2012 Electoral Math -- Debate Effects Remain To Be Seen
Among Barack Obama supporters, panic seems to be setting in after his first debate performance was roundly panned. National polls have pulled back into a neck-and-neck contest. This is all fun for the pundits, who (pre-debate) were on the verge of declaring the race all but over (and, hence, boring), but we hasten to remind everyone that this is not how we elect presidents. The national popular vote is meaningless -- just ask Al Gore. Presidential elections are won and lost state by state, which is how this column series examines things.
Having climbed up on our high horse, however, we're going to immediately climb right back down and offer up a bunch of caveats before we begin. State-level polling has a higher chance of inaccuracy, to begin with. Also, state polling is done much less frequently than national polling, so it can take longer for the numbers to show movement. While the polling data has been streaming in on the state level pretty steadily for the past few weeks, it is never as up-to-date as the national polling, so trends take longer to show up. Also, one general caveat about all polling, national and state: it always takes longer for voters' reactions to show up than the casual poll reader might think. So while the debate was last Wednesday night, the effects are just beginning to show up in the polling now, and the effects of the unexpected downward movement of the unemployment number on Friday still largely has yet to show up at all. These things take time, folks. The true picture of where we are won't solidify much until next week, to put it another way.
One last note before we move on to the charts -- from now until the election dawns, we'll be putting out one of these Electoral Math columns once a week, every Monday, to keep closer track of how the race looks. OK, let's get on with it:
[Click on any of theses graphs to see larger-scale versions.]
The first chart shows the total percent of vote each candidate would get if all the polls are correct and the election were held today. Obama, in blue, starts from the bottom, and Romney, in red, starts from the top. While there was a minor move up for Obama to begin with, it was countered with a minor move back down, and then the numbers settled back to almost exactly where they began. Obama actually improved slightly, by adding New Hampshire's four electoral votes (henceforth "EV"). Percentage-wise, Obama climbed up to almost 65 percent of the total EV, then slipped back to finish at almost 62 percent. Romney has not managed to top 40 percent in this graph since mid-August.
But this is a general overview, and does not show all of the movement happening. Ten states moved around in our categories this time, and five of them were good news for Romney versus only two for Obama (the remaining three moved, but then moved back again). Mitt Romney is showing momentum coming out of the first debate, but still has quite a ways to go for it to truly be significant when counting noses in the Electoral College.
Romney strengthened his numbers in two states (Arizona and Indiana), while weakening in none of the states in his column. Obama got stronger in two states as well (New Mexico and New Hampshire), but weakened in three states in his column (Michigan, Ohio, and Virginia).
Three states wobbled and fell back. Obama briefly got stronger in Wisconsin, Romney briefly got stronger in Missouri, but both were back by the end of the period. North Carolina flipped between the two candidates, but wound up back (barely) in Romney's column at the end.
Let's take a closer look at all of this movement, starting with Mitt Romney's chart:
[Definition of terms: "Strong" means 10% or better in the polls,
"Weak" means 5% or better, and "Barely" is under five percent.]
The last time we took the pulse of the race was almost two weeks ago (the vertical lines in the graph represent each of these columns). At the start of this period, Mitt Romney firmed up several states and pulled out of the slump he had been experiencing among his base. This was before the debate, it bears mentioning.
Indiana and Arizona both moved into "Strong Romney" on the same day -- Arizona, notably, from as far down as "Barely Romney." At the start of the period, Missouri moved up to "Weak Romney" but would fall back to Barely a week later. During this period, Romney's Barely category disappeared altogether, because with all the good news for Romney, North Carolina moved briefly to Obama's column, before moving back to Barely Romney to finish up.
Romney did get some bad news as time wore on, however, when Arizona slipped back to Weak from Strong -- which was still higher than it started. Missouri moving back to Barely was also a step back for Romney, but the addition at the end of North Carolina brought Romney back to the same 206 EV overall total that he started with. Romney did strengthen his standings within these states, though, hitting an all-time high in his Strong category of 158 EV before slipping back to end up at 147 EV -- which was 11 EV higher than he started.
More importantly, Romney ended his downward slide among the crucial "Strong Plus Weak" category, which had fallen to an all-time low of 170 EV. This rose to 191 EV before slipping back to end at 181 EV -- the same net gain of 11 EV.
