>Kevin:
>Try
wag.com (joke).
Wild
Ass
Guess.
com. I wouldn't put too much faith in anyone's "scientific" method for estimating software costs. It's like trying to explain/predict your teenager's behavior. The best of best WAGs would be try really really hard to come up with an accurate estimate then multiply by 2 (or maybe 3 depending upon how you feel at the moment). Then the tail end of your project will probably be wagging your estimates all over the place. I couldn't resist the corny mixed metaphors.
Sounds like Murphy's Law of Estimating:
Take your original estimate of time, then use the next highest unit and multiple by 2. So a 2 week job takes 4 months. Betcha that's closer to the truth than most would like to admit to.