the nature of game development is that sometimes you will need to multiply a physics number by -1 to fix it. you won't know why. you can either agonize over it all day and still not know why, or just surf that sweet, sweet -1 wave into the sunset
I have a Ph.D in computer science, a bag of SIGGRAPH papers and patents, and I still also encounter the "this must be multiplied by -1 against all reason and I don't know why."
Worse: in his classic book "A Trip down the Graphics Pipeline", Jim Blinn (the guy who invented Bump Mapping amongst a bazillion other things) also discusses how HE runs into this all the time; he refers to as "equalizing the number of sign errors." I know about this because John Carmack mentioned it once, and he also encounters this phenomenon and (at the time) just resorted to multiplying things by -1.
So, really, there's no hope for any of us at all.
