Skip to comments.OH Real-Time Election Simulation (Romney 50.11% Obama 49.89%) Methodology Included
Posted on 10/06/2012 8:42:28 AM PDT by parksstp
This is probably the most detailed simulation Ive ran, since I was easily able to find the latest voter data at the OH Secretary of State site.
Basically, what Ive done is run a simulation for all of OHs 88 counties and the percentages Romney and Obama need for each county to have a shot at the state.
Romney: 2,781,909 / 50.11% Obama: 2,769,785 / 49.89% Total Votes: 5,551,694
Results of the 2004 Presidential Election (Bush v Kerry)
Results of the 2008 Presidential Election (McCain v Obama)
Results of the 2010 Midterms (Kasich v Strickland)
Results of the 2011 Question 2 (Union Issue)
Using the data at the Secretary of State Website concerning the total number of registered voters within each county, I compared that number to the total number of Bush/Kerry and McCain/Obama votes to determine a turnout-percentage (Website says 04 was 72% and 08 was 69% and 10/11 was around 49-50%). I then apply whatever the trend has been in that county for the percentage.
In garnering real time results, I assume Romney started around 1-2% points higher than McCain, but lower than Bush for most counties. The exception is with what I call the PUMA Counties (McCain took a higher share than Bush some SE Ohio counties where Romney runs slightly ahead of McCain and ahead of Bush). After the debate, Romney picked up more ground to where he is now about 0.5-1% behind Bush in most counties or even, and around 3% or more above McCain (depending on the percentage gap between McCain and Bush). I tried comparing the 2010 Gov race percentages, but it is difficult, because Stricklands perception that he is a conservative Democrat in the state allowed him to over-perform in the results.
Another interesting result is on Question 2. Even in heavily GOP counties, with few exceptions, Kasischs proposal was rejected. The only way this proposal got rejected is sizable number of Republicans had to vote against it. It doesnt seem to effect these voters from still voting Republican however.
Looking at each county, Ive estimated the vote totals and percentages I think Romney and Obama would receive. Prior to the debate this week, Romney was probably about several thousand votes down, but now its flipped in the other direction. OH is really, really close. Theres no way either candidate has more than a 3 or 4 point lead in this state. Romneys lead will increase based on GOP turnout, while Obamas total may increase based on lack of enthusiasm for Romney. With 30 days to go, anything can happen.
The bottom line is while the raw vote numbers may change, if Romney hits the percentages Ive listed for each of the 88 counties, he should be able to win. Hopefully, this document will prove useful as a tool on Election night to see how the state is going. If any candidate overperforms the numbers I have listed, you can probably see how the state is going to go even if the networks dont call it.
Have you considered:
Ohio Poll Analyst: Voter Turnout Modeling Predicts Big Ohio Trouble for Obama
Pajamas Media ^ | 10/05/2012 | Paula Bolyard
After hearing a report that Ohio voter registration rolls have shrunk by 490,000 , Cooper wanted to see if media reports that Obama is enjoying a wide lead in the state would stand up to a statistical analysis based upon actual voter registrations and historical voting patterns.
He decided to analyze Cuyahoga County, which includes Cleveland and is the states most populous county. He says its representative of the state because it has a large enough number of registered voters and you can statistically determine how many total votes each candidate will get in Ohio based on how many votes are cast for each party in Cuyahoga County. In fact, from 1980 to 2008 an average of 17% of the total Democrat votes cast in Ohio presidential elections were cast in Cuyahoga County. In recent years, this has declined (to around 15.6% in 2008). During the same time period, an average of 9.3% of the total Ohio Republican ballots were cast in Cuyahoga County. That number declined to 7.5% in 2008. Since 1980, the average voter turnout for Cuyahoga County is 64.57%. Obviously, this is a fairly significant sample size.
Cooper found that Cuyahoga County has 208,207 fewer total registered voters in 2012 compared to 2008. The county has lost 48,872 Democrat-registered voters and 194,199 independent-registered voters. The difference and this is significant to Coopers prediction is made up by a surprising gain of 34,864 Republican voters since the 2008 election.
Thanks for all the work you have done. That is quite a lot of work on that spreadsheet.
I did, which would put Cuyahoga at about 587,000 votes for 2012. If this happened, Obama would be sunk. But there was also a decrease from 2004 to 2008 where there were only 12,000 less votes cast. 641,000 may be too high. I just have a hard time seeing that drastic a drop when all 88 counties have been consistent over the past decade in slight changes up or down. Not saying it can’t happen, I just don’t have any results to base it on.
That cuts both ways.
'08 was an unusually high turnout year. There WAS a lot of enthusiasm for Zero. Even some "republicans" got swept up in the enthusiasm.
That enthusiasm for Zero has disappeared. I was talking to someone a few weeks ago who was pretty heavily involved with local dem politics, and he said there was no way he could vote for Zero (of course he didn't call him Zero).
The point being, there are numerous people who identify themselves as dems who WILL NOT vote for Zero.
There are also MANY republicans that did vote for the jug eared one, who vehemently regret their past mistake.
Look at Dumbo's historic Rasmussen Presidential Tracking Poll. 1 month after inauguration, his approval numbers were high, including the strongly approve. Now the Strongly disapprove are about 12 points above the strongly approve. That means the strongly's are the ones more likely to vote. The strongly disapproves MIGHT NOT be enthusiastic about Romney, but they are vehement against Zero. Also, those who aren't strongly for Zero, are more likely not to vote at all.
Granted Rasmussens poll ISN'T Ohio, but I think some things can be generalized here. According to Ras, Obama has a 50-49 approval. But lets look deeper. Out of 50% approving 28% strongly approve, 22% just approve. The 28% is a lock, but the 22% who somewhat approve, I'd say half, minimum of a 1/4 won't show up.
On the flip side 49% disapprove of Zero, but out of those 49%, 41% Strongly Disapprove, leaving only 8% who somewhat disapprove.
This 8% are the ones who disapprove of Obama, that aren't enthusiastic about Romney, that might not show up.
Even IF you could say that NOBODY was enthusiastic about Romney 82% of those STRONGLY Disapprove of his opponent, 16% are maybe ambivalent.
I’m in northern Ashland county.Last election,the Obama lawn signs were everywhere.This time you can hardly find one.Even my neighbor who’s a die hard union guy and lifelong Dem doesn’t have an Obama sign out.Lots of local candidate signs-but no Obama.
Park's numbers look good. There have been about 5.5 million voting in Ohio, so that's on the money. The problem is that there are about 8 million already registered voters. That's enough for either party to change the outcome by virtue of a solid door-to-door, ground campaign, and with early voting, it is no longer necessary to have that ground campaign happen only on election day.
The dramatic turn-around of Romney's fortunes after his strong debate performance suggests that 2 more such performances -- assuming Obama remains weak -- will put Romney over the top without a ground game. I wouldn't count on that, though. Either Obama will improve his performance or the media will improve their pro-Obama spin.
If I were a republican leader in Ohio, I'd be putting drivers on the street carting registered voters who don't normally vote to the location where they can vote early, and I'd double those efforts on election day.
After all, 2.5 million registered voters who don't vote suggests that 1.25 million of those, about half, are republicans. If they're already registered, why not find some way to get them to cast a ballot?
>> Im in northern Ashland county.Last election,the Obama lawn signs were everywhere.This time you can hardly find one. <<
Similar situation in the left-leaning suburbs of northern Virginia, where I spend a lot of time. My off-the-top-of-the-head estimate is that now, we’re only seeing 10% to 20% of the Øbama signs we saw in 2008.