Could 98% be an artifact of the 100's of millions of people who have tried social games? 2% is at least in the high millions if not low 10s of millions of people.

I (and it appears a significant number of others) ended up paying for the Economist - the 10 articles free / month was conditioning enough.