Ok - I'll stop waxing philosphically... How about I point you to some links:
http://www.paralite.net/plastic_louvers.htm
Indicates a 75% light transmission from plastic eggcrates - ie: 25% loss of light.
Here's an interesting one - it talks about parabolic louvers - the eggcrate's bigger brother:
http://www.eclipselightinginc.com/pdf/Parabol.pdf
Look at lumens across all angles in LSI test 18705 and 18706 and LSI 18746 and 47 respectively which are the same bulb but with a specular parabolic louver and without...
I'm not an expert by any means - in fact I know very little aside from what I just read for a half hour or so. But in my 30 minutes of digging I couldn't find ONE "eggcrate" manufacturer that stated that their products increased light output. It seems to me that they'd be jumping up and down to say that... What they were saying was that it dispurses the light more evenly and the effective light at the wider angles from the bulb is more... But not that the output of the fixture itself is increased. I suppose if you were running flouresent bulbs with no canopy, the light that would have been reflected onto the floor, would potentially be pushed straight down into the tank - but I'm not sure I know of anyone that runs that way... Basically, I would think that the 25% loss that you took from the shading caused by the eggcrate would far outweigh any potential benefits of a slightly more focused beam... But I could be completely wrong.
Sorry if this get's on your case Anthony.>> I guess your "make us proud" comment or potentially your plea for me to stop bantering incited me to stop in and do some digging. My sources aren't exactly scientific papers, but google felt they were important
For what it's worth - I did go out and put a piece of eggcrate under a double NO fixture just now and visually I noticed LESS light coming through. But then again - my eyes are not light meters...
Hope you take this the right way - I've just never been the type to take something that to me sounds a little hard to swallow and believe it.
After doing some additional reading - lots of it.. and doing a couple visual experiments (albeit very late at night

) I have a tough time agreeing with you.
Is that okay?
Sorry man.