If Obama has a 50 point lead with 19% of the Ohio early vote... Math geniuses welcome!

[Alekei_Firebird]

By how much does Romney have to win the remaining voters to pull out a win? At what point does it become all-but-mathematically-impossible for Romney to overcome such an early deficit?

Right now, Obama leads 76-24 with 19% of the voting already completed. If Romney maintains a 51-46 lead with the remaining 81% of the voters, what would the final numbers look like?

Here's my quick math:

About 5.6 million people voted in Ohio in 2008. Assuming that demographics and turnout are relatively similar, that would mean that about 1.1 million people have already voted. Of those 1.1 million, about 836 000 have voted for Obama and 264 000 have voted for Romney.

If Romney carries the remaining 4.5 million voters at a 51-46 advantage, then he'd have 2.3 million votes and Obama would have 2 million votes.

Final tally would be:

Obama: 2 836 000 (50.6%)
Romney: 2 564 000 (45.7%)

Just like in 2008, Obama wins Ohio by about 5%.

Is my math horribly flawed?

[RedSpartan]

1. I look at it this way...

And we get to almost the exact same result...

Let's pretend there are 100 votes in Ohio.

19%, or 19 votes, have been cast, and Obama leads with them 76%-24%. So Obama has 14.44 votes and Romney has 4.56 votes.

81%, or 81 votes, have yet to be cast, and Romney leads with them 51%-45%. So Romney will have 41.31 votes and Obama will have 36.45 votes.

Add them up, and Obama has 50.89 votes and Romney has 45.87 votes -- essentially the same 51%-46% that PPP is reporting in total.

So unless I am mistaken (and I admit I likely am), it seems to me that if these numbers hold, Ohio is very, very, unlikely to go for Romney unless he can get the 81% of people who have not voted yet to go for him by closer to a 56%+ level. I can't see that happening.

[writes3000]

3. More than 56% percent. That's the math I got as well.

[SheilaT]

21. The only thing I don't get is

Last edited Sun Oct 14, 2012, 02:48 AM - Edit history (1)

why, out of 100 votes, your math only gives us 97 votes in the end?

edited for a typo.

[RedSpartan]

22. For the same reason that PPP's numbers don't add up to 100...

Third party votes and/or undecideds.

[SheilaT]

24. Oh.

But I would have thought doing the math the way it was done would have resulted in 100 votes. Never mind.

[tarheelsunc]

2. Let me try

19*.76 = 14.44

50 - 14.44 = 35.56

35.56/81 = 43.9%

Assuming no third party candidate receives a vote and that 76%-24% is the correct margin as of now, Obama needs 44% of the remaining vote to win.

[Azathoth]

7. Sounds about right. n/t

[HockeyMom]

4. If Obama can take Ohio,

and more of the MidWest, then screw Florida which is, and especially now, is a very corrupt state.

[cheezmaka]

34. My calculations exactly!

If Obama takes Ohio and also wins PA and WI, he will win even if he "lost" FL and every other battleground state.

[Sekhmets Daughter]

5. No not horribly. :-)

[Applan]

6. Another way of looking at it

If the 19% represents likely voters then Willard will need to carry the remainder of the votes at 56% to 44%

But if the 19% represents all eligible Ohio voters, and they get a 60% turnout as in 2008, then Romney would need to carry over 62% of the remaining votes.

Either way it's not looking too good for the rich git is it?

[alcibiades_mystery]

8. Just as a note, these are extrapolations from polls, not actual vote results

Just saying, a word of caution is important here. Fight for every goddamn vote until the second the polls close.

[dkf]

9. What does this 19% represent? Is that of the expected vote? Can turnout affect it?

[Azathoth]

10. 19% of Ohioans have already voted

76% of those, or close to 15% of all Ohio voters, have already voted for Obama.

[dkf]

11. Is that of all registered voters then? Seems kind of high.

[Azathoth]

13. All likely voters I assume n/t

[FBaggins]

23. Presumably LV, but yes - it seems high.

Every polling firm has their own minor tweaks on the likely voter screening... but I think it's fair to assume that anyone who says they already voted gets through the screen.

As for your reasonable assumption - I can only find reporting on 175k early votes so far. There are good reasons to think that this is low (some counties have not fully reported and some votes may be in the mail), but that's only 3% of 2008's total turnout. Additionally, early voting was just under 30% of the electorate (in Ohio) in 2008 - with the bulk of it coming just before election day.

So it does seem highly unlikely that 19% have actually voted already.

I don't know what that means for their sample. I'm unaware of any studies that compare responses to the "have you already voted?" question to actual voting. Maybe people claim that they've voted because they think it will make a political caller hang up (not all "pollsters" who call turn out to actually be pollsters). But that's just a preliminary guess.

[LisaL]

25. There were three OH polls.

The earliest one said 10% already voted. The second poll said 18 % already voted. This current poll says 19 % already voted. Obama had a huge lead in all three polls for the early voters.

[FBaggins]

29. I'm not saying that they're lying.

Just that the number is odd. There's an actual count of how many votes have been cast, and it doesn't seem to match these polls.

Also, I don't know how well you remember the 2010 election, but there were these kinds of projections based on polls and early voting figures. They were constantly used to show us that reality wasn't as bad as what the polling seemed to say it was.

