McPalin and the Electoral Map

Now that the conventions are over, the dust will settle in a few days, and we can see where we really are with regard to the Presidential contest, and most importantly, to the key state races.

The national tracking polls have re-adjusted to the Republican rebound coming out of its convention--they have the race back to a statistically-insignificant one-or-two-point lead for Obama (down from 6-8 in midweek)-- but the state-level polls largely have not. Thus they are showing a large lead--too large--for Obama. For example, 538.com, which works off a percentage probability outcome for each state and Monte Carlo simulation of results, is showing Obama up 310-228, and electoral-vote.com has Obama up 301-224 (with Virginia a tie).

When the state polls fully reflect a return to "normality"--or more accurately, the real starting point for the 60-day general election campaign--they will not end up back at the starting point before the conventions, which I would argue is where the national polls are right now. There will be some shifts due to the naming of the VP's and the course of the two conventions. For example, Delaware is more a lock than ever because of Biden, and Alaska (which had actually been showing itself quite close) should now be written off by Obama. Those are obvious, but there are some shifts, particularly reflecting the Republicans' adoption of scorched-earth attacks on Obama (except by McCain himself) and the McCain-Palin ticket (or McPalin, for short) as it emerges from the convention.

McPalin is a brand: it's feisty like McCain would like to be, if he hadn't squandered all his useful positions of distinction from the Bushites in the primary contest; it's warlike, down-home, it worships traditional values but is hugely vague on religious stance and ethical practice. It has no new initiatives but a fierce desire to define its legacy through opposition. It's at the same time feminist and sexist.

In general terms, McPalin will make the Republicans more competitive in the West and North Central. Biden and Palin will probably be a wash in the key states of Pennsylvania and Ohio. I have the feeling Palin will be a net liability in Florida and, more broadly, in New England and the Middle-Atlantic.

The Democrats didn't really do much in their convention to change the map. They strengthened and unified their own party, and made a credible play for undecided moderates and independents. McPalin did, too, as long as you exclude any of those undecided and independents who actually pay attention to policy positions.

McPalin and stoner(: The Blog)

I will make some revision in my base prediction, previously announced in The Wisdom of Two-forty-five. I hate to do it, because I am still backing that philosophy--get a reasonable route to just over 270 votes without Ohio, then work like the devil to get Ohio, too. In my view, it's what Campaign Obama still has as its basic strategy. However, the realities have moved, and my calls a little, too.

What I'm really predicting here is what the polls will be indicating going into Election Day. There are two or three big uncertainties which the polls will not have in them: 1) due to the "likely voter" approach of polling organizations, there may well be some underweighting of first-time and sporadic voters; 2) their use of landline phones--almost exclusively--will mean that cell phone users' without landlines will not be represented in polling results; and 3) the dreaded "David Duke Effect" (DDE), in which white racist sentiments will not be fully reflected in poll results due to concealment--intentional or unintentional--of those sentiments. 1) and 2) will tend to cancel out 3), but, like the conventions' effect, not evenly throughout the country. I should try to account for these better in my final pre-Election forecast.

The base-case prediction I had before was 293-245 Obama, with Obama picking up all states Kerry won, plus these states Kerry did not: Iowa, New Mexico, Colorado, and Ohio, The long and short of the following discussion is: move CO and OH to undecided, and bring Nevada (NV) over to Obama. That makes it 269-240, with 29 expected to be still up for grabs on Election Day. Two-sixty-nine, of course, is one vote short of Electoral College victory, so if McCain ended up winning CO and OH, and all the electors on both sides "kept faith", it would end up a tie and go to the House of Representatives, where Obama would win (he will have more states' House-member majorities in any reasonable scenario). A win, still, though an extremely nervous one.

Down to Cases

These are ranked in order of the greatest-likelihood of a switch in party from the 2004 outcomes, and color-coded for Bush-Kerry result.

