Well, I agree with them. No system will earn legitimacy if the judges are not accountable. Now that I'm understanding more of the technicalities of this system, I don't think there is any reason for the random selection of judge's marks. Each judge marks each element separately and the marks of each judge appear on a protocol, apparently. You just don't know which judge's marks counted in the point totals.

With these detailed protocols, it's entirely possible to use statistical techniques to analyze each judge's marks on each element and compare it to the others to uncover any unusual patterns or inexplicable decisions in marking. This actually does make it tougher to cheat, but only if you also know which judge's marks contributed to which point totals. (heehee ... have I lost ya?). Bottom line is that with so much easily quantifiable data available, random selection is redundant, IMO.