I dunno, King. On one hand, I very much agree with you that it's ridiculous that "anybody can profit from collegiate athletes' success except the athletes and their families". On the other hand, you can't simply dismiss rules by hand waving. On yet another hand, I also fully agree with you that his parent's living quarters in no way affects his on-field performance and shouldn't hurt his collegiate record.

Still, a price must be paid for breaking the rules, and it seems to me that the person who should pay is the person who most knowingly broke those rules: Michael Michaels. He knew he was screwing around. He should be barred from recruiting and/or pay a hefty fine. Reggie Bush and USC should get some sort of hand-slap, but keep their awards and wins intact.

Personally, I didn't think Stephen Colbert was all that funny at the podium. His timing was off: the audience didn't laugh, and he kept allowing time for the yuks to die down. Too bad, since he delivered sizzler after sizzler. Bush had this "deer in the headlights" look, and the audience was clearly unfamiliar with Colbert's persona. As comedy, it didn't work. As politcal satire, it was brilliant. Much better than last year's Cedric the Entertainer.

I have to give props to whoever in the Correspondent's Association chose him to be the keynote speaker. Maybe they figured that Bush wouldn't see beyond the faux bluster. I'm not sure Bush got any of the jokes, but he knew his ass wasn't being kissed, and Bush didn't know how to handle it.

CNN had a report on Colbert's routine Monday or Tuesday. Lou Dobbs and co. completely missed the point, as usual. As Stephen remarked to the assembled, "Reality has a well-known liberal bias." Too true to be funny, and the conservative news media desperately needed to hear it. Today's "journalists" are as clueless as a Bush.

I wonder if Bob Novak still has his blue stained suit...

What Manjoo fails to account for -- and therefore removes much of his objections -- is the claim that the 2000 election was also flawed. (And the 2002 election, by implication, as well, though that was not a presidential year.) His first point, about counties voting differently on state-wide elections than presidential elections, is dismissed on this point. He further doesn't account for the fact that after the 200 elections predicted an Al Gore win based on exit polling, the pollsters tightened up their polling methods for the 2002 and 2004 elections. Exit polling isn't the election, but Manjoo fails to counter Kennedy's observation that the Democrats surveyed were more likely to be reticent than the Republicans.

Robert Heinlein predicted that a religious, rightwing extremist named Nehemiah Scudder would win the 2012 election and there would be no elections after that, resulting in the Second American Revolution. He was only 12 years off.

Farhad Manjoo, replying to my comment on his article, asks, " I'd honestly like to know if you can tell me what about this explanation doesn't work for you."

What doesn't work is that your explanation falls short of any standard of "proof".

First of all, you don't answer my main objection to your article, that the 2000 elections were flawed as well. That was another instance where the best polling minds in the world called the election for Al Gore. When the dust settled, the guy with fewer votes managed to get into office. After that, the pollsters got together and tightened the procedures. Remember the Voter News Service? It's one thing to blow a prediction. It's another to be off by "as much as 9.5 percent " (according to the Kennedy article). You're comparing rotten apples to rotten apples.

Two election cycles after they went to work, the people who make their living on polls can't predict the outcome of an election after it's happened? One election? Maybe. A string of predictions over six years? Highly unlikely. (I don't believe Elizabeth Dole won either, but that's a different story.)

Second, you don't seem to counter Kennedy's high-level sources with anyone I trust. Kennedy cites Lou Harris -- 'Ohio was as dirty an election as America has ever seen,'' -- and John Zogby, who specifically answers your point about Mitofsky.

Third, your response isn't "proof", it's speculation. Your Mark Lindeman keeps saying things like, "it's possible" that Kerry voters in Bush strongholds "would" have been more likely to respond. The data, as reported by Kennedy, is that of a "careful examination of Mitofsky's own data by Freeman and a team of eight researchers." Exit polls are not "gospel", as Mitofsky points out, but surveys are used in almost all aspects of American life, and we're good at it.

To sum up my objections to the methodology in your article: You speculate about flaws in research but offer no proof, and the part of your analysis is to compare 2004 with the equally dirty 2000 and show how the exit polls were similarly wrong. We know Bush has no moral compass, and this is just one more disgrace.