Very interesting, thanks for the insight. I want to note that there is still value to this data. Although the coverage may not be accurate (covering all users),

Very interesting, thanks for the insight. I want to note that there is still value to this data. Although the coverage may not be accurate (covering all users), it sounds like it is precise (covering a very specific subset of users that access foreign sites, which StatCounter tracks). The fact that these users have moved so dramatically to Android in just 13 months says something. Is there a pro version of StatCounter that gives absolute figures, to see if there has been a dramatic change in tracked users over this time period?