I've noticed this too over the last 2-3 days, coincidentaly (or not?) with Leviathan consistantly lowest.
I'm not incredibly versed in ping/latency/etc. but if the numbers on the site are the results of dozens of pings, and Leviathan is, on average, the lowest, wouldn't that strengthen the idea that Leviathan is the better server for OP and me?
If this is flawed logic don't hesitate to tell me why and elaborate... though I'm fairly unversed this is interesting to me.
Edited, Jul 9th 2013 10:57pm by NeptuneMagnus
Without collecting literally thousands of samples, you can't really generate an accurate model.
Since you know that all the servers are in the same place though, we can make some assumptions about what a difference in ping might mean. Odds are it's an aberration created by nothing more than how their network infrastructure is laid out in the datacenter. Either way, a 10ms difference isn't meaningful.