I don't know whether it was faulty polling (nobody has a good track record of identifying people who have already voted), or faulty analysis... or just that Democrats vote early far more often even in losing races... but I'm much more comfortable talking about their headline number and just leaving off estimates of how far up we are in actual voting.

[LisaL]

37. Some might have voted via absentee ballots.

Which could be in the mail thus not counted in the amount of those who already voted.

[FBaggins]

39. Yep.

Just not anything close to enough to account for the gap.

[regnaD kciN]

12. Let's not count our chickens just yet...

Recall that all these figures are extrapolated from a poll, not official vote counts. It's taken from less than one-out-of-five respondents - if the PPP poll has a margin of error, I would think that the margin of error for a small subset of that poll would be considerably greater.

In other words, while it's reasonable to conclude that a significant bloc -- if still a minority -- of Ohio voters have already voted, and those who have show a significant tilt to Obama, starting to calculate detailed scenarios from those exact, down-to-one-hundredth-of-one percent figures is pretty much a fool's errand. Let's be happy that the PPP poll apparently shows us with a lead beyond the margin of error, and be happy with that.

[LisaL]

15. All three recent polls show Obama has a large lead among early voters.

[Doctor Jack]

14. Uhhh.....let's see.....

...Ok, got it, 280%.....that can't be right. I never should have learned math from Paul Ryan!

[Blaukraut]

16. Let's pretend that nobody has voted early, and O is leading 51 to 46 (PPP today). Looks good, no?

So I'm no math genius, but I doubt Romney can improve on the 51% to 46% advantage he has with those who haven't voted yet, unless Obama does something really stupid. Romney's momentum has stalled, and from the looks of it, he never had it in Ohio in the first place.

[Doctor Jack]

17. A 50 Point Lead? So its a tie?

[brush]

18. Not flawed

No it's not flawed if Romney were to get the 51-46 percent margin, but it's very highly unlikely if in the sample of actual voters that we have he's already behind 76 to 24. I'm thinking the final vote will more likely reflect the actual 76 to 24 numbers than 51-46 estimate of some poll.

[DonCoquixote]

19. wartch those absentee votes

The ones the GOPloves because they are easily hackable...Florida always seems to get a wave of absentee voters that always vote GOP, surprise surprise.

[bemildred]

20. Assume the 19% number is correct:

then 19% of 76% is ~.14 or 14% of the votes.

and

19% of 24% is ~5% of the votes.

So Obama has a lead of 9% of the votes.

iF the 19% number errs, the numbers would change in proportion, i.e. if it's really 25% complete, you would get 19% and 6%, so O's lead is 13%, bigger. If the true number is 10%, you'd get 7.6% and 2.4% and the advantage would be 5.2%, much smaller.

So it really depends on how accurate the 19% of the total vote estimate is.

[Historic NY]

27. Arithmetic.

[thevoiceofreason]

28. It depends on what the 19% number represents.

If the 19% number means 19% of the expected turnout, then your math is correct.

If it means 19% of the registered voters, then we have to guesstimate what the turnout will be this year. The best data I can find is that the 2008 Ohio RV turnout was 65.5%. So assuming that is the case this year, 19/65.5 of expected voters, or 29%, have already voted. And the poll says President Obama leads that group by a 76-24 margin. That means that President Obama has 22% of the total expected turnout (29% X .76) already, and Rmoney has 7% (29% X .24).

So, for the remaining 71%, Rmoney needs to end up with a percentage of that amount which will give him 43.1% of the total, to give him 50.1% (7% + 43.1%) and win Ohio. Therefore, Rmoney would need 60.7% of the remaining 71% to end up with the 43.1% (60.7% X 71%) he needs to get over the top.

Net net: If the 19% having voted already means 19% of the RV's, and if turnout is similar to 2008, Rmoney needs 60.7% of the remaining votes. Good luck with that, brittle boy.

[thevoiceofreason]

30. And further, if the 19% means of RV's

Then Rmoney's 51-45% advantage among those who have not yet voted is woefully inadequate (a phrase I am sure he has heard in many contexts in his life). That would mean that he is going to get 51% of the remaining 71%, or 36.21%. Add that to his current 7% gives him a total of 43.21% - a badass buttwhipping.

[Alekei_Firebird]

31. Romney will need to win Ohio by double-digits

The numbers may be a little off b/c it's a poll, but multiple polls have given Obama at least a 40-point advantage with about 20% of Ohio's early votes.

Romney will likely have to win by an overall margin of double digits then.

God, that's gotta be disheartening for his campaign. I love it!

[davidn3600]

32. Depends on where in the state those numbers are from

If most early voting is taking place in Cuyahoga or Franklin counties, that lead not surprising. I think in Cuyahoga county the Ds outnumber Rs by 2 to 1. So you would expect a large Obama lead in that county.

[bemildred]

33. Good point. nt

[LisaL]

38. Yes. Cleveland especially.

That's where democrats in OH get their votes.

[bushisanidiot]

35. You need to figure 60% voter turnout, not 100%

there was 60% voter turnout 4 years ago. this year it will likely be less. maybe %58.

[LisaL]

36. While the turnout is not going to be a 100 % (it's never is)

I am not sure as to why it would be less.
I think in early voting OH is ahead of 2008 already.