After reading the guide in this forum section on 'the best transcoding program', I got the impression that when you use deep analysis with Shrink, the compression using a variable bit rate algorithm but without the deep analysis, it is a constant bit rate. Is this correct? If so, of course it depends on the original dvd bitrate as well, but I expect you would find more scenes with pixelation and other compression artifacts after using deep analysis than if not. Without deep analysis, the overall picture quality would be lower but at the same time, you wouldn't have to deal with random points in the movie where artifacts show up.
This post was inspired by my recent burning of Family Guy DVDs using deep analysis. Obviously with animation, it is much easier to detect artifacts in the video. I figure just using a CBR may actually be better for the burn.