so you don t end up with a situation where there is clustering at the toppe. four or five are above 5% and a bunch below that threshold. you are talking about 2% and you don t end up with a sheer lottery like the four top people in one night and not on the other night. you make sure there is a miss across the spectrum of polling on both nights. that s exactly right. that s why we are going to have in effect a draw of one set of participants and another draw. that s how we try to solve this goal we have. both nights, we will have great people on the stage. let me ask you a personal question. i have known you for a long time and for a while you were ....