Wednesday, December 31, 2003

previous entry | main | next entry | TrackBack (0)

Being Andrew Sullivan -- day two

Early this morning: I’m chagrined to see this morning that although Blogger says everything is hunky dory with my posts, I can’t see them on the public website – which is what led to the post below.

Chagrined is too weak a word – I’m freaking out. If I break a glass in my house, I’m slightly annoyed. If I break a glass at someone else’s house, I’m mortified beyond belief [So, um, how often do you go on glass-breaking binges?—ed. Come closer and find out]. Andrew’s been generous enough to loan me a valuable piece of real estate on the web, and it appears as though I’m letting my dog piss all over his lawn.

A few e-mails and calls later, the source of the problem is identified – is being upgraded to a new server, and the transition is proving to be a bit bumpy. The remote DNS servers around the web are taking their own sweet time to process the change in IP addresses. So, it’s not my fault (relief). It's not Blogger's fault either (surprise).

I see Josh Marshall taking Dean to task, and decide it’s worth posting. Dr. Dean is beginning to remind me of a scene from The West Wing when President Bartlett dealing with the fallout from concealing is case of remitting, relapsing multiple sclerosis. He admits to his press secretary that he never revealed his condition because he never thought he was going to win, so why bother? I've got to think there's a part of Dean that's shocked about being the front-runner.

I’ll admit to some reluctance about going after Dean too much, however. I’ve been hitting him pretty hard as of late, even though I defended the candidate over the summer. It’s not that I disagree with anything I’ve posted. It’s just that there are diminishing marginal returns to this kind of criticism, and I don’t want to sound like a broken record. Plus, Dean’s still got the nomination locked up – and if/when he wins, expect to see a passel of stories about how he’s acquired an invulnerability to media criticism.

11:00 AM: I get a bunch of e-mail in response to my political predictions post saying that a Yale economics professor, Ray Fair, had the 2000 race as close to even. I consider amending the post, but then I see that his 2004 model has George W. Bush winning 58.3% of the vote in November. That seems a tad high to me, even if Howard Dean is the nominee. I make a mental note to single Fair out if I'm wrong, though.

2:00 PM: One of the most useful articles in political science is a 1984 American Journal of Political Science essay by Mathew D. McCubbins and Thomas Schwartz entitled: "Congressional Oversight Overlooked: Police Patrols versus Fire Alarms," McCubbins and Schwartz's argument is that actors who seek information can gather it in two ways: actively seeking it through aggressive searches (police patrols) or trusting interested parties to get the information to you (fire alarms).

Going through the e-mail, there’s no question that big blogs attract a lot of fire alarms in the form of e-mail links. Does this make it easier for big blogs? Yes and no. There’s some good nuggets of information – the links to the Michigan and Conference Board projections for 2004 came from an e-mail. However, there’s a hell of a lot of chaff to go through as well.

On this post, I also link to the Easterbrook book, but I'll admit to wavering. I've been a big fan of Easterbrook's policy analysis in the past, particularly this TNR essay that's a key component of the new book. Last week, however, I made the mistake of linking to an Easterbrook post about the environment when it turned out he'd screwed up an important fact (he has yet to correct it). In this case, however, he appears to be standing on the shoulders of other researchers, so I go with it.

4:00 PM: That David Adesnik -- he thinks that flattery will get him a link. We'll show him!! Oh, wait...

5:00 PM: Ashcroft recuses himself from the Plame investigation. Post on it or take son to bookstore? Survey says... bookstore!!

posted by Dan on 12.31.03 at 12:30 AM


Not that I'm assuming you put a lot of faith in the Fair's model but..... Ray Fair's model is a bit off. It is inherently built to Bush's favor, in fact the only way under the model for Bush to get less than 50% of the vote next year is if real per capita GDP falls by 10% over the next three quarters.

