I'm guessing most ISPs wouldn't have the resources to look into such an issue specific to one application, and many wouldn't spend the time if they could:
What we believe, based on a lot of analysis so far (and that analysis is ongoing, by the way) is that we're seeing a protocol bug in the Blizzard WoW code - a faulty interaction between the WoW client and the WoW server clusters. We've found a way to work around that faulty interaction. Its actually still happening, but with the configurations we're trying, it happens in a way that no longer impacts latency during game play.
As such, we are using our contacts into Blizzard (which we have, obviously, due to our extensive game server efforts over time), to see if they're able to address and (we hope) fix the issue.
This is one of the reasons why Whirlpool and other forums are filled with disturbingly obsessed Internode fanatics.
All timestamps are Melbourne time.