<$BlogRSDUrl$>

Maintaining Optimism in the Face of Reality. Occasional observations on the state of the world, society, business and politics. Usually anchored by facts, always augmented by opinion.


Two Days Out, The Money's Still On Bush  | e-mail post

I traded some e-mail with someone who seems very concerned about how the election will play out. I thought I would try to put some information out to ease his mind, and hopefully some other people's minds.

First, let's look at what's locked up: Bush has 191 electoral votes locked up and Kerry has 149. This leaves 198 up in the air. Let's look at a few different metrics.

I'll start with the money. I have already written about the wisdom of looking at election markets and gambling sites in "Election Outcome: Like Deep Throat Said, 'Follow The Money'." Tradesports takes action on the state-by-state results, the electoral vote count, and the basic "who wins?" bets. On an electoral vote basis, the action breaks down like this:

State
E Votes
Bush OddsBushKerry
Arkansas683.0%60
Colorado975.0%90
Florida2759.0%270
Hawaii420.5%04
Iowa753.2%70
Maine47.6%04
Michigan1724.1%017
Minnesota1035.5%010
Missouri1183.4%110
Nevada574.0%50
New Hampshire431.0%04
New Jersey1510.1%015
New Mexico563.0%50
Ohio2050.0%00
Oregon711.6%07
Pennsylvania2125.1%021
Washington117.0%011
Wisconsin1043.1%010
West Virginia585.0%50


Pickups75103


Solid191149


Total266252

So, not quite enough to put Bush over the top, and Ohio is a true wildcard. The one state with even money between the candidates. Note that the wild percentages indicate what you would expect, if Bush has a 10-point lead in the polls, he should have very high odds, as it is winner-take all on the gambling front.

Tradesports has Bush winning 270 or more electoral votes at 52% and had Bush favoring Kerry for reelection at 55-45.

The Iowa Election Market also shows Bush favored for the winner-takes-all contest, at a similar level.

Let's look at other polls. Anyone who reads me regularly knows I am not a fan of Zogby and that I sound like Dustin Hoffman's Rainman talking about Qantas regarding Mason-Dixon polling. So, what does Mason-Dixon say about these races? Well, they don't cover all the races, but most of the races they don't cover are expected to go to Kerry, so, being charitable for now, I will concede all of those races to Kerry. Here are the results from the most recent Mason-Dixon polls:


E VotesBushKerryUndecidedSpreadBushKerry
Arkansas651%43%6%860
Colorado950%43%5%790
Florida2749%45%5%4270
Hawaii4



04
Iowa749%44%7%570
Maine4



04
Michigan1745%47%6%2017
Minnesota1048%47%5%110
0
Missouri1149%44%6%5110
Nevada550%44%4%650
New Hampshire4



04
New Jersey15



015
New Mexico549%45%5%450
Ohio2048%46%5%2200
Oregon744%50%5%607
Pennsylvania2146%48%6%2021
Washington11



011
Wisconsin1046%48%5%2010
West Virginia551%43%6%850





Pickups95103





Solid191149





Total296242


According to Mason-Dixon, things are looking very good for Bush. An important thing to note is that Bush has appreciably larger spreads in states he is winning, other than Ohio and Minnesota, while Kerry only has a healthy spread in Oregon. Kerry's other states are very close.

Minnesota is probably the most in contest, with polls showing a very close race, but a recent Humphrey Institute poll shows Bush up 47 to 44. Of course, the Star-Tribune has Kerry up 49-41, which is simply ridiculous. No poll has shown an 8-point gap in Minnesota, since, well, since the Star-Tribune had Kerry up 9 in early September. Twenty total polls between then and now, and only the Saint Cloud State poll is close to that (and sorry SCSU, but you're no Quinnipiac or Rutgers in the polling business). I've written about the systematic bias of the Strib, back in September. It will be a squeaker, here, that's for certain, but it is still competitive, and the fact that both candidates probably know their way around here like natives after the past month is evidence of that.

Ohio is the wildcard in most of these equations. It's hilarious, actually, the papers there are going nuts on polling. The Cleveland Plain Dealer did a 1,500 likely voter poll (showing Bush up 48 to 45) and the Columbus Dispatch did a 2,880 likely voter poll (showing a tied 50-50 race). Even John Zogby shows Bush up 49-44 in Ohio, but I think he might be bulking the numbers up in the hopes of driving Democratic turnout in Ohio, it's the largest Bush-favored spread in a while. I don't have enough data on the Columbus Dispatch poll, but it is very odd that it's a dead heat for Bush, while presumably correlated elections are blowouts. Republican Senator Voinovich is leading his race 62-38 and State Issue 1 (which is both a ban on same-sex marriage and civil unions) is leading 63-37. (One thing I question about the Dispatch poll is the lack of an undecided component.)

I feel pretty good, because I think Bush will likely shock everyone by bringing New Jersey home. If that happens, as I've said, we can all go to bed early. If Bush wins New Jersey, and the other races play out as predicted, Ohio and Minnesota can still go to Kerry and Bush still sails into another 4 years at 281 to 257.

Tomorrow's the last day to explain the facts to people though. Look at some of my posts over the last week to help people know the facts ["Last Chance to Get Some Facts Out"]["There is No Ban on Stem Cell Research"]["The Artisan v The Idealist"]["Nothing to Fear But Fear Itself"].

The level of misunderstanding in this election season, combined with high voter turnout among a lot of, honestly, politically ignorant individuals, is the biggest risk Bush has.

Updates: Federal Review ran some Monte Carlo simulations it would appear, and with 20,000 trials, they assign a 70.4% probability to a Bush victory. Data Seers seems to have done a similar thing, but came up with Bush at 87.8% probability of victory.

e-mail post | Link Cosmos | [Permalink]  |  | Sunday, October 31, 2004
Comments: Post a Comment

This page is powered by Blogger. Isn't yours?