1) Iowa (IA)--Obama's successful prototype organization from the primaries is still flying.
2) New Mexico (NM)--Hispanics have already come home, giving Obama a slight edge, so it's down to turnout and Albuquerque. McPalin will help on the margins with right-wing turnout; any edge to the Republicans from the large military vote (heavy in Albuquerque, but also in other parts of the state) was already in McCain's count before. This is not to say that McCain gets all the military vote.
3) Nevada (NV)--Palin's rudeness to Sen. Majority Leader Harry Reid (even worse, she botched the reading of the preface identifying Reid, as if to say "I don't even know what his title is, he's so insignificant") has been taken personally. So has McCain's advocacy of dumping all the "new-clear" waste in his neighboring state's Yucca Mountain. These look to negate a slight natural advantage--see also the comment regarding Hispanics under NM.

4) Ohio (OH)--My basic starting point is the observation that the Democrats gained electoral control of the state in 2006, so the question is whether McPalin & McCampaign can win it back. At this point, it's very much an open question; apparently, the state Republican organization has rallied around its cause.
5) Colorado (CO)--McPalin will revive flagging interest among the large evangelical community in the Colorado Springs area, which will counter the huge Obama organizational bump coming out of having the DNC there. It will be a turnout war, and the outcome at this point is uncertain.
6) New Hampshire (NH)--The only really strong McCain state in the Northeast, and thus McCain's best chance to pick up a Kerry '04 state. I don't think McPalin will help that much--most of the state's population lies outside the Mooseburger Effect Zone (MEZ--see below). It will be close, but Hillary will go to bat here for Obama with positive effect.
7) Virginia (VA)--The Democrats have the trendline in their favor, and I think Biden will help more than Palin in this state. The DDE may be small--Virginians of that mindset tend to wear their racism on their sleeve, rather than concealing it.
8) Michigan (MI)--McCampaign's big target for an upset, and one can certainly see reasons why, starting with Obama's snub of their unsanctioned primary and ending with Kwame Kilpatrick's sentencing. On the other hand, McPalin has largely given up on contesting the economic issue, except for promising handouts to car companies (but two can play that game!)
9) Florida (FL)--I am somewhat surprised to say it, but Florida is back in play. Charlie Crist didn't get the nod (I think it might have clinched the state for McCain), and I think Palin will turn off more security-conscious voters wondering which ticket is really more risky, than she will gain among confused soccer-mom undecideds (not much hockey down there). Florida will be tilted toward McPalin all the way but polls will remain Too Close To Call through Election Day. DDE may tip things to McPalin in the end.

10) Wisconsin (WI)--I don't know why this state is so close, but it is. So be it (as I suck up my courage and say): McPalin will try hard there, but fall short. As did Dubya, both times.
11) Missouri (MO)--Probably out of reach for Obama.
12) Pennsylvania (PA)--Obama has done everything right to put this state in his column. Republicans will be heartbroken again if they put heavy resources into winning it.
13) Minnesota (MN)--Was close, then Obama was pulling away. McPalin may cut it down (don't forget that the original home of Bullwinkle is Frostbite Falls, Minnesota!), but it still looks like a 5-7 point win for Obama.

14) - 16) Montana and the Dakotas (MT, ND, SD)--If one had an ESPNation-style poll on the question: "Mooseburgers--Sounds Like Food?" these would be among the contested states which could win a majority for "Yes". The polls were getting very close in these states (some had Obama ahead) prior to Palin's nomination, but I see the Republicans rallying to win these states in the Mooseburger Effect Zone. Also in the MEZ are Idaho, Wyoming, Nebraska, and Maine, but none of these are close enough for MEZ to make a difference.

17) - 19) North Carolina, Georgia, Indiana (NC, GA, IN)--These are close states for which DDE may end up being decisive. I wouldn't urge Obama to invest much in these states, except as a feint, over which I don't think will cause McCampaign to lose too much sleep. If they get any (from what I can see of the Obama campaign, they never do).

None of the other 31 states have more than a 1-2% chance of switching sides, in my view.

(This post's title links to 538.com, which along with electoral-vote.com and a piece by archdaemon Karl Rove in Newsweek, are references for some of the points in this one.)