2008 State by State Electoral Predictions Compared

The 2008 election is over and Obama solidly trounced McCain. Several sites kept a running total of polling results by state, including pollster.com and RealClearPolitics. Nate Silver at fivethirtyeight.com went a step further, analyzing the polling data and explicitly making predictions of the voting totals by state. I also found RJ Elliot who gives a state by state prediction of the voting percentages for Obama and McCain. There are no doubt other sites with predictions. I found one, but they wanted money to view their predictions, so they are omitted from this comparison.

I took the final estimates from these four sites to compare to the actual vote totals. A caveat should be kept in mind: Silver was making an explicit prediction about the election, while RealClearPolitics and pollster.com are aggregating and summarizing polling data, though presumably with the purpose of predicting the election. RJ Elliot also was making predictions, though I didn't see any explanation of his methodology. For actual results, I took data from Dave Leip's Atlas of U.S. Presidential Elections. I didn't try other sources, this was the first that I found.

Final cavil: I collected fivethirtyeight.com and RealClearPolitics' predictions on the day after the election. The pollster.com and RJ Elliot predictions were collected later, on Thanksgiving or the day after. While I don't think have any suggestion that any of these predictions were changed after the election, there certainly is the possibility of that happening.

Results
Herewith the results. I took the predictions for Obama and McCain's state by state percentage, subtracted from the actual percentage, squared the difference, averaged across states, and took the square root. This gives a root mean square error (RMSE) of prediction difference from truth for both the Obama side and the McCain side of the estimates. The results are in the image. Hopefully the image shows up in the right place.


For each site there are two results, for Obama and for McCain. For each row there are two numbers. It turns out that RealClearPolitics (RCP) only predicted results for 38 states out of the 50 states plus DC. The others predicted for all 51. The first column (starts 3.0, 2.9) includes all 51 states plus DC plus the national vote treated as a 52nd state. The second column is the predictions only on the 38 states where RCP made predictions. So the first column you can compare 538, pollster and Elliot. In the second column you can compare all 4 sites.

The states where RCP did not make predictions were the very partisan and small states where the winner was very clear throughout the entire election (think Utah and DC) and where there were very few polls. Both 538 and pollster do much better in the accuracy of their predictions for the subset where RCP was making predictions. On the 38 states where RCP made predictions, we see that 538 had the smallest RMSE followed by pollster then Elliot with RCP having the worst predictions.

Across the 52 predictions (50 states + DC + national), Elliot edges 538 perhaps barely, both of which definitely beats pollster. The RCP result in the first column is not comparable to the other numbers. I inspected the individual residuals (differences prediction minus reality) in the states where RCP did not make predictions. These residuals were often very large. The largest residuals for 538 were from the states (Alaska, Arkansas, DC, Hawaii, Louisiana, Nevada, Vermont, Wyoming), defined as being in error by more than 4% on either Obama's or McCain's vote total. DC, Hawaii, Louisiana Vermont and Wyoming were not predicted by RCP.

Election 2008: Tie Ball Game

At pollster.com, they have their own calling of the presidential race based on current state polling. Their count has 208 electoral votes for McCain, 202 for Obama and 128 electoral votes in toss-up states. If you go through the toss-up states, there are 61 votes for McCain and 67 for Obama, leading to an exact electoral college tie.

Here are who is ahead in each of the 128-electoral-vote-totaling toss-up states:
JM
MT - 3
NV -5
IN - 11
OH-20
WV - 5
VA - 13
NH - 4

BO
CO - 9
MN - 10
WI - 10
MI - 17
PA - 21

Total: 67 Barack Obama, 61 John McCain
Added to 202 BO and 208 JM
gives us a 269-269 tie.

Update: Was still tied yesterday using the same mechanism, though the strong and leaning Obama and McCain states had shifted quite a bit. And I did a little editing above.

Red Meat and How Obama Needs to Run Against McCain

I've been feeling that Obama's campaign has been missing the red meat of attacks at Bush. Tonight Tuesday at the Democratic convention, we've been getting some well placed licks in, but there needs to be more.

When people complain that they don't know what Obama stands for, I think they really mean, lay out what Bush stands for and why you are different. Not just who you are for, but what you are for, what you will do and most importantly how that is different from who Bush is for, what Bush has done and what Bush wants for the country.

TalkingPointsMemo has an excellent further discussion that they need to really explain that McCain is worse than Bush. Not the same as, but much much worse. As a marketing gimmick, 'the same as' is bland and boring. The Democrats need to explain what Bush is for, what McCain is for and why that will be even worse for the country than Bush. Not just as bad as, but much much worse.

Why McCain is dangling Pro-Choice Ridge & Lieberman

Why is McCain dangling pro-choice veep prospects Ridge and Lieberman in front of the hard right anti-choice crowd? Mark Halperin on CNN was saying that he thought McCain was softening up the wing-nuts to more easily accept Minnesota governor Tim Pawlenty or former Massachusetts governor Mitt Romney. Romney was pro-choice until he switched to run for president as a Republican.

