I still think that not allowing ties in each judging category leads to big problems with the overall rankings after all 11 judges' scores are added up together. I'd like to hear what a statistician would say about this system. Say you have say 5 groups deserving a 9.4 in one category. It really shakes out to 9.2, 9.25, 9.3, 9.35, 9.4, 9.45 or worse. That .25 spread on what should be a tie is a big problem in tight competitions. Now if they all actually got a 9.4, would those 5 groups tie in the final score? It seems unlikely given there are 10 other judges' scores.
Trying to score the top 10 groups with only 21 points that range from 9.0 to 10 that does not allow ties is just too inaccurate. Doing that across 11 judges is really inaccurate. The only scoring I trust is the ranks of individual judges. The way the 11 judges add up to a winner does not seem accurate to me at all.