New Study on Election Trends

[phk]

For the Dakotas it could be a Bush over-performance. In 2000, I had thought of a massive Gore under-performance there though.

[WhyteRain]

1988: Right-leaning votes (repercentaged): 53.95%
1988: Left-leaning votes: 46.05%

1992: Right-leaning votes: 56.85%
1992: Left-leaning votes: 43.15%

1996: Right-leaning votes: 49.84%
1996: Left-leaning votes: 50.16%

2000: Right-leaning votes: 48.77%
2000: Left-leaning votes: 51.23%

2004: Right-leaning votes: 51.19%
2004: Left-leaning votes: 48.81%

2008: Right-leaning votes: 46.32%
2008: Left-leaning votes: 53.68%

By allocating third party votes to either side, we see less dramatic swings across time than if we just looked at the two-party vote.  We see that in 1988 it is basically a direct reflection of the Bush-Dukakis vote (although I did allocate Ron Paul with the right and Lenora Fulani with the left).  In 1992, the right-leaning percentage actually goes up from 1988 because I chose to classify Ross Perot as a right-leaning candidate.  In 1996 and 2000, the right-leaning and left-leaning coalitions were about evenly matched.  The left-leaning percentage actually goes up from 1996 to 2000.  This shows the country may not have been ready to make a move to the right.  They were happy with the economy but some voted against the Democrats and in favor of Nader because of frustrations with Clinton.  The right-leaning percentage takes a slight uptick in 2004 partially because of a rally-around-the-flag tendency.  Then in 2008 after frustration with Bush and the Iraq War, the right-leaning coalition drops below 47%, the lowest of all 6 elections.  What is more interesting, however, is to compare each individual state against the national average.  I will do that in the posts that follow.

Just a few comments:

Re. 2000, you might want to account for the fact that some states counted their absentee ballots and some did not.  For example, cash-rich California (/sarc) decided to count its absentee ballots while financially foundering Texas (more /sarc) did not.  If it had been the other way around, then Bush would have "won" the national popular vote.

Re. 1992, I don't think all of Perot's voters were "right-leaning".  I did an interesting exercise once in which I assigned 2/3s of Perots votes to Bush and 1/3 to Clinton.  It resulted in a very, very tight race that would have been decided by a handful (~3,000) voters in Iowa.

Re. 2008, you say voters had soured on the Iraq War.  That was 2006, not 2008.  By 2008, voters knew that Bush had won his second victory in Iraq (okay, not the MSM zombies; they were clueless as always).  2008 was decided by McCain's miserable perfrormance -- remember how he ended the campaign having to attend his Vice-Presidential candidate's rallies, just so he could see what a large crowd looked like? -- and by the fact that Democrats implicitly promised to stop calling everyone "racists" if they voted for Obama.

[WhyteRain]

1988: R+0.15
1992: D+0.73
1996: D+1.85
2000: D+2.1
2004: D+3.06
2008: D+4.59

Bad news for Republicans.  This state is trending liberal.

NOOOOO! As a conservative citizen of Michigan, that's quite distressing news. I was hoping there might be a chance for redemption, but alas, it doesn't look like it.

Sorry, Michigander (love that name), I have Obama faring badly in this election -- losing six states that Kerry won in 2004 -- but even I have Obama winning Michigan (basically only Detroit and Ann Arbor) this year.

[ChipGardnerNH]

WhyteRain, could you share your 2008 map?  As far as 1992 goes, I do not think all of Perot's votes came from Bush.  It's possible a third or even as many as half came from Clinton.  But I don't think the people who would have otherwise voted for Clinton but voted for Perot were liberal types, or the same type of people who voted for Ralph Nader.  They were probably right-leaning people who would never have voted for HW Bush because of the bad economy and could have been available for Clinton, but they were center-right people, not liberals.  I think just about all liberals voted for Clinton in 1992, not Perot.

[WhyteRain]

WhyteRain, could you share your 2008 map?  As far as 1992 goes, I do not think all of Perot's votes came from Bush.  It's possible a third or even as many as half came from Clinton.  But I don't think the people who would have otherwise voted for Clinton but voted for Perot were liberal types, or the same type of people who voted for Ralph Nader.  They were probably right-leaning people who would never have voted for HW Bush because of the bad economy and could have been available for Clinton, but they were center-right people, not liberals.  I think just about all liberals voted for Clinton in 1992, not Perot.

Chip, I didn't create a 2008 map.  Besides 2012, the last time I created a map was in 2000, when I predicted Gore to win 281 to 257.

"As far as 1992 goes, I do not think all of Perot's votes came from Bush.  It's possible a third or even as many as half came from Clinton. "  <---- that's exactly what I said, but didn't you say you counted all Perot votes as "right-leaning (meaning GOP)"?  The economy wasn't bad in '92.  The recession had ended (in fact, GNP growth for 4Q 1992 was 5%, something Obama has never approached despite the steep recession).  I guess there was a PERCEPTION among the MSM zombies that the economy was still bad.

[ChipGardnerNH]

I meant to say 2012 with Romney winning some Kerry states but not Michigan.

[ChipGardnerNH]

Right-leaning doesn't mean GOP necessarily.  Some right of center people may never have voted for Bush in 1992 because of the perception about the economy, but given the choice between Clinton and Perot, they went with Perot.  Liberal people would probably be pretty unlikely to be attracted to Ross Perot.  The only thing they might possibly like about him would be the protectionism.

[Oldiesfreak1854]

1988: R+0.15
1992: D+0.73
1996: D+1.85
2000: D+2.1
2004: D+3.06
2008: D+4.59

Bad news for Republicans.  This state is trending liberal.

It may be trending liberal, but it's still a swing state.

[BritishDixie]

Good study, although I would split Perots 50-50 for 1992. But I think most of his voters would have gone for Dole in 1996.