Let's take a look at Barack Obama's chart for comparison:
Once again, I caution everyone that this chart is not going to really show much reaction to either the debate or the unemployment numbers until next week, so don't draw too many conclusions from the fact that this period was rather mixed for Obama. He had some ups, he had some downs, but it won't be until next week that we can draw any sort of definite conclusions from the data, with the possible exception of movements in the past two days or so.
Obama started out with some good news from New Hampshire, which then immediately got better. Entering the period, New Hampshire was tied, but it then moved to Barely Obama and then advanced again to Weak Obama, where it stayed put. This, however, was the best news of the period for Obama, and New Hampshire only has 4 EV. New Mexico also strengthened for Obama during the period, back to Strong Obama. Wisconsin flirted with moving from Weak to Strong, but then fell back again at the end, in a post-debate poll. Obama briefly captured North Carolina into Barely, but Romney had snatched it back by the end.
The three truly sobering states this time around for Obama were Michigan, Ohio, and Virginia, all of which slid back. Michigan fell from Strong to Weak, Ohio from Weak to Barely, and Virginia was briefly in the "Tied" category before Obama got it back into Barely. Virginia and Ohio certainly bear watching in the next week, that's for sure, since any movement is likely to show up there (and in Florida) before other states.
Obama's overall numbers weren't all that bad, however, with the exception of one trend. In Strong, Obama fell from 215 EV back to 199 EV, recovered to 214 EV, and then fell back to 204 EV -- losing only 11 EV even with all that movement. Overall, as mentioned, Obama actually gained 4 EV to wind up with 332 EV total -- but that was down from the high of 347 EV he set during the period, and until today (with Virginia's move) had been down to 319 EV. The more ominous trend for Obama was in the Strong Plus Weak category, where he fell from 284 EV down to 257 EV -- a loss of 27 EV (which translates to the loss of Virginia and Ohio, offset by the gain of New Hampshire).
Obama is still looking pretty strong in this snapshot, but he's lost some ground -- and the fallout from the debate hasn't really registered completely yet. Team Obama better hope that the expected losses next week from his debate performance will be balanced with gains due to unemployment getting below eight percent. Even though weakened, Obama still holds a very healthy lead over Romney in the Electoral College at this point. Comparing Strong Plus Weak, Obama still holds a 76 EV edge over Romney, but that is down from a whopping 114 EV lead from last time around -- an almost 40 EV drop.
My Picks
Going more on my gut than just hard polling numbers is tough this time around, because it's hard to know how the data will react in each state. One positive development is that states are being polled much more frequently now, and we have recent polls from over 35 states. This polling frenzy will only intensify as we approach the actual election, making gut predictions easier. As always, different categories are used in this section to avoid confusion. Full lists of the states in each category can be found at the bottom of the column.
Likely States -- Obama
Safe Obama (15 states, 186 EV)
No change in this category from last time around. A case could be made for moving a state or two up to "Safe Obama," but we're holding off to see what happens in the next week. One state (Hawaii) was polled for the first time this year, confirming that it's an easy lock for Obama.
Probable Obama (5 states, 61 EV)
As with Safe Obama, a case could be made that a few more states deserve to be listed as "Probable Obama," but we're going to wait and see. The significant news here is the loss of Virginia -- all the way down to "Too Close To Call."
Likely States -- Romney
Safe Romney (19 states, 156 EV)
As with Obama, Romney neither gains nor loses any states in his Safe category. Two states confirmed this status with their first polls of this election cycle (Arkansas and Louisiana).
Probable Romney (3 states, 25 EV)
Last time around, Arizona moved downwards to just "Lean Romney" due to a bad poll. This time around, it moves right back up to "Probable Romney" due to stronger polling.
Tossup States
Lean Obama (5 states, 43 EV)
As always, most of the action took place in all of the "Tossup States." Starting with the "Lean Obama" category, one state moved out and one state moved in. Florida, at this point, has to be seen as Too Close To Call. I'm feeling good about Colorado, for no concrete reason, so I'm moving it up from Too Close To Call to here for now. A case could be made for either Nevada or New Hampshire moving all the way up to Probable Obama, but I don't feel as strongly about either one quite yet, so they're staying put, as is Iowa. Ohio seems to be teetering on the brink of a move down to Too Close To Call, but for the time being I think Obama still has an edge.
