When 100 people with an interest in new music come together, their "top 20" aritsts are likely to be dicated by the choices of a minority. In the group Thoughtful about music in West Yorkshire
there are 109 members ... but last week's 6 most listened-to artists were scrobbled by 17, 12, 12, 10, 10, and 10 people respectively. So even leading choice Joanna Newsom
did not show up in the listening figures for 92 of the members. Our biggest star is a minority taste even in a group with non-mainstream taste.
Maybe the interesting activity is at the bottom end of the Chart - artists we have only listened to once? I use random play quite a lot, and I also use the last.fm radio function, so I don't listen to whole albums all that often. Having done a little bit of statistics training in the field of social research, I can see that it would be possible to use a variety of measures to create "ecleticism" charts - an expression of how dispersed individual listening paterns were.