10 May 2013

Electoral Tidbit: OHIO

Not just for statistics nerds, this stuff can be really, really interesting. So, have fun. If you take time to read it thoroughly, then I bet you will really enjoy it!



In 2007, I did electoral bios on all 50 states plus DC, going into the 2008 election. Here is the 2007 OHIO BIO.
In 2011, I updated all of those bios, going into the 2012 election. Here is the 2011 OHIO BIO.

Right now, I am starting to record all the county data for each state to compare 2012 to 2008, just as I did four years ago to compare 2008 to 2004 - and just as four years ago, I will start with the battleground states, in ascending order of winning margin. So, Florida, North Carolina, Ohio and Virginia will be the first states to be newly analysed.

But until those analyses are done (and this will take about a year's time to do), here are some juicy tidbits about the Buckeye state that most people do not know. 

Remember, we are talking pure numbers here, not ideology, not reasons for "why" or "how". I am only talking about results, about historical record.

OHIO Statewide presidential votings stats, 1856-1912 (40 cycles, from Dave Leip's uselectionatlas.org, with friendly thanks. The colors are reversed at his site: red=DEM, blue=GOP). Leip is considered the gold standard for electoral statistics. His numbers are absolutely correct.

1.) Ohio has strongly tended to "hug" the national margin in presidential elections, meaning that the results from Ohio were very close to the national average in most cases. And in 2012, it was no different. In 2012, President Barack Obama won nationally by +3.85% over Mitt Romney. That puts the national average between the results from Ohio (Obama +2.97%) and Virginia (Obama +3.87%), with Virginia being a nearly perfect match to the national average.

2.)!!!!  The percentage of "other" votes in OHIO in 2012 was exactly identical to 2008: 1.82%. Two identical percentages in one category for a state in back to back elections is a statistical rarity. In fact, excluding Oklahoma, where the "other" value has been 0.00% for most of it's electoral history, I have never seen it happen in any other state in two back-to-back elections, ever. 

3.) The statistical rarity in the "other" vote also means that the partisan shift (or "swing") from 2008 to 2012 must occur absolutely in equal 1/2 parts between the two big parties, and indeed it did. After millions and millions and million of dollars spent on Ohio on trips and advertising and what-not, the result was that Obama lost exactly 0.80% on his winning percentage, and Mitt Romney gained exactly 0.80% on the GOP losing percentage, making a shift of 1.60% toward the GOP (actually, it is 1.61%, ask me and I will explain why). This swing is lower than the national swing, which is also a trait of Ohio politics.

4.) President Obama is the first Democratic President since FDR (1940) to win Ohio twice with over 50% of the statewide popular vote. Both of Bill Clinton's wins were minority wins. FDR's statistic, however, is very weird, because we are comparing his 3rd election to his 2nd and not his 2nd to his 1st (which was a minority win) and since Obama will never be up for a 3rd election, a true comparison in unfair in this case. But indeed, both 1936 and 1940 were majority wins for Roosevelt. 

5.) Ohio joins a small list of four states which, for the second re-election in a row, gave the winner a smaller margin than in the first election. These are, of course, four of the nine Obama pick-up states from 2008.

Ohio, 2000: Bush, Jr. +3.51% (minority win)
Ohio, 2004: Bush, Jr. +2.11% (majority win)

Swing = Bush -1.40% (or DEM +1.40%)

Ohio, 2008: Obama +4.58% (majority win)
Ohio, 2012: Obama +2.97% (majority win)

Swing = Obama -1.61% ( or GOP +1.61%)

Please notice how close both of those swings are to each other, when we compare one re-election to the last re-election, in spite of the fact that they were for opposing parties. So, for all the money sunk into Ohio in both 2004 and in 2012 for hotly contested re-election campaigns, in both cases between 0.7% - 0.8% of the Ohio electorate changed sides. That should tell us how landlocked Ohio can be.

This phenomenon also happened in Nevada, Colorado and Virginia.

This has happened in Ohio only three other times in history, but not back-to-back as was the case when comparing the re-election of 2012 to the re-election of 2004: Roosevelt in 1940, Wilson in 1916 and Grant in 1872.

6.) You can clearly see that the Obama's winning margin in 2008 (his first election) was larger than Bush's in 2000. 4.58 - 3.51 = 1.07%

You can also clearly see that the Obama winning margin in 2012 (his re-election) was larger than Bush's re-election in 2004. 2.97 - 2.11 = 0.86%.

This is the first time in history where two back-to-back Democratic winning margins in Ohio were larger than the two previous margins by an immediate Republican predecessor. Compare Clinton 1992 and 1996 to Reagan 1980 and 1984 (or even Bush 41 1988), or FDR 1932 and 1936 to Coolidge and Hoover 1924 and 1928. So, a lot of unique electoral history was made in Ohio, which you can see when you sift through the numbers.

That being said, and also having pointed out that Ohio tends to "hug" the national average, once again, in 2012 a Democratic candidate scored a lower winning margin in Ohio than his national margin. This has occured with every Democratic winner in our history except LBJ in 1964 and Wilson in 1916. Conversely, Bush's 2004 win in Ohio (+2.11%), which also "hugged" but was under his national average (+2.46%), is a rarity for Republicans and has only happened to Nixon (1972), Taft (1908) and Lincoln (1860). And in the case of the 1972 Nixon +23 point blowout, no one noticed the difference.

Actually, a three-point race is not really all that close. And polling throughout all of 2012 was pointing to an Obama win in the Buckeye state. Of 117 polls, Obama won 98 of them, there were 7 absolute ties. Romney won 11 polls throughout the year. And the end-polling average pointed to Obama +3.16, less than two-tenths of one percent away from where he actually landed!

7.) How close is a +2.97% win in the history of Ohio politics?


1892 - Harrison +0.13%
1948 - Truman +0.24%
1976 - Carter +0.27%
1944 - Dewey +0.37%
1876 - Hayes +1.14%
1992 - Clinton +1.83%
2004 - Bush, Jr. +2.11%
1968 - Nixon +2.28%
1888 - Harrison +2.33%
1932 - Roosevelt +2.85%
2012 - Obama +2.97%
2000 - Bush +3.51%
1884 - Blain +4.05%
1856 - Fremont +4.30%
1940 - Roosevelt +4.41%
2008 - Obama +4.58%
1880 - Hancock +4.72%1896 - McKinley +4.78%This makes Obama's re-election win in Ohio the 11th closest of the 40 cycles between 1856-2012, or in the lower 25%ile of Ohio elections. Look at the names of candidates who won Ohio by a leaner margin than Obama did in 2012. Very enlightening.

8.) President Obama is the only Democrat in history to win Hamilton county (Cincinnatti) two times in a row. I researched this: not even FDR was able to do this.

9.) No presidential candidate has won Ohio in the +5.00% to +6.00% margin-range. McKinley won with +4.78% in 1896, which rounds to +5, but is not really +5.00%. So, election results in Ohio are either really within the margin of error in polling or they are way outside the MoE, but not on the borderline.

Hope you enjoyed these tidbits. Crossposted here at my politics blog.

No comments:

Post a Comment

Constructive comments and critique are always welcome. Please keep it polite and respectful.