Lean Romney (1 state, 10 EV)
Romney lost one state here, but it's good news for him, as Arizona moves up to Probable Romney. Which leaves Missouri as the only "Lean Romney" state this time around. Missouri's going to be interesting, because of the "reverse coattails" from the hotly-contested Senate race, but so far Mitt seems to be holding his edge.
Too Close To Call (3 states, 57 EV)
Colorado moves up to Lean Obama (which may be unjustified, we'll see), but two Obama states moved back down into Too Close To Call this time -- Florida and Virginia. These, to put it mildly, are going to be crucial states for both candidates in November -- the battlegroundiest of the battlegrounds, so to speak.
Final Tally
Barack Obama's convention bounce fades into the past. Mitt Romney's debate bounce hasn't quite shown its full colors yet. Meaning it's an unstable time in the polling, and things could break either direction in the next week (reminder: these columns will now appear weekly, so check back every Monday from now on).
Obama holds 20 states containing 247 EV by my reckoning, which stacks up favorably against Romney's 22 states with only 181 EV. Out of the 110 EV in the nine remaining battleground states, this means Obama needs 23 EV while Romney has a much bigger 89 EV hill to climb.
Obama can win with only one tossup state -- Florida's whopping 29 EV haul. Even if he loses Florida, he can win with any two of: Virginia, North Carolina, or Ohio. Romney, on the other hand, has to pretty much run the table of the battlegrounds. To put this another way, if Romney won the biggest five of these nine states (Florida, North Carolina, Ohio, Virginia, and Missouri), he'd still need one other state to push him over the finish line. Romney may have had a great debate, but he still has a long way to go before he makes the electoral race anywhere near as competitive as the national polling seems to suggest these days.
[Electoral Vote Data:]
(State electoral votes are in parenthesis following each state's name. Washington D.C. is counted as a state)
Barack Obama Likely Easy Wins -- 20 States -- 247 Electoral Votes:
Safe States -- 15 States -- 186 Electoral Votes
California (55), Connecticut (7), Delaware (3), Hawaii (4), Illinois (20), Maine (4), Maryland (10), Massachusetts (11), New Jersey (14), New York (29), Oregon (7), Rhode Island (4), Vermont (3), Washington D.C. (3), Washington (12)
Probable States -- 5 States -- 61 Electoral Votes
Michigan (16), Minnesota (10), New Mexico (5), Pennsylvania (20), Wisconsin (10)
Mitt Romney Likely Easy Wins -- 22 States -- 181 Electoral Votes:
Safe States -- 19 States -- 156 Electoral Votes
Alabama (9), Alaska (3), Arkansas (6), Georgia (16), Idaho (4), Kansas (6), Kentucky (8), Louisiana (8), Mississippi (6), Nebraska (5), North Dakota (3), Oklahoma (7), South Carolina (9), South Dakota (3), Tennessee (11), Texas (38), Utah (6), West Virginia (5), Wyoming (3)
Probable States -- 3 States -- 25 Electoral Votes
Arizona (11), Indiana (11), Montana (3)
Tossup States -- 9 States -- 110 Electoral Votes:
Tossup States Leaning Obama -- 5 States -- 43 Electoral Votes
Colorado (9), Iowa (6), Nevada (6), New Hampshire (4), Ohio (18)
Tossup States Leaning Romney -- 1 State -- 10 Electoral Votes
Missouri (10)
Too Close To Call -- 3 States -- 57 Electoral Votes
Florida (29), North Carolina (15), Virginia (13)
No polling data since July:
(States which have not been polled since the beginning of August, with the dates of their last poll)
North Dakota (7/11), South Carolina (1/13), Tennessee (5/9), Utah (6/21)
No polling data at all, yet:
(States which have not been polled so far this year)
Alaska, Delaware, Idaho, Kansas, Mississippi, Washington D.C., Wyoming
Electoral Math Column Series Archive:
[Sep 26] [Sep 17] [Aug 22] [Aug 8] [Jul 18] [Jun 25]
-- Chris Weigant
Cross-posted at Business Insider
Cross-posted at The Huffington Post
Follow Chris on Twitter: @ChrisWeigant
Of course the problem is that at this time of the election campaign just as the debate noise is flushed out - around Thurs this week - there is another major event to introduce more noise!
