Marginal Voters

The Baseball Crank ponders the election statistics, and unearths a new interest group - "Marginal Voters for Bush".

Our reaction - first, the author is a baseball stats guru who is ducking the news about Derek Jeter's Gold Glove. End the cover-up!

Secondly, very interesting. I eagerly await his follow-up state by state numbers, since I have been puzzling over a few questions myself:

(1) Did Karl try to turn the crank everywhere for Rep voters, in order to get the cosmetic benefit of a popular vote victory? A comparison of battleground with non-battleground states might be interesting. Or, did Bush show lots of new votes even in states he won solidly in 2000?

(2) Did the gay-marriage amendments affect turnout? Was the increase in turnout in amendment states higher than we might have expected? This is tricky - how might one measure expectations? A battleground state like Ohio that also had an amendment on the ballot will surely show an increase in turnout, but there are two obvious reasons for it.

UPDATE:  Paul Freedman, writing in Slate, rejects "the gay marriage myth":

These [gay marriage] measures are alleged to have drawn Christian conservatives to the polls, many of whom failed to vote last time. The theory is intriguing, but the data don't support it. Gay marriage and values didn't decide this election. Terrorism did.

MORE:  Skepticism round-up by Glenn.

Comments

[Gerry]

I believe that the 72 hour plan was put in place everywhere, not just in the key states.

It relies heavily on volunteers, so while it does have a cost, it is not prohibitively expensive. It does/did impact the popular vote total, but the real key is that it helps grow an infrastructure that can be reused, and builds experience that can be reused.

New York was not going to be competitive in 2004. But there will be a nice Senate race in 2006, and the experience and watered grassroots will come in handy. We did not win New Jersey, but the rusty machinery there will benefit from the oiling it just got in future elections. To regain competitiveness in Illinois, we have to be prepared for the future. And Connecticut, and so on.

[John Thacker]

Bush showed real improvement in his total number of votes in lots of states, not just battlegrounds or ones he won in 2000, but in plenty that he lost in 2000 and 2004 as well. CT, NJ, PA, MN. Of course perhaps those were the "wrong" sorts of people in those states.

We weren't that far off from a Bush blowout in electoral votes, as Kerry won a lot more close ones.

Close Kerry wins:
New Hampshire (50-49)
Wisconsin (50-49)
Pennsylvania (51-49)
Michigan (51-48)
Minnesota (51-48)

That's not too big a swing for Bush to capture 348 to 190 for Kerry.

Close Bush wins:
Iowa (50-49)
New Mexico (50-49)
Nevada (50-48)
Ohio (51-49)

Kerry had to take Ohio. Even a swing to give him all the first three would just make a 269-269 tie, and Bush wins on the tiebreaker. With Ohio, it's 289-249, pretty close to the 286-252 it looks like today.

Florida (52-47), Colorado (52-46), Missouri (53-46) are the next group for Bush; Oregon (52-47), Washington (53-46), New Jersey (53-46), and Delaware (53-46) are Kerry's next group.

[Appalled Moderate]

Anecdotal evidence from georgia --

The races in ga were not close (the senate race was a GOP runaway, the house races were uninteresting, there were few statewide races and Bush was a shoo in to win the state). But there were long lines everywhere.

Explanation -- there was a ballot resolution defining marriage as being between a man and a woman.

So in georgia, at least, I would say that the initiative probably juiced up turnout.

[Crank]

D'oh! Broken link! Try this one.

As for the Gold Glove, I demand a recount!

[TexasToast]

advertisement

Religious groups vowed that gay marriage would send evangelicals flocking to the voting booth. Exit polls showed that 22 percent of voters named "moral values" as the most important issue to them—ranking it higher than the economy and the Iraq war. Of them, 79 percent voted for President Bush. In Ohio, 24 percent of those surveyed identified themselves as "white evangelical/born-again Christians."

From Newsweek

[Dr. Manhattan]

Re: Jeter's Gold Glove -
Flukes do happen. Roger Maris hit 61 HRs one year, after all.
And according to two sets of advanced defensive metrics, Jeter's defense did improve by a massive amount this year, so the choice wasn't quite as crazy as it would've been in years past (the source of the improvement is TBD).