Now if the economy were to retract a total 3% during the first 3 quarters I would find it to be astounding and I can assure you that Bush's re-election would seriously be in doubt. Fair's model would have Bush getting re-elected by almost 55%.

posted by: Laddy on 12.31.03 at 12:30 AM [permalink]

Wow, that's interesting. Fair's model is apparently an empirical fit of some sort -- does this suggest that the empirical really says Bush (or rather the incumbent) has that much of an advantage out of the gate?

posted by: Charlie on 12.31.03 at 12:30 AM [permalink]

I tuned in late, but see (day 1) that Andrew gave you permission to do anything. I personally find the color combinations of Daily Dish difficult. White on black or dark blue is so hard to read, I have to hurry to the link.

posted by: Norma on 12.31.03 at 12:30 AM [permalink]

"Wow, that's interesting. Fair's model is apparently an empirical fit of some sort -- does this suggest that the empirical really says Bush (or rather the incumbent) has that much of an advantage out of the gate?"

No, it says that Fair's model is not selecting the correct variables (or is taking an incomplete set) in order to create his fit. It also points up the fact that there are so few elections that any empirical model is highly questionable. The important variables shift significantly in even one election cycle, much less in 10 or 12 cycles.

So it's trying to create a curve fit to a moving target on the basis of very limited data, which is something of a fool's errand. In every election, there are likely to be unique factors that no model will adequately capture.

posted by: Doug Turnbull on 12.31.03 at 12:30 AM [permalink]

I don’t want to sound like a broken record

Coulda fooled me. (See: endless Plame posts)

posted by: Eric Deamer on 12.31.03 at 12:30 AM [permalink]

Dr. Daniel,

I think the situation of Dean Master nomination is now only a matter of time (early March). So, it may be good to have a Daily Dish sub-section focus on Dean Watch. I think his flip-flops (e.g., Discovering Jesus years after a bikepath fight, Overthrowing Saddam during Clinton years and today, Closed energy meetings in VR).

I think it is sad that even as we move to 2004, the World's most perfectable Democracy (i.e., there are flaws but these are all workable) has yet to focus on becoming the World's most open and participatory democracy.

I yearn for that day, sadly because of liberal-hypocrites this is taking a long time...

Ali Karim Bey

posted by: Ali Karim Bey on 12.31.03 at 12:30 AM [permalink]

Not to worry, Daniel, the problem with the blog isn't you, I don't think. is completely down at the moment, and has been so apparently since sometime last night, so far as posting goes. I expected something to show up on my personal blog at about 6 pm yesterday and it didn't happen. They appear to be completely trashed at the moment.

Elkhart Indiana

posted by: Dave on 12.31.03 at 12:30 AM [permalink]

See as I made a note of Fair to you, I'm willing to place a friendly bet (with any interested party) that Fair's last (sequential) prediction will be within the margin of error.

I'd also note that -- not claiming his model as inerrantly correct -- I simply claimed that he was correct in that case, when you had said none were.

Oh well.

posted by: Mikhel on 12.31.03 at 12:30 AM [permalink]


The abbreviation for Vermont is VT, not VR.


I concur, light text on dark background is bad.

posted by: Gary Utter on 12.31.03 at 12:30 AM [permalink]

“In every election, there are likely to be unique factors that no model will adequately capture.”

Yup, that’s most certainly the case. I’m not sure what to make of Ray Fair’s election theories based solely on economic factors. Needless to add, I hope that he’s right about President Bush’s reelection chances. My gut instincts tell me that the current incumbent is favored by an eighty percent margin. It would be even higher if the liberal media were not out to damage him. A Democrat candidate always has this single factor in their favor. The “elite” media are always on their side!

The President presumably has little to fear regarding the economy. It’s the terrorist issue that remains the wild card. All bets are off if the Bush administration might be blamed for a major terrorist attack on American soil. And there is always the unlikely possibility of someone finding a recent photo of President Bush enjoying sex with a farm animal.

The Democrats are their own worst enemy. They don’t have any realistic chance of selecting a 1992 Clinton style leader. The left of center has a lock on the nominating process. Joseph Liberman’s chances are less than 5%. Howard Dean is the clear favorite with Richard Gephardt having an outside chance. Karl Rove is probably laughing his rear end off. It appears that he will be collecting some easy paychecks.

posted by: David Thomson on 12.31.03 at 12:30 AM [permalink]

Post a Comment:


Email Address:



Remember your info?