There is no doubt Romney got a bump and right-leaning pollsters were (obviously) quick to accentuate it, making analyzing the effect harder.
I don't know how your EV analysis doesn't see it more clearly though CW. Most EV analysis I looked at before the debate was starting to get to the point where STRONG Obama was >270. Now, post debate, Obama is at the point where strong + leaning is now only just around 270 (your graph does show this but it doesn't show the peak in Strong Obama that happened pre-debate).
Other bumps, just to pick a few I follow:
- Sam Wang @ Princeton Election: Obama pre-debate 6% 'meta (EV) edge'; post 2.2%.
- Nate Silver @ 538: Obama pre-debate around 85% probability to win the Electoral College; post-debate 75%.
- RCP average was around Obama +3.1 pre-debate and is now +0.5 post-debate.
Funnily enough today I have seen people complaining that RCP is right-leaning because of their treatment of the Gallup poll (ignoring the entire poll being +5 Obama by only including in their average the part of the poll that is post-debate - they have never done this before). I expect the shift to Democrats believing pollsters are biased instead of Republicans to be totally complete if Romney (at last) manages to obtain a lead in the national average on RCP in 2012 for the first time!
Personally I think Obama was +3-4% (in Electoral College EV) pre-debate and is probably around 1-2% post debate/job numbers.
The problem Romney has is that Romney v.8.0 (or whatever version he is at now) from the debates appears to be liked (at last!) but he may have changed his position on every single policy too late...
Double posted on Huffpo because my stuff does not always post there.
You slay me not giving the numbers? I use RCP averages:
NH Obama 6
MO Romney 5.2
NV Obama 4.6
IA Obama 3.2
OH Obama 3.0
NC Romney .8
VA Obama .3
CO Obama .2
FL Tie
So assuming NH, MO, and NV don't change that is 257 to 191
That means Romney needs to win OH, VA, NC, and FL and can lose IA and CO
or Romney needs to win IA VA, NC, FL and CO and NEVADA, and can lose OH
But OH is really the one to watch and the 3% Obama lead includes a pre-debate +8 Obama and +4 obama poll and two post Debate polls that are +1 R and +1 O (although one of those was slanted R)
So OH is closer to Romney then it looks and trending.
Bottom line the national Polls are closer to the states and the race is FAR from over.
michty6 -
I really need to add a "this is my source" bit to this column series. My data is from electoral-vote.com.
tinsldr2 -
See my response over at HuffPo, where I mention your reliance on "Romney math"... your numbers just don't add up, sorry.
First case: adds up to 266EV for Romney.
Second case: adds up to 269EV for Romney.
270's the mark to hit...
:-)
-CW
OK folks, [especially new commenter mikesamerica] -
I realize I've been slack once again, but I just went back and answered comments reaching all the way to last Monday, so go check and see what I've had to say about the past week's free-for-all.
:-)
-CW
Notice the shift Pew has made:
The Pew poll's sample of likely voters included 36% who identified themselves as Republicans, 31% as Democrats and 30% as independents. Its September survey included 29% who said they lean more toward the Republican party, 39% Democratic and 30% who said they were independents.
http://online.wsj.com/article/SB10000872396390443294904578044993155023484.html?mod=WSJ_Election_LEFTTopStories
Yes, this is what people have been saying for months in the face of Republican criticism of these 'skewed' polls: party affiliation in polls is very, very fluid! Only Rasmussen actually skews their polls to adjust for their own perception of party affiliation, almost every single other pollster just reports what the people in their polls are actually saying about their affiliation.
Oh sorry I thought you were CW for a second there. Well I've written it out so I'm posting it now...
Hay everyone,
Although I am not quite ready to jump into commenting yet, but I just had to stop in and tell everyone how utterly appreciative I am of the support seen here.
As much as I like to think of myself as emotionless and objective, I have to admit that the comments here brought a tear to my eye...
*All* the comments were appreciated, but I would like to give a sound-out to our host for his kind words and to Ninjaf for her especially touching comment..
