No. At least I don't and I have never been instructed to do so. Normally there is a range that you score specific skills and execution so its pretty clear what scores they should be getting. =).
I don't think people are saying that judges are
told to hold back scores, or that judges
know that they're doing this, more that they may subconsciously do it, not to hurt the teams going at different times, but just because they're humans and they make mistakes! (:
That being said, I think that someone made a good point about the best team still winning. Assuming there was no favoritism or anything and scores were just held back, the best team
should still score the highest out of everyone, even if it's not a perfect score or all the scores are consistently lower than expected, the ranking should (hypothetically) be the same. However, when you then compare the results between divisions is where it's probably inaccurate because if they were holding back in division A but not really holding back in division B, Division B's scores will obviously be higher.
*that's assuming there's no favoritism and
only the holding back of scores. Which is probably not realistic.
I love the idea of playback w/ no sound. I feel like that would make the scores SO much more accurate. And playback for #s (difficulty) because I know I'm not a judge and I'm not trained, but I know there's no way I would be able to count while watching a routine and keep track of the numbers and judge overall impression at the same time. I would either be focused on #s or on impression... but maybe it's just because I'm not a judge?