This post is mostly for family and friends looking for some insight while watching the numbers come in, but I thought I'd bring my expert political geek side over to the blog to help anyone looking to make sense out of the results of the US Presidential Election on Election Night. I'm releasing it early so you can study it and impress your friends! I follow the numbers daily so somebody else might find them helpful, too. You know the cast so we'll jump right into what to look for...
The easiest way to put it is that there are really only 6 states that are being fought over: Virginia, New Hampshire, Ohio, Florida, Iowa, and Colorado. (You can maybe throw Nevada in there--Romney would like to think Wisconsin--but the "math" is in the other states for the most part.)
The bottom line for the casual viewer is that if any of the big "eastern battlegrounds" go to Obama then Romney has been given a very tough path for the rest of the night: OH, VA, FL.
There is, of course, the possibility that Obama could win Ohio and NH and then Romney could somehow pull out victories in Iowa, Colorado, and Nevada...a 269-269 tie if Romney also won FL and VA.
And it goes without saying that any strange issues like Romney winning PA or Obama winning NC would set the tone for an Electoral landslide, but both those are unlikely.
If I were forced to put money on the line, I'm guessing the most likely outcome is going to be the Obama-wins-Ohio-plus-Nevada/Iowa combination. That's statistically the best guess.
A word about the odds...Personally, when weighing the different probability systems, I tend to view a sliding scale where numbers in the 50% range are what we genuinely call "tossup," numbers in the 60% confidence range are lean/likely, and we can be fairly certain of anything above 70%.
Ohio
Ohio, for instance, had a ruling yesterday which will open the polls for the 3 days before Election Day. In 2008, something like 100,000 mostly Democrats voted in those 3 days and that ruling is seen as a strong influence on which way Ohio will go. Ohio is right on the high-60's but near 70% range, but for me that ruling pushes the odds pretty low for Romney. The story in Ohio is whether Romney can make enough gains in Early Voting and Election Day voters to break the Obama Early Voting advantage.
Tossups
As far as the other contested states, the true tossups appear to be VA and CO. My gut hunch is telling me they will split those 2 stateswith Romney getting CO and Obama getting VA. (OH plus VA would certainly make a nice, early win for Obama.) Both VA/CO are in the 50% range and will probably continue to flip back and forth between the two. My gut tells me FL is going to go for Romney, but what is actually going on is up for debate. Early Voting turnout is high. Democrats have a registration advantage. It may be close depending. Others have tried to put Iowa in the tossup category, but the advantage here is probably a little bit more towards Obama.
Lean/Likely
New Hampshire and Iowa are tricky and I had conversations about them with a friend yesterday. They both lean Obama, but present us with a mix of polling despite Democratic "state fundamentals." Nevada it depends on who you believe. It's a tossup, likely, or firm Obama because some forecasts have it approaching nearly 80% confidence. When you look at the state fundamentals it has a strong built-in Democratic lean, but as with the Romney lead in FL I wouldn't count any chickens before they hatch.
Weirdness
I'm not going through ALL the situations possible here. This is just a guide for what to watch for. But I wanted to mention 2 things specifically and they both relate to Tipping Point. The underlying assumption is that if you win certain states you probably have won others. A good example is Mitt Romney's talk this week about how they feel PA/MI are not possibilities and what paths to 270 they might have if Ohio is a blue state. They insist Wisconsin is on the table then..they'd be hoping to win NH, IA, etc.. But the problem is the same, really, as Obama losing Ohio. If a state falls, it is a reasonable guess that the metrics at work will give you the more likely states down the line. It's not unthinkable, but would be highly...weird...for Obama to lose Ohio, but win VA or FL or CO. Which is what makes a few states so key, must win, etc.. If you'd like, there are a few websites that publish the chart of Tipping Point for each candidate. Sometimes it is the same state, often it isn't. Lately it has been Nevada or Ohio for both.
That's really the President's advantage this election is that he doesn't need any of the states currently unlikely to go his way. All he needs to win are the states thought of as blue anyway. The pressure is on Romney to gain turf, Obama simply needs to hold his ground. But this says nothing about turnout. A high turnout in FL because of its population could make all the math moot if Obama has those EV in his pocket.
Final score?
My forecast for Election Night is that it will probably end up Obama 281 to 294 versus Romney getting somewhere between 235 and 257. (For reference, the Tipping Point final score for a Romney win would be 273 to Obama's 265.) The stats folks are a little more optimistic about Obama giving him over 300.
I'll update this post in the march to Election Day if the underlying foundation moves much.
