Following on from this post I tried checking the number of casual/regular players using this method:
(can only check for accounts, not individual players)
Carry out a player name tag search on a number of three-letter combinations chosen randomly (max result each time 99). Copy all the results together into a text file until a significant sample of 5000 accounts has been reached.
Then search for entries where "last played online" is in minutes, hours or days (therefore within last month).
Also get number of accounts online in last day (minutes or hours).
So get percent of players online within last month, and last day.
Multiply by total number of accounts (approx 1,300,000 obtained from awards data).
14.7 percent of accounts within last month = approx 190000 accounts (inflated by Xmas though)
3.1 percent within last day = approx 40000 accounts
This is much more than I would have thought, and means that Quake Live is a reasonable success.
I checked different samples other days with similar results.
Is there any error in this ?