[Cecil Turner]

"Exit polls showed that 22 percent of voters named "moral values" as the most important issue to them"

Are those the same exit polls that showed Kerry winning by a landslide?

Pre-election polls showed Florida voters picked "terrorism" as the single biggest issue, and "Iraq" second. By splitting those two issues, it might be possible to miss that a (30+26=56%) majority were primarily concerned with national security. But claiming now it was all about gay marriage is hardly "reality-based."

[Paul Zrimsek]

There ought to be a pun in here somewhere about "the glove that dare not speak its name."

[TexasToast]

My point has been - and continues to be - that the new voters who turned the tide in places like Ohio were largly evangelicals who came out to vote against gay marriage. Kerry won the "independants" (ie the "center"), 55-45, the youth vote was the highest ever and supported Kerry by a hefty margin, but the rural, evangelical, new voter swung this election.

I thought it would be the Iraq war or the WOT, myself, but you and I were wrong and it appears that Karl Rove was right. The question remains, how are you guys going to "pay your debt?"

[John Thacker]

TexasToast-- Of course, Bush won a higher percentage of Democrats than Kerry did Republicans.

the youth vote was the highest ever and supported Kerry by a hefty margin-- The youth vote was the same percentage as every other vote, and supported Kerry by the same percentage they supported Gore. (Well, 55% instead of 54%.) The youth vote was the highest ever because every age group was the highest ever, because turnout was the highest since 1968 and the population has grown.

Bush won 25% of Jewish voters, compared to 19% in 2004. Guess they all cared about gay marriage. Bush won 23% of gays, compared to 25%, strange if the election was all about gay bashing. (What, are they "self-hating gays?") Won a higher percentage of gun owners. Etc.

You've given no evidence that a higher percentage of first-time voters came out because of gay marriage. (And in any case, "moral issues" can include, for example, abortion. In my mind, it includes Kerry's immoral realist foreign policy and his immoral statements in favor of protectionism.) It's a reasonable hypothesis, but hardly proven.

For one thing, get some data comparing to 2000.

[John Thacker]

Now, it is true that in 2000, the Democrats had great turnout compared to the Republicans, especially in Florida. Republicans have argued since that it was because of a great GOTV effort by the Democrats, which the Republicans have now matched.

How do you explain 2002, anyway? 2002 and 2004 gave essentially identical results, without gay marriage on the ballot anywhere. Isn't it merely possible that the Republicans have improved their Get Out The Vote machines to match those of the famed Democratic ones?

[Bradley Hampton]

Interesting question, and like Appalled Moderate, my evidence is anecdotal, but here goes.
There were no Senators up for election, no big initiatives on the ballot, and a state that was absolutely never in doubt for the President - Texas. Rove probably never called anyone in this state, nor turned out an "operative" - because he didn't have too.

2000 Election
-------------
Bush: 3.797m votes
Gore: 2.429m votes
Nader: 0.137k votes
Total: 6.363m votes

2004 Election
-------------
Bush: 4.503m votes
Kerry: 2.820m votes
Other: 0.038k votes
Total: 7.361m votes

Over 1.097m more voters in 2004 than in 2000, and President Bush picks 706k of them.

More anecdotal evidence: lines, lines, lines. Long lines even in a "safe" state to vote.

Thoughts: this is the President's home state, so the effects are somewhat more difficult to measure, but I don't think that can explain 1 million more new votes. In my opinion, the Iraq war was still first and foremost, but don't discount that the judges/gay marriage/abortion issues. They absolutely count, and big down here. (Note: in my judgement, like Kevin Drum's, it started with the Massachusett's Court decision to legalize gay marriage.)

Finally, something I haven't seen really discussed and/or blogged: the absolute assault on the President for the last four years probably pissed people off as well, so much so that it may have been something of a motivator to vote. How many anti-Bush movies and books were there? How many times have the elites and MSM inferred or implied that the President is dumb (amongst the least of the charges they've levelled)? How many times have you gone to the left side of the blogosphere and read vitriol so foul directed at the President of the United States that you want to bathe afterwords? Maybe it didn't mean many more votes overall, but it made me feel a little better when I can proverbially "poke a stick" in the hatin' left's eye...

