) but it kinda broke everything? Not sure why, but I guess it's related. Would be great if you could look into it, or look over my code and see what's going on there..?

function() {
  var a = 5; var b = 2;
  for (var i = 0; i++; i < 11) {
    if (i % 2) a = Math.power(a, b);
  }
  return a + b;
}

I don't know why this function got 16 points?
8 lines(8) + 1 if statament(2) + 1 method(4) = 14 point, where are the rest 2 points?

for loop = 2 points?