Whether this is true or not, I propose a second benefit from this major dangle: it speaks to the uncommitted electorate McCain is a `maverick' and `independent' in that he could support a pro-choice running mate. It makes a play for voters for whom being anti-choice is a deal breaker. I believe that McCain's anti-choice position is not widely known. Even during his interview with Warren at the Saddleback church, people could overlook what his beliefs on choice will do to abortion availability in this country.

This is all about politics. It is a play for voters, and not a true tell on his feelings about abortion.

Leaking His Lead Away: Obama Versus McCain

I imagine the Republican party, while not happy about the state of Senatorial, Gubernatorial, Congressional etc races, can't be unhappy in the slightest about the presidential race.

McCain has kept it close against Obama. Obama has had a modest lead for some time since he clinched the nomination, but the lead has slowly been leaking away. Pollster.com has Obama at +2% or so at the moment over McCain. Suppose we have a continued leakage away.

Coming up, we can expect the conventions to lead to a slight McCain bounce. There is usually a bounce for each candidate after the conventions, but the Republicans always hold their convention second so as to squash the Democratic bounce. I predict that will happen again.

We have the usual Republican right-wingnut parade of ridiculous charges against Obama. We'll see if that has any effect on the race. It seems to me that the best the democrats can hope for is for no effect. I don't see how this can be a positive effect for Obama. Obama has been fighting this quite strongly, but who knows if it can be strong enough. There are a 1-2% (at least) of voters likely to be swayed by all this.

We have the possibility (probability?) of racial effects on voting versus preference. This has variously been called the Bradley effect among other things in American politics. This will be unmeasureable even after the election, though it may get blamed for any Obama loss.

This analysis implies that Obama needs something to lead a move back towards himself. The democrats will do fine at all lower levels of the voting.

Team Obama is investing heavily in GOTV (get out the vote) efforts, more so than in all previous elections. That should match the Republicans in their GOTV, but the Repubs have a lot more experience. This is one of the key blessings of Obama's background as a community organizer, the realization that individual actions can be put together to create a large movement.

There will be the debates. Obama was never seen as doing as well as Hillary, but McCain is a verbally skillful fellow and will quite possibly more than hold his own in the debates. Obama likes to be nuanced - McCain will be blunt. And blunt often sells well in politics, even idiotic bluntness.

Other possible game changers: seems unlikely. Any verbal gaff by Obama will be leaped on by the Repubs. In contrast, McCain gaffs have been routinely ignored by the media. I guess team Obama needs to start spreading the McCain news around more -er- liberally.

McCain could have a Dole-type moment showing his age. Recall that Dole fell off a small stage in Chico California. This is possible, but there isn't much time left for this. And you can't plan for it.

Money-wise, McCain may seem to be raising less money than Obama, but the Republican party will be spending most of their cash hoard to support McCain, and money-wise the race currently seems about even. Even if Obama manages to edge ahead by 70million or so, that won't be enough to generate anything close to a land slide. Could it turn in to a substantial edge? I don't know. Small edge? Maybe.

General conclusion: the presidential race is a toss up.

Clinton to Fight for Vice President Slot

Talkingpointsmemo.com has a not that Clinton has explicitly authorized an advisor/supporter (the head of BET, Black Entertainment TV) to help her run for the vice presidential spot on the ticket. That is a lot of hubris. Commentators on CNN last night after her non-concession were saying that if she had a 50-50 slot at the veep slot, now it was 80-20 against.

Obama's problem with her joining the ticket is that he doesn't look presidential if she takes the veep slot, rather, she looks like she grabbed it whether he wanted to offer it or not.

Her goal in this is not necessary to have Obama win. It is to have her in the strongest position to run for president next time.

Other news reports suggest that Obama would be willing to have her in his cabinet, and to take the lead on health care legislation. Then if it doesn't go through, its her fault. And if it succeeds, she was a success, and so is he.

Does it appear there is a negotiation going on in the press over her place in the Obama administration? He'll offer cabinet, she'll take vice president.

I guess he's getting his first test on talking to hostile leaders: should he talk to Hillary directly without preconditions?

Obama's wealth rises rapidly

LA Times has a story today on Obama's wealth rising rapidly.

Barack Obama's wealth has more than doubled during his presidential campaign -- and has shot up tenfold since he entered the U.S. Senate three years ago, his financial disclosure filed Friday shows.

The easiest way for your wealth to shoot up tenfold or to double in a year, is to start with next to nothing. If you have a dollar today, and next year you have a hundred, your wealth just shot up 100 fold. But don't go congratulating yourself yet.

Addendum May 18, 2008:

The correct statistical interpretation of the quote data unquote presented in this article is that Obama was far and away the poorest of the 3 (and then there were 2) presidential candidates 3 years ago, and 1 year ago. The implications of the article are that he is on the fastest upward track, hence he is the elitist with the most-est...

I Got That Wrong

Indiana barely goes for Clinton.

North Carolina is a strong result for Obama.

I didn't predict that very well at all. Zogby actually did quite well with their late polling estimates.

