Posted on 05/14/2013 2:28:52 PM PDT by parksstp
Prior to the 2012 Election and after the first Romney/Obama Presidential Debate I started creating prediction models for each of the contested "Battleggrounds". OH was the first and most detailed state I reviewed because it seemed like it could go either way.
The original model spreadsheet I posted on FR contained county by county results for the 2004 and 2008 Presidential Elections, the 2010 OH Governor Race, and the 2010 Referendum Union Question. I used this data to develop the probable voter turnouts for each county and determined the number of votes and vote margin the GOP candidate would need to win OH.
The model left no room for error for either candidate, and if a candidate could persistently beat the model, even if it was just a small edge, the advantage would be decisive.
The model gave Romney a slight statewide victory by around 12,000 votes. In reality, he lost OH by over 178,000. So the natural question to ask was what happened and where did the missing votes go?
This analysis took longer since CNN decided to not adjust their vote totals to match the OH Sec of State, which I did not realize differred until I thought I was finished and short thousands of votes.
For the most part with the exception of the large blue counties, my model stayed within 5000 votes for each county, and for a majority was within 1500 votes of the actual margin. Voter turnout was a little harder to predict but for many counties, I was within 1000 (note: I only count voters in my models casting votes for the 2 Rep/Dem candidates, not the other parties).
As to the results:
Romney underperformed, which is no surprise, but for most of the non-deep blue counties, the underperformance was slight, and in most counties less than 1,000 votes. Yet, since OH has 87 counties and the model only gave him a 12,000 vote edge, these slight underperformances were decisive even before the larger blue counties came in.
Stark County was Romney's best performing "blue" county and overall, winning a net gain of 4,511 Votes. Romney also performed well in Portage, Athens, the rest of the SE part of the state, and for the most part held his own in Trumbull, Ottawa, Wood, Erie, and even Summit County, where he was within less than 1500 votes of the model.
Franklin County and the Columbus area was by far the most damning result in the state. Romney lost a net vote of 60,339 votes in Columbus, more than Cuyahoga (32,241) and Hamilton (27,449) combined. Lucas County (Toldeo), was also a net loss of 17,478 votes from the model. A Republican candidate hoping to win OH cannot lose these 4 counties by more than a combined total of 355,000 votes and expect to survive. Romney lost these 4 counties by a combined 480,266 Votes.
Contrast this with Bush/Kerry, where the incumbent lost the 4 counties above by only 240,000 votes, and managed to win by 170,000 votes.
This is how Barone and his crew were able to call OH before the rest of the precints in SW OH came in that got Rove worked up. Obama had racked up such a large lead in Columbus, even had Romney evened Hamilton County which was what most analysts thought was his maximum performance level, it wouldn't have matterred.
As for Columbus, the results were truly puzzling. Obama got the same number of votes and then almost 100% of the increase in Dem/Rep voters from 2008/2012 who voted for the Dem/Rep candidate, while Romney failed to acheive even McCain's level. Most analysts (myself included) expected Romney to garner 43% in Franklin County, but he was almost 5 points off.
In Cuyahoga County, voter turnout was down by 50,000 from 2008, yet Obama still managed to come with 10,000 votes of his 2008 vote totals, while Romney could not eclipse 200,000 and once again performed behind McCain.
The trend above would make one expect to see not just a slight, but significant decrease for Romney across the state. But that didn't exist. While Romney was off for most of the counties, it wasn't by much, and those margins would not lead someone to believe of the massive defeat in Franklin and Cuyahoga.
Before people start claiming voter fraud, it should be noted in the red counties surrounding Columbus, these red counties seem to be Romney's worst producing in the state, while he did at or exceeded expectations in the East, West, and SW parts of OH. Without a more thorough study of demographics in the surrounding areas of Columbus, I can't offer anymore information.
I'll say this though. For 2016, OH is going to be very difficult and is dangerously approaching IA status where there will simply be not enough conservative voters to carry the state in majority. To win OH, the Republican candidate MUST make Hamilton County even, at a minimum. They also must reduce margins in Cuyahoga to under 220,000, Franklin to under 70,000 and Lucas to 50,000. They also need to pick up, on average around 1,500-2,000 in the designated red counties. The Democrats, meanwhile, can abandon the Coal Unions in the East and Southeast as they no longer appear to need them or the PA counterparts in the SW.
A strategy for 2016 has to be constructed now. For OH FReepers associated with the Tea Party, please use the spreadsheet data as an aid with planning.
The first tab, you can scroll across to the last two columns that say 2012 Predicted and 2012 Actual. You can view how Romney/Obama fared in the model and how the county results this time compare to previous results for President, Governor, and Question 2. Counties in Red are GOP, Blue is liberal.
The second tab on the spreadsheet labeled "Missing Votes by County" shows the number of votes gained/lost by the GOP candidate to Obama in the prediction model. Summing this gives Romney the 12,000 vote margin victory he would have had, so when looking from where the missing votes are at, this is a tool. Many counties are under 1,000 votes, but the 4 largest blue counties were by far the most damaging.
The GOP in OH is in a sad state. They just chose a former lobbyist for the LGBT cause to head up the party. As such, I’m not sure folks are very motivated to support the state party.
Columbus = State Capital = Lots of GOV Employees = Obama Lovers.
Plus the place has been getting steadily bluer since I left many years ago. No surprise Romney got his arse kicked there.
At the national level, the state is trending blue. For statewide races, at this point its an even bet. It depends on the state economy. Ted Strickland clobbered Ken Blackwell back in 2006 in the ‘Rat tsunami, using a “turn the state around” theme. But when he failed to deliver by 2010 the voters sent him to the sidelines by a narrow margin. But if Hillary is the nominee in 2016, she’s going to clobber whoever the ‘Pub nominee is. The national ‘Rat ground game here is too strong.
When you have Dems voting 6 times each in Cuyahoga County, it is hard to win elections. How may precincts went 100% for Obama and what was the turnout in those precincts? I firmly believe the Dems stole Ohio.
Agreed. The fact that the GOP is turning its back on conservatives only exacerbates the problem.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.