It's heartening to see that, when the chips are down, Left, Right, Democrat, Republican, Liberal, Conservative doesn't mean diddly squat...
In the heated discussions that are sure to come in the future, I plan on keeping these last few days' comments uppermost in my mind..
Thank you, one and all....
Michale
Okay, I must've missed something. Is everything okay, Michale? Evidently not. But whatever is going on, I'm sending good vibes your way. xxxooo
Chris1962 -
See:
http://www.chrisweigant.com/2012/10/04/debating-formats/#comment-27699
-CW
An interesting observation, at least to me.
In tracking seven odds makers who predict the winner of the upcoming election (the 538 Blog plus six upstanding off-shore on-line betting establishments), I find surprisingly strong support for the old saw that Republicans can't win the Presidency without winning Ohio.
I've stumbled onto a very simple, yet uncannily accurate, procedure for predicting the relationship between odds of victory in each of the 50 states and DC and the resulting chance of a candidate winning the electoral college. The math behind the observation is a bit complicated, but the procedure itself is very simple.
In column one, rank order the states and DC by descending odds that Obama will carry that state on election day (column 2). In column 3, tabulate the electoral votes associated with each state, in column 4 compute the cumulative electoral votes as you descend column 3. Now, trace your finger down column 4 until you meet or exceed 270 electoral votes. Cross laterally to the probability for that row in column 2, which will almost always be very close to the probability the odds maker gives Obama for winning the election with 270+ electoral votes.
That state, or one of the states if there is a tie, is almost always Ohio! It seems some old saws are mostly true.
Thanks for that, Chris. Check your email.
Thanx CB...
Very much appreciated...
Michale
On another note...
http://news.yahoo.com/state-dept-reveals-details-benghazi-attack-062900114.html
It seems the official story from the Administration is more in keeping with the reality of what occurred in Benghazi..
There was no protest. Everything was quiet when the attack happened...
This is a major black eye for the Administration's vaunted Foreign Policy prowess...
On the plus side, however, it IS heartening to see the Obama Administration come clean and tell the American people the truth about what occurred. It's likely that Obama was under intense pressure to just ignore the incident and hope it fizzles out. The fact that Team Obama told the truth, THIS close to the election, is a plus in the INTEGRITY column for Obama..
Michale....
COMMENT NOTE: From now until the election, I am going to try something new. For every slam against the Obama Administration I post, I am going to also try and come up with something positive, something good that Obama has done.. Hopefully, this will show a more even keeled approach and give credit where credit is due.
We'll see how it works out...
Michale
The University of Colorado has updated their predictions that are based on their algorithm that has accurately "picked" Electoral College picks 89% of the time...
http://www.colorado.edu/news/releases/2012/10/04/updated-election-forecasting-model-still-points-romney-win-university
The updated prediction shows an even bigger win for Romney and was completed prior to the PR debacle that has hit Team Obama in the last few weeks...
It should be noted that this study is *peer reviewed*...
While *I* am of the opinion that the peer review process doesn't mean as much as it used to, I know that ya'all are really big on the peer-review stamp, so that's why I mention it...
It should also be noted that this study of the Electoral College approach is the only one of it's kind in depth and broad spectrum analysis, which might explain why it's the only one that has this prediction.
Time will tell how accurate the prediction is, but if the algorithm is as accurate as it's past "predictions", well.......
An 89% success rate is hard to argue with...
Michale.....
As far as the Jobs Report numbers.....
http://online.wsj.com/article/SB10000872396390444897304578046260406091012.html?mod=WSJ_Opinion_LEADTop
Do I believe that the Obama Administration would cook the books to make Obama look better??
Abso-frakin-loutly...
Do I believe they did it in this case.
No I don't... They would be stoopid to do it for the exact reasons that we have seen. They would be caught......
But, that's not really the issue... The issue, I believe, is the idea that the Right is being crazy just SUGGESTING it..
Postulate a scenario where President Bush had just gotten royally shellacked in a debate by a Democrat opponent..
Further postulate that, a couple days later, an unprecedented and nearly impossible drop or rise in a statistic that benefited Bush occurred...