Why? My guess: the biggest issue on the table, gas tax holiday pushed late deciders to Obama.

In contrast, at the last few primary rounds, the big issue on the table, Rev Wright, bittergate, pushed late deciders to Clinton.

Pennsylvania Primary in One Day

Pollster.com has Hillary ahead by 7% with approximately 7% undecided. In the absence of a major shift, 7% undecided are not all going to Obama. And where has his "Yes we can" gone? I haven't heard it in weeks. Since bitter-cling-gaffe in fact.

Before any one was paying attention, she was ahead by 20%. As more attention was paid, the gap naturally narrowed to about 12% and in the last 2 weeks, maybe Obama closed to within 5% of Clinton. But if I can tea-leaf-read the wiggles, that has now expanded again. We'll see if the undecideds again break for Clinton in a big way. I'd guess 10% final margin for her, but I don't really put a lot of stock in my prognostication.

Look to Obama's camp pointing out that this was friendly Clinton territory, and that she was ahead by 20% not long ago. All true.

Look to the Clinton camp to point out that he didn't win by 20% and that he outspent her by 2 or 3 to 1. All true.

The truth is it is a big state. Big states are harder to change people's preferences merely because of their size. And like every state in the country, when the campaign started, she had a sizable advantage. When Obama wins a state, he has overcome, at some point in the distant past, virtually zero percent name recognition, and near complete Clinton domination. (I'm ignoring Edwards in all this discussion. ) So figure that when he wins, he needs to pick up the entire margin of victory during the run up to the primary or caucus. Hillary has merely needed to maintain her edge, which is generally easier, and which she has mostly managed in states where she lives, has lived, neighbor states and states that are large where it is difficult to get Obama's message out.

The networks are playing this as a "Will Obama win? Will Clinton Lose?" But Clinton has won, they just want you to tune in for the exciting finale for 4 hours on their news channel.

I'll be glad when this one is over.

Time to move on to Indiana and North Carolina.

I'll be glad when the primary is all over.

Ask a Trashy Question

The ABC silly-debate discussion is still reverberating around the blogosphere. I hope the echo deafens Gibson and Stephanopoulos.

The next time a moderator asks a stupid gotcha question, the candidate should turn the question around and accuse the moderator of promoting that repugnant point of view. If you ask gotcha questions, you are promulgating those canards. Moderator asks: You know someone who said something mean. (Implying therefore you're mean). How do you respond senator? (Or more likely, disavow this person Senator!)

Senator: Well Mr. Moderator, you should not be bringing up nasty bits of meanness unless you seriously endorse it. You are directly promoting the position of those people who are bringing this issue up over and over. Is that your intention? The people banging this issue like a drum are not interested in the answer to your question. They are interested in occupying my time with inanities and filling the voters ears with fear mongering hate messages. They are interested in obscuring the profound policy differences between me and my opponents. They and you are not interested in any answer to any foul canard posed as a question. They are solely interested in sliming me so as to steal from the American people an otherwise easily won on the merits of my positions election. (ok, so not that last in so many words).

By raising the issue you are lowering the quality of our national debate. You should be asking us important questions about Iraq and international policy, global warming and the environment, how I will run my White House, what goals I will set for my administration. Why are you asking about the nasty remarks made by that person? Do you ask John McCain about Reverend Hagee? Will you? Would you? If not, and I doubt you will, then you are engaging in partisan politics, under the false guise of non-partisan-pretense.

You should be ashamed of your self Mr. Moderator. Slimy attacks do not make America a better country and your pretense that these attacks go to my electibility are not obscuring your partisanship to the American people.

October Surprise:Iran

Someone important thinks that we're in for a Fall, possibly October surprise. Pat Buchanan thinks we're likely to start bombing Iran.

Another possible October surprise: From the LA Times: Finding Bin Laden.

The Bush administration may go for a twofer. Both will substantially help McCain over the Democratic nominee.

Crook to Prosecutor: The 2009 White House Changeover

Supposing a democrat wins the election, are we all certain that Bush and Cheney are going to go gently into that dark night?

The next president, if democrat, will be picking up the pieces of the last 8 years of policies that he or she strongly disagrees with. At least some of those policies are going to appear to be covering criminal activity: those billions of dollars going in airplanes to Iraq, nobid contracts to Halliburton, signing statements, torture, Cheney's secret energy negotiations, vice presidential visitor logs, paying reporters to report biased news, leaking CIA agent's identities, lost emails, cronyism. The list will be long.

Cheney and Bush and advisers might well be worried about criminal prosecutions. From the Cheney-Bush perspective, the cats will now be in charge of the hen house.
While deleting lots of emails is one way to interfere with prosecution, and taking massive quantities of documents with them is another, there is still likely to be plenty of evidence lying around. Its hard to commit crime on this massive of a scale without leaving some evidence lying around.

I recall a statement in the news from someone at the Pentagon during the night that Nixon resigned. The statement was to the effect that the Pentagon was making certain that all troop movements were authorized and previously scheduled, and that no extra commands that skipped the normal chain of command.
Presumably we will have the same situation in 2009.