I'm under the impression that the original Barrowman formulas made a number of conservative assumptions, meaning that they might make something look worse than it is. I expect that the calculations that Rocksim does are a little more refined, and might be overall more accurate (generally speaking). I've also been led to believe that the Barrowman equations are best suited to fairly traditional shapes, and the more a design diverges from a standard configuration, the less accurate they'll be - or, perhaps, the more conservative they'll be. The Patriot is a pretty conventional shape, though, so it strikes me as odd that the calculations are that far divergent. I haven't come across that phenomena myself, most of the time the difference is minor. That said, comparing calibers of stability might be leading us astray here - what are the actual differences in predicted CP location, delta and as a % of the overall length?