The server is set the same.
What I'd do is the following (I don't know how many tests you've run), to avoid statistical or network congestion anomalies:
1. pick a server (set it on BOTH phones)
2. run a test on gnex, then on charge. do this 3 times, alternating between phones (reduces the risk of the network being less loaded vs. if you'd run it 3 times in a row on each phone
3. now pick another server and repeat step 2
4. do one more different server and repeat step 2
You'll then have 3 data samples for each phone on each server and if the trend is the same, you've got data you can show to the VZW store if you want to take the phone in to exchange it.
As an optional step, you could try the server(s) again
outside, because at least in my experience, being inside a structure definitely seemed to adversely affect the nexus' 4G (and even 3G) performance.
Please post the results here, as I'm curious (and I'm sure others are as well!).