There was a great bit written by Ed (I think), who was working for Holset at the time, about water/methanol injection.
I can't remember the whole thing, but I can remember some of the points:
If you want the most from the water/methanol, you need to map for it or all you're doing is enriching...