Wow, I didn't mean to start a controversy. However, the questions are valid and I will address them as best I can. Some of the data is with students, so I will ask them about their numbers after school today. I would love to hear from engineers with expertise in bearings and loads as to what might be happening within both the ball bearings and the Igus. Perhaps there is already a thread in the archives about bearing loads experienced in a typical FRC competition? I will search for one, but if anyone reading this is aware of one, please post the link. My own expertise is woodshop...
The weight is less because there is less material in the hub itself, as well as in the wheel. We weighed both the complete 1.5 x 4 wheel assembly and the 1.25 x 4 assembly about 3 weeks ago. The difference was about 1/2, but we didn't write it down, I don't remember the exact number, and all our new wheels are either on the competition bot or the practice bot currently, while we gave our other wheels to another team. Sorry!
Lacking the two flanged bearings also removes about 40g per wheel. The hubs are spoked as opposed to the solid hubs that WCP/VexPro produces. Please click on the links to the two pictures I posted to see how these hubs differ from the WCP hubs. They are also about .1" thinner on the flange, which might not be clear from the photographs.
We initially built the practice bot with the 4x1.5's and were planning on using the same wheels on the competition bot. We purchased the assemblies from Banebots on a whim and, as I said, loved that they were lighter and gave us more space in the drive train. My efficiency statement was based on the current draw of the two robots. The two robots were constructed using the same types of parts (ball shifters, CIMs, pneumatics, etc.) but (obviously) have differences too, many of which are specific to the individual parts and systems, and would compound. You are right that I ought not to attribute too much of our results just to the wheels, but we did have a notable increase in battery life, along with fewer problems with the wheels. Again, I'll ask the kids after school what their numbers were. We have since switched out the larger Colsons on the practice bot and replaced them with a second set of the smaller wheels to make the driving experience as close as possible to the competition bot.
Without having the background to address the theory on this problem well, I will venture to guess that there are several factors that might make the Igus bearings work well. First, the axle bolts tend to have some warping (though we bought the straightest ones we could find). Second, the ball bearing/hub assembly could potentially have some misalignment. Third, the types of loads applied to a bot in competition might include some that would impact the performance of the bearings. Again, I'd love to hear from someone who knows more about this subject.
Returning to my original point: I really only wanted to say that we like these wheels, and your team might like them too. I understand that I threw out some statements that deserve scrutiny, and that no wheels (including these) are miracle workers. Thanks for your time and patience.