Unless big numbers are better, which is quite possible. But I think the latency has more effect, anyway, if I'm right about the meaning of the letter.
I noticed when we were doing the original bug test - the network traffic between two players increases somewhat when you're in close proximity with someone. Framerate/perceived grip loss, if it is happening (not the bug, but painful anyway), seems to me to be magnified when you're close to someone you have high latency with. Could be observer bias, mind.
My own connection (and this is likely true for other Australians) has 100-150ms more latency to Europe than US and Canadia.
Remember, it's not the ISP connection itself, but the whole route to the PSN server in question, or the other player(s).
I am just guessing, based a bit on what I know about our network connections in the past to assign meaning to those codes. They're just about the only easily spotted plaintext in the traffic, and they occur at timed regular intervals, unlike the driving data.
I'd like to see us try it with various hosts In a public lobby - more data
EDIT: It seems to use ICMP (ping) when setting up - some of the "router incompatibility" or NAT/DMZ issues could simply be no reply to these, due to security. That will in turn probably cause the mesh to have to try to get around it...
I haven't had a chance to check the capture for today properly yet.
Argh, more wall'o'text. Must stop.
More on topic, I didn't really have any hassles network wise. No practice under real grip online with lots of surface water more of an issue, shall we say...