This antichess game: who is going to win?

I guess Watkins has solved it further as
12.Nxb6 cxb6 (forced, all others probably lose and it should be easy for a computer to figure these fast lines out. Simple lines such as axb6?? can even be solved as a one-line forced capture sequence by me) 13.Kxg2

Now we probably have reached what Watkins called "14 remaining lines" but I'm not sure.

This is a mystery he and people that do get his program run on their own computers know: It is in his latest 40MB file.

http://magma.maths.usyd.edu.au/~watkins/LOSING_CHESS/e3b6.html

One issue Watkins may have is that any non-forced move is hard to reduce since there are too many possible moves. Though Watkins has clearly established a white advantage, but white advantage != all forced white wins known

Who is actually interested in compiling his program and see how far he has actually gone besides me?

Guys, who will win from this point?
http://en.lichess.org/a1ZMI9AJ/black

(Quote from http://magma.maths.usyd.edu.au/~watkins/LOSING_CHESS/e3b6.html)
Please read Dr.Watkins' page for the 14 remaining moves.

I consider the last unsolved line of Watkins to be a fact, hence not protected by copyright but his program etc are his original work.