The True Relevance of Polling Data

Bush will win. You heard it here first. I'll explain why I'm comfortable making this prediction by referring to polling data, but that's a pedestrian use of that information. I will offer my reading of these tea leaves to our Commander-in-Chief as a service to him, myself, and my country.

But first, why should Bush lose? I pass over his many faults, which can broadly be classified into two categories (1) his religious agenda and (2) his nurturing of the nanny state. After all, I am a single-issue voter in this election, and that issue is the war against the Islamofascists. Had the Democrats run a credible pro-war candidate who offered to fight this war more effectively, we would see America "changing horses in the middle of the stream." However, the Democrats didn't and America won't. Bush will win with a solid majority of the popular vote as Dick Cheney predicted earlier this week. He'll also garner a better Electoral College majority than in 2000 and perhaps a solid one, but not a landslide.

What are the why's and what about them are important?

I'll start by recounting some interesting poll numbers pointed out by T. Bevan on 10-27-04 at RealClearPolitics. A slim majority (51%) of Americans thought that the U.S. was mistaken to send troops to Iraq while a large majority (70%) think they have a clear idea about what the war in Iraq is all about. Bevan offers two factors that may contribute to the disparity: (1) "I suppose you could attribute at least a few percentage points to some hard core lefties who think they're quite clear on what we're fighting for and it's not something they approve of - like oil or American imperialism." I agree with him on that. Lord knows, that's all the Chomskyites here at work ever talk about. (2) "Maybe another point or two believes we are fighting to establish democracy in Iraq - which we most certainly are doing - but separates that particular task from battling global terrorism and thus concludes that establishing a democracy in Iraq is simply not worth doing." He's on the right track here.

I'd add that the execution of the war has left lots to be desired. First, consider this question I'd like to ask of Bush: "Why, exactly, is there still such a place as Fallujah?" Second, I thought we were going to stop negotiating with terrorists. We have chatted with the leaders of the terrorist "insurgency" however many times too many. (The correct answer is zero.) Part of the group who might approve of our presence in Iraq must be thinking something like, "If we're going to act like this in a place where the troops are in harm's way, we shouldn't be sending them there at all." In their book, it was a mistake to send troops, but it is not a mistake in principle to be there for our stated reasons. Further, you have people (like me) who can see a commitment to being in Iraq as part of establishing a non-Islamofascist beachhead in the Middle East. Bush's commitment to "democracy" in Iraq could lead to an Islamic Republic there. This will bother those people -- though it doesn't seem to bother Bush! Those of us who agree that a "forward strategy" is a good idea want a secular state in the Middle East. Failing that, we should run Iraq as we did Japan after WW II or even run it as a colony. The objective here is not liberation, though that will occur anywhere we run the show just as a by-product of us being there. The objective, Mr. President, in a war is to make the enemy unable to harm the American people. Furthermore, some would hold that an invasion of Iran would have been more appropriate than one in Iraq. (They have an excellent point, by the way.) They will also probably call being in Iraq a mistake.

The short-range meaning of this is bad for Kerry. As disappointed as lots of us are with Bush's execution of the war against the Islamofascists, we aren't ready to replace him with a committed pacifist who will simply not fight at all. The long-range meaning is this: start fighting the bloody war, Mr. President! And it can be bad for America. I'm not sure that the Democrats are capable of learning from a Kerry loss that maybe national defense is a good idea. And the Republicans won't learn that what Bush is doing isn't good enough either -- unless those who favor the war, but think Bush isn't aggressive enough, hold their feet to the fire.

To understand the short-term implications of the poll numbers, I invoke the Mystery Pollster's Rule of the Incumbent. The gist of his rule is that in a normal election, the important number in a poll is the incumbent's level of support. Thus, "the incumbent rule tells us that, at any given moment, the President's percentage of the vote relative to 50% is a better indicator of where the race stands than the margin separating Bush and Kerry. It also suggests the appropriate way to read the final polls just before the election (and these are my ranges – others may differ): If the average result of all the final polls (including undecided) puts Bush's percentage at 50% or higher, the President will likely win. If Bush's percentage is 48%-49%, the race is headed for a photo finish. At 47% or lower, the President will likely lose (add 1% to these ranges in any state where Ralph Nader is not on the ballot)." I'd also consider the job approval numbers as Mystery Pollster recommends. "[T]he two numbers I watch most closely are Bush’s job rating and percentage of the vote, as I believe these are most indicative of his ultimate support. That both are hovering just at or above 50 suggest an ultimately close contest, with Bush receiving just about the support he needs to win."

