I think several people her have fallacies with mecanum. Our team decided week one to use them this year, and we did spend a little too much time perfecting it, but in the end it turned out for the better. We were a feeder station/shooter and we found the mecanum great for lining up. This was the problem we had last year. Even with all the summer practice I got with last years bot (hours upon hours) it still was not all that great. So our team debated week one and saw the omindirectional system as very advantageous, versus the challenge last year. Our original intent was to dive under the pyramids and such, but we never got to end up doing that because of the slight variations in pyramids. However, during competition when our shooter failed, we were plenty good at defense. (Youtube video. Not sure how much defense is shown..
http://www.youtube.com/watch?v=3M_YGV9kKP8). We actually pushed several teams with the kitbot around. We did find it easier to drive aroudn them though. We discovered (all too late), that speed would have definitely helped. I think a lot of teams underestimate speed. When are robot can drive around you, or simply quickly out maneuver you with a mecanum, it really makes a lot of things easier. Our teams has nothing bad to say about our mecanum, just make sure that you use a gyro to help with all the subtle problems. We used 6" AM mecs with TB minis
Also forgot to include that when we decided to add a climber, rather than messing with other systems, we were able to mount our climbers on the side and strafe onto the bar