[Bradley Hampton]

John Thacker,

The breakdown by age group is interesting, but even more interesting is the breakdown by income (see here).">http://www.cnn.com/ELECTION/2004/pages/results/states/US/P/00/epolls.0.html">here). The only income Kerry dominated was for those earning $30,000 or less. He wins by one percent (50% to 49%) with those earning an income between $30,000 to $50,000, with anyone earning $50,000 or more, Kerry was spanked...

[John Thacker]

Right, re: gay marriage initiatives, you have to note that:

They were only in 11 states, yet everywhere showed massive turnout increase, and everywhere (except four states, including WY, I believe) showed Bush doing better than in 2000 in his vote percentage, and most everywhere showed Bush doing better relative to Kerry than he did relative to Gore, despite the utter collapse in the Nader vote. One notable exception was Ohio, one of the few states with a two-party swing against Bush-- and this was a place with the supposedly all-powerful gay marriage initiative. Sure, it's easy to argue that without the initiative, Bush would have lost, but it's got to be a harder case.

Are there that many evangelicals in CT, where Bush sliced a 17 point disadvantage in 2000 to 10%, who came out despite no marriage initiative for example?

[doctorbong]

"Moral issues" is a pretty broad category. Though I'm sure that it includes red meat issues to social conservatives, it's not unreasonable to think that other campaign issues come into play under the heading of Moral Issues.

Certainly candidate character edges into moral issues. Also, our attitudes about war are closely related to what we consider to be moral, so terrorism and Iraq can't be far from the minds of "moral issues" voters.

[sym]

bush would have lost if it wasn't for the gay marriage amendment on the ballot in ohio. I'm sure that there were at least 150,000 voters that would have stayed home without the amendment, since it's such a small percentage of the Ohio vote. The amendments didn't have to boost turnout much to swing the whole thing for Bush.

[Cecil Turner]

"bush would have lost if it wasn't for the gay marriage amendment on the ballot in ohio."

I suppose that's possible. But he certainly would have lost if John Kerry had had a coherent defense strategy. From Andrew Coyne

the pollsters only managed to elevated "moral values" to number one by dividing up the other issues into subcategories. Thus "Iraq" and "Terrorism" are treated as separate issues, though grouped together as, say, "national security" they would have claimed the top spot, with 34% of the total. Likewise "taxes" and "economy" were named by a combined 25% of voters. Had "moral values" been split into "abortion" and "gay marriage," the spin would have been rather different.

It's worth noting that a majority of voters don't agree they're two different issues, and they broke overwhelmingly toward the President--and confidence in the President's terrorism policies was even more indicative:

                          Bush/Kerry

IS IRAQ WAR PART OF WAR ON TERRORISM? 

Yes (55%)                  81%  18%  

 No (42%)                  11%  88%  

TRUST BUSH TO HANDLE TERRORISM?

Yes (58%)                  85%  14% 

 No (40%)                   4%  94% 

It's also worth noting that the "hate" vote went to Kerry by better than 2 to 1:

YOUR VOTE FOR PRESIDENT WAS MOSTLY...

For Your Candidate   (69%) 59%  40%  

Against His Opponent (25%) 30%  70% 

[sym]

but cecil, the "Iraq is top issue" vote was a Kerry landslide, and the terrorism vote was an even bigger bush landslide. if people voted for Bush because of the war on Iraq, why didn't they say so? Taxes and economy also were cited by mainly rep and dem voters respectively. There was no partisan divide between abortion voters and gay marriage voters.
Yes, Kerry could have won big if he was as good a politician as the President, but he would have won a nice close electoral vote victory (and a big popular vote loss) if it wasn't for the Ohio initiative.

[Cecil Turner]

"but cecil, the "Iraq is top issue" vote was a Kerry landslide, and the terrorism vote was an even bigger bush landslide."

Perhaps because the people who believe Iraq is part of the WoT called the issue "terrorism"? Again, more than half say it's one issue, and they gave the election to the President. You can pretend it's all about gay marriage if you like . . . but it's not very convincing.