PS I've said nothing about Congress, but I'd be surprised if the Republicans lost the House or Democrats lost the Senate. Whoever is President for the next term is most likely to have a split Congress to deal with. (I'll amend this to say that there are a FEW of the folks who are predicting 300 and above for Obama who think the races may break for a Democratic House. I'm suspicious of this. I'm remaining unmoved until I start seeing big Blue State advantages coming in next Tues.)
The easiest way to put it is that there are really only 6 states that are being fought over: Virginia, New Hampshire, Ohio, Florida, Iowa, and Colorado. (You can maybe throw Nevada in there--Romney would like to think Wisconsin--but the "math" is in the other states for the most part.)
The bottom line for the casual viewer is that if any of the big "eastern battlegrounds" go to Obama then Romney has been given a very tough path for the rest of the night: OH, VA, FL.
- With Virginia, Obama would need just Iowa or NH to win
- With Florida, Obama would cross the 270 Electoral College threshold by itself--no extra help needed
- With Ohio, Obama would need to simply close the deal in NV to win
There is, of course, the possibility that Obama could win Ohio and NH and then Romney could somehow pull out victories in Iowa, Colorado, and Nevada...a 269-269 tie if Romney also won FL and VA.
And it goes without saying that any strange issues like Romney winning PA or Obama winning NC would set the tone for an Electoral landslide, but both those are unlikely.
If I were forced to put money on the line, I'm guessing the most likely outcome is going to be the Obama-wins-Ohio-plus-Nevada/Iowa combination. That's statistically the best guess.
A word about the odds...Personally, when weighing the different probability systems, I tend to view a sliding scale where numbers in the 50% range are what we genuinely call "tossup," numbers in the 60% confidence range are lean/likely, and we can be fairly certain of anything above 70%.
Ohio
Ohio, for instance, had a ruling yesterday which will open the polls for the 3 days before Election Day. In 2008, something like 100,000 mostly Democrats voted in those 3 days and that ruling is seen as a strong influence on which way Ohio will go. Ohio is right on the high-60's but near 70% range, but for me that ruling pushes the odds pretty low for Romney. The story in Ohio is whether Romney can make enough gains in Early Voting and Election Day voters to break the Obama Early Voting advantage.
Tossups
As far as the other contested states, the true tossups appear to be VA and CO. My gut hunch is telling me they will split those 2 states
Lean/Likely
New Hampshire and Iowa are tricky and I had conversations about them with a friend yesterday. They both lean Obama, but present us with a mix of polling despite Democratic "state fundamentals." Nevada it depends on who you believe. It's a tossup, likely, or firm Obama because some forecasts have it approaching nearly 80% confidence. When you look at the state fundamentals it has a strong built-in Democratic lean, but as with the Romney lead in FL I wouldn't count any chickens before they hatch.
Weirdness
I'm not going through ALL the situations possible here. This is just a guide for what to watch for. But I wanted to mention 2 things specifically and they both relate to Tipping Point. The underlying assumption is that if you win certain states you probably have won others. A good example is Mitt Romney's talk this week about how they feel PA/MI are not possibilities and what paths to 270 they might have if Ohio is a blue state. They insist Wisconsin is on the table then..they'd be hoping to win NH, IA, etc.. But the problem is the same, really, as Obama losing Ohio. If a state falls, it is a reasonable guess that the metrics at work will give you the more likely states down the line. It's not unthinkable, but would be highly...weird...for Obama to lose Ohio, but win VA or FL or CO. Which is what makes a few states so key, must win, etc.. If you'd like, there are a few websites that publish the chart of Tipping Point for each candidate. Sometimes it is the same state, often it isn't. Lately it has been Nevada or Ohio for both.
That's really the President's advantage this election is that he doesn't need any of the states currently unlikely to go his way. All he needs to win are the states thought of as blue anyway. The pressure is on Romney to gain turf, Obama simply needs to hold his ground. But this says nothing about turnout. A high turnout in FL because of its population could make all the math moot if Obama has those EV in his pocket.
Final score?
My forecast for Election Night is that it will probably end up Obama 281 to 294 versus Romney getting somewhere between 235 and 257. (For reference, the Tipping Point final score for a Romney win would be 273 to Obama's 265.) The stats folks are a little more optimistic about Obama giving him over 300.
I'll update this post in the march to Election Day if the underlying foundation moves much.
PS I've said nothing about Congress, but I'd be surprised if the Republicans lost the House or Democrats lost the Senate. Whoever is President for the next term is most likely to have a split Congress to deal with. (I'll amend this to say that there are a FEW of the folks who are predicting 300 and above for Obama who think the races may break for a Democratic House. I'm suspicious of this. I'm remaining unmoved until I start seeing big Blue State advantages coming in next Tues.)