Is it outside the realm of possibility that the entire Left would react in the exact same manner the Right acted in the here and now?? Back in 2003, a current Obama official himself suggested that the Bush Administration had "cooked the books" in response to a favorable Unemployment figure.
I think it's safe to say that we ALL agree that the Left would react in the exact same manner that the Right has, given the same set of circumstances...
In other words, while it's unlikely that Team Obama would really cook the books, the suggestion that the books COULD be cooked is not the hysterical irrational paranoia that some would make it out to be...
Michale.....
In keeping with my old-but-re-affirmed fair and balanced attitude... :D
foxnews.com/politics/2012/10/09/ap-publishes-unflattering-pic-romney-bending-over/
AP is no more or no less guilty of political bias than the Right was when the Right made an issue of this picture from Obama..
oddee.com/_media/imgs/articles2/a97675_Barack-Obama-staring-at-Mayora-Tavares1.jpg
In other words, we've got a lot of problems MUCH bigger than this...
Michale.....
http://www.realclearpolitics.com/video/2012/10/09/actress_stacey_dash_on_receiving_racist_hate_for_supporting_romney.html
Where is the condemnation from the Left???
Michale.....
Where is the condemnation from the Left???
I mean, Obama called up Fluke when she was attacked by the Right....
Is Obama going to call up Dash??? I mean, after all, there are no RED states, no BLUE states, only UNITED states....
Right????
On the plus side.... At least Obama didn't refuse to buy Dash a beer...
(lame, I know, but that's all I could come up with... :D)
Michale......
As hard as it may be to believe, I can forgive Obama a lot of things...
http://www.thegatewaypundit.com/2012/10/confirmed-obama-sent-same-form-letter-to-parents-of-all-fallen-soldiers-marines-seals/
But not that.....
Michale.....
Michale [17, 18] -
I'm in the middle of posting today's article, just have a second...
You might want to reconsider your example. Fluke not only condemned the comments, I think she actually called Dash up personally. I can provide a link, just too rushed right now.
-CW
You might want to reconsider your example. Fluke not only condemned the comments, I think she actually called Dash up personally. I can provide a link, just too rushed right now.
If true, that is to her credit...
But that doesn't excuse the hypocrisy from Team Obama over this???
That doesn't excuse the silence from the Left in general..
Where's the condemnation from everyone here??
If you are a black woman who supports a conservative candidate or philosophy, you are vilified in the most vile and racist manner imaginable.
And the Left (from President Obama on up to everyone here) seems to be OK because, after all, silence gives assent...
The Left would go utterly and ballisticly batshit if a Democrat was treated in such a manner by the Right.. Not that it has ever happened in recent history, but if it had, the Left would lose their minds.
Yet, when elements of the Hysterical Left acts in such a vile, disgusting and racist manner, the Left overall, doesn't say diddly...
So, the ONLY logical conclusion is that the Left really isn't against racism at all.. The Left is just against racism when it comes from the Right. When racism comes from the Left, it's A-OK with the Left.
It's the ONLY explanation that fits the facts...
But, I am willing to admit I am wrong. I am hoping like hell that I *AM* wrong..
Does anyone have any links from Left Wing blogs/sites/commentaries that condemn the vile and racist treatment of Dash and Love???
Michale......
Does anyone have any links from Left Wing blogs/sites/commentaries that condemn the vile and racist treatment of Dash and Love???
I didn't think so....
{sssiiiigggggghhhhhhh}
That's really a shame...
Michale.....
I have a link to some NEW vile coments made by the Left. Warning: They're downright disturbing.
http://twitchy.com/2012/10/12/left-wing-sicko-hopes-someone-impregnates-paul-ryans-13-year-old-daughter/
I have a link to some NEW vile coments made by the Left. Warning: They're downright disturbing.
http://twitchy.com/2012/10/12/left-wing-sicko-hopes-someone-impregnates-paul-ryans-13-year-old-daughter/
Ya know, this really doesn't bother me, beyond the natural disgust that ANY human being would feel..
What REALLY bothers me is that no one on the Left takes this kind of garbage to task, UNLESS it comes from the Right..
What really REALLY bothers me is that the Left pretends that THEY are all goodness and light and it's only the RIGHT who wallows in this kind of crap...
But, as usual, the facts say different...
Michale.....