A Seat-of-the-Pants Guesstimation of Gut Voting

The serious discussion is over. I now go on to the obligatory prediction of election results. I credit an acquaintance with inspiring the term "gut voter." He was all set to vote for Gore in 2000, then entered the booth and couldn't bring himself to do it. He went with Bush. For the record, I refused to vote for President that year.

So here are my state-by state predictions. 70% of Americans think they understand why we're in Iraq and 51% think Iraq is a mistake. Let's say about half are pro-war and half anti-war. So a 9.5% of the electorate are like me: they want better than Bush's prosecution of the war. We'll say that about one in ten hasn't decided who to vote for or are even leaning Kerry out of disgust. I say that this 1% will enter the voting booth and let their gut do the talking. They'll vote for Bush. I'll go out further on a limb and say that the gut vote would make the higher number of Bush's stated support or his job approval be his starting point in the election. I don't have these numbers, so I'll fudge and just add 1% to his stated level of support. Another species of gut voters are unenthusiastic Kerry supporters. Blacks who still don't support Bush outright will alone easily make up about 1% in this category, the stay-at-home voter. In the interest of being conservative, we'll just call the stay-at-home vote 1%. We'll add 1% for Nader and other candidates. (So we have 4% total in fudge factors.) Wherever Bush exceeds 50%, he gets the electoral votes. We give the rest to Kerry.

[10-28: I had to correct the below for a math error. Having done this, it is apparent that Bush could be in trouble without at least one of the 50% states and/or the northern CD of ME.]

Here are the results My polling numbers came from electoral-vote.com about 7:00 pm Central Time on 10-27-04. My apologies for the lousy format. I hope to fix this when I have more time. The last number on each line is the number of Electoral Votes I project for Bush for that state. Electoral vote oddities could change the total as follows: Maine could give Bush an extra EV should he carry its northern Congressional District. A faithless elector from WV could drop him one. I show Bush winning in Colorado, so if their vote-splitting referendum passes, he'd lose 4 there. Interestingly, three states come in at exactly 50% using this projection (which my gut tells me might be a tad conservative): Hawaii, Ohio, and Wisconsin. By a slightly more generous estimate, then, Bush gets these and wins with 293 electoral votes. Without these, he has 269, the barest minimum he will need and then he'll need to hope that CO and WV hold up.

State Votes Bush Bush+4 Bush EVs

Alabama 9 53 57 9
Alaska 3 57 61 3
Arizona 10 50 54 10
Arkansas 6 51 55 6
California 55 44 48 0
Colorado 9 47 51 9
Connecticut 7 38 42 0
Delaware 3 38 42 0
District of Columbia 3 11 15 0
Florida 27 49 53 27
Georgia 15 57 61 15
Hawaii 4 46 50 0
Idaho 4 59 63 4
Illinois 21 42 46 0
Indiana 11 56 60 11
Iowa 7 45 49 0
Kansas 6 57 61 6
Kentucky 8 56 60 8
Louisiana 9 52 56 9
Maine 4 39 43 0
Maryland 10 43 47 0
Massachusetts 12 36 40 0
Michigan 17 44 48 0
Minnesota 10 44 48 0
Mississippi 6 51 55 6
Missouri 11 52 56 11
Montana 3 57 61 3
Nebraska 5 61 65 5
Nevada 5 49 53 5
New Hampshire 4 41 45 0
New Jersey 15 43 47 0
New Mexico 5 48 52 5
New York 31 36 40 0
North Carolina 15 54 58 15
North Dakota 3 55 59 3
Ohio 20 46 50 20
Oklahoma 7 61 65 7
Oregon 7 43 47 0
Pennsylvania 21 45 49 0
Rhode Island 4 36 40 0
South Carolina 8 57 61 8
South Dakota 3 55 59 3
Tennessee 11 53 57 11
Texas 34 60 64 34
Utah 5 64 68 5
Vermont 3 40 44 0
Virginia 13 50 54 13
Washington 11 45 49 0
West Virgina 5 49 53 5
Wisconsin 10 46 50 0
Wyoming 3 65 69 3

Total 538 293