It's the same thing. Just the latter implies you're identifying an assumption or two as variable and modeling outcomes of some of them changing. They're all forecasts. To argue otherwise is semantic pedantry.