Another Decided Decides

Tim Cavanaugh | October 27, 2004

My old colleague Steven Johnson, onetime impresario of Feed and Plastic and author of Mind Wide Open, posits an electoral college winner who is also, believe it or not, his own candidate of choice. Prediction:

My gut right now is that Kerry will win fairly decisively -- not in the popular vote, but in the ultimate electoral college tally. It won't be Reagan/Mondale, but neither will it be 2000 redux. My best guess is Kerry wins 299 to 239: he picks up the big three (Florida, Ohio, Pennsylvania) and grabs Wisconsin along the way. Bush gets pretty much everything else that's seriously in play right now: Iowa, Arkansas, New Mexico, Colorado. It's enough of a margin that there's not a major recount battle, even if a handful of those states show less than a percentage point of difference. I wouldn't be surprised in the slightest if Bush won the popular vote, though I suspect Kerry will squeak out with that as well. So let's say the final tally is 299/239 and 49%/48.5% both in Kerry's favor. Check back in a week to see how I did.

At the moment, Electoral-Vote.com has a moderate Bush win, though the trend over the last couple days is toward Kerry. I am on record in various places as expecting a Bush win for very basic reasons, but prophecy was given to fools. My only prediction anymore: All this talk about another popular/electoral split is a lot of hooey. People are gearing up to fight the last war, and they're going to be disappointed. Whoever gets the electoral vote will also get at least the popular plurality.

So there!

Update: As the original Jason Bourne notes in the comments, Electoral-Vote.com's actual prediction page has a Kerry win. I was only looking at today's count. I missed the prediction page's count because my eyes were drawn to the word "Depress" in the button labelled "Depress for current map."

Comments

|10.27.04 @ 7:15PM|#

Tim Cavanaugh,

Electoral-Vote.com predicts a Kerry win: http://www.electoral-vote.com/pred/index.html

|10.27.04 @ 7:26PM|#

If Electoral-Vote.com has predicted correctly, then this will be the first time a Democrat has been elected President with a southern state under his belt. Wow, this sort factionalism (northern tier states & Pacific coast states v. the southern states and the Rocky Mountain states) reminds me (roughly) of post-bellum national politics in the 19th century, particulary the election of 1896.

|10.27.04 @ 7:28PM|#

Hmm, depending on whether I use JB's link by cutting and pasting or Tim's by clicking on his link, I get Kerry (the former) and Bush (the latter). What gives? My cache maybe?

|10.27.04 @ 7:33PM|#

Nevermind, I see the difference.

How are the assumptions going to hold up, though? Do we know all the reasoning behind them?

|10.27.04 @ 7:33PM|#

Lowdog,

One is the "prediction," and the other is how it "stands today." A lot of those states fall out of the Bush column based on how he thinks undecideds will break (2 to 1 for Kerry); in nearly even states like Iowa or Ohio, such a prospect means Kerry takes EVs out of the Bush column where Bush leads by a mere 1%-2% in the polls. Of course, such an assumption also assumes that the Bush lead isn't higher but not reflected due to the margin of error issue.

|10.27.04 @ 7:34PM|#

Lowdog,

Well, undecideds do tend to break for the challenger the way he describes. I don't know how strong his other assumptions are.

Nathan|10.27.04 @ 9:20PM|#

Rasmussen has a different take on how undecided break.

Of course, this is all just poll masturbation. The only one that counts is the one held next Tuesday. Until then its all just academic.

|10.27.04 @ 9:56PM|#

My prediction... Kerry in a landslide

|10.27.04 @ 11:04PM|#

My prediction: Kodos, by the narrowest of margins.

Tim Cavanaugh|10.27.04 @ 11:09PM|#

Of course, this is all just poll masturbation. The only one that counts is the one held next Tuesday. Until then its all just academic.

Oh thank you, Professor, for clearing away the fog with your hardheaded, no-nonsense wisdom. Here we all were thinking we were actually selecting the next president in a smokefilled backroom intrigue.

|10.27.04 @ 11:16PM|#

Doug,

Well, that would have pollsters scratching their heads for a while. I still remember that in 1996, a week or so before the election, Clinton was leading in the polls by 14%-16%; when it came time to vote, he won by 7% or so, and barely missed the 50% mark he (or at least his staff) so coveted.

Nathan|10.27.04 @ 11:16PM|#

Oh thank you, Professor, for clearing away the fog with your hardheaded, no-nonsense wisdom. Here we all were thinking we were actually selecting the next president in a smokefilled backroom intrigue.

You're welcome. You guys often need it.

|10.28.04 @ 12:55AM|#

I've given up predicting.

Oh, and as of Saturday, when I take custody of the ballots and voting machine for the precinct that I'm working, I'm going to abstain from commenting on the election until I turn everything in on Tuesday. Even though I have lots of checks and balances, I just don't feel comfortable expressing partisan opinions while I have custody of the ballots. Maybe it's just superstition, but that's how it is. So I guess I'll be restricted to the handful of pop culture threads starting Saturday morning.

Oh, and the neat thing about being a poll worker is that on Tuesday night, when all is said and done, I'm always so tired that as long as the outcome is clean I declare victory and go to sleep. Even if all of my least favorites win and all of the ballot measures go the opposite of how I voted, as long as it's clean I figure it's good enough.

So here's hoping for a clean result, one way or another.

|10.28.04 @ 1:53AM|#

"So here's hoping for a clean result, one way or another."

I thought you were a Democrat.

|10.28.04 @ 8:21AM|#

Considering how Bush lost the popular vote but won the electoral vote in 2000, I think it would be great if Kerry lost the popular vote but won the electoral college this time around; that way, both major parties would have an incentive to do away with the electoral system and allow Americans to directly vote for president.

|10.28.04 @ 8:36AM|#

You'd think that Reason would pay more attention to the market-based indications which say right now that Bush has the edge. Guess that was the "old" Reason.

|10.28.04 @ 8:45AM|#

Matthew Cromer,

That's not guaranteed any greater measure of success than the polling enterprise (in part because any market of political futures is necessarily going to be impacted by, well, polling - so you have an abstraction of an abstraction).

|10.28.04 @ 8:57AM|#

Electoral-vote.com has updated its main page:

http://www.electoral-vote.com/

Looks like Kerry picked up Ohio, but Michigan has moved to the tied column. Neither candidate (according to the web page's numbers) has enough EVs to win.

Prediction: http://www.electoral-vote.com/pred/index.html

|10.28.04 @ 9:35AM|#

Jason,

The election market evaluates all factors, including how much bias might exist in each poll, track records of the polls, etc.

The market-based approach has a better track record than any other method at predicting election results.

|10.28.04 @ 12:15PM|#

Matthew-

The Iowa markets only predict the popular vote. I know, I know, the popular vote isn't what counts, but the people who run those markets decided to sell contracts on the popular vote. Say what you will, but that's the product that they're buying and selling, so that's the future even that they're forecasting. People buying those contracts aren't predicting the electoral college, they're predicting the popular vote. In a very close election a split is more plausible. In 2000, if I'm not mistaken the Iowa markets predicted a Gore victory, and he did indeed win the popular vote. As we've seen, however, the popular vote isn't what matters.

|10.28.04 @ 1:36PM|#

Matthew Cromer,

The market-based approach has a better track record than any other method at predicting election results.

It does? Care to demonstrate that, because they were wrong regarding the EV tally in 2000.

thoreau,

Zogby predicted a Gore victory (popular vote wise).