Science, Statistics, Politics, Current Events, Photos and Life.

Sunday, November 30, 2008

Beating a path for Obama

Bill Bradley, former Senator from New Jersey and hall of fame basketball player deserves credit for lighting the path to the presidency for Barack Obama. Bradley first put 3,028,451 cracks in the highest ceiling in the land during his historic run for the democratic nomination for president in 2000. Bradley's candidacy paved the way for basketball-playing Barack Obama to run and win his party's nomination for president.

Could we therefore see a female candidate, not necessarily H. Clinton, in the year 2016?

Grading 2008 election prognostications on a curve

Mark Schmitt at The American Prospect grades various theories of how the election turned out. Less numerical than our evaluation.

I think there are some verbal miscues in my previous posting, but I think I'm just gonna leave 'em alone.

Friday, November 28, 2008

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.

Monday, November 10, 2008

Today's sub-Headline

From a teaser at the LATimes online:

At Henri-Chapelle in Belgium, Meuse-Argonne in France and Sicily-Rome, countless tombstones tell the stories of those who gave their lives in battle during both world wars.

Bad enough to give your life in one world war.

This was a teaser for an LA Times article on European cemeteries for US war dead. They didn't use that sentence elsewhere on the web site as far as I could see.

Saturday, November 8, 2008

California Prop 8 and Public Health

California's proposition 8 passed, outlawing same sex marriage.

This passing will have negative public health implications. Allowing male homosexuals to marry could (should?) spread monogamy into the gay community and could therefore substantially reduce HIV transmission among men who have sex with men. Married & monogamous homosexuals in turn can act as role models for younger homosexuals.

In California, the primary mode of transmission of HIV is gay sex. In Africa, it is heterosexual sex, and in Russia, as I understand it, it is needle sharing in the intravenous drug community. HIV has several ways of spreading, it is not just a homosexual disease.

Removing stigma against homosexuality will allow older homosexuals to find a path to happiness and health (lots of sex with lots of different partners is not a path to health, and I would guess it is not a path to happiness either). In turn, having older homosexuals as positive role models for younger homosexuals will allow positive health behaviors to take root in the homosexual community.

Friday, November 7, 2008

Ayers the Education Professor on the Election

We'll be seeing more careful interviews of, and comments from, William Ayers and Jeremiah Wright in the coming days and months. After the abuse heaped upon them by the McCain campaign and independent organizations hostile to Barack Obama's candidacy, it will be time to straighten out the record.

The Washington Post published a small interview with Ayers. Some selections:Asked Tuesday if he wishes he had set more bombs, Ayers answered, "Never." He also said he had regrets.

"I wish I'd been wiser," he said. "I wish I'd been more effective. I wish I'd been more unifying. I wish I'd been more principled."

History has shown of the Vietnam War that "those who opposed it were on the right side," Ayers said. But he said some of his early rhetoric was "juvenile."

This certainly contradicts the NY Times quote.

Ayers blames the "liberal media" for failing to dismiss the Republican assaults. He called the media's performance "kind of shameful" and likened the situation to the 2004 episode when Swift Boat Veterans for Truth created a narrative that helped doom the candidacy of Sen. John F. Kerry (D-Mass.).

"The dishonest narrative," Ayers said, "is that guilt by association has some validity."

This calls out the media and all those claiming that the media has a liberal bias (O'Reilly? Drudge? Coulter? Limbaugh?). A fair and unbiased media that had some spine might have fact-checked the Ayers meme a little more accurately. Most of what I saw, on supposedly liberal media replayed Republican ads with little in the way of discussion.

William Ayers, the former Weather Underground leader who became an issue in the 2008 campaign, said Tuesday that he is not close to Sen. Barack Obama and that Obama's opponents had turned him into "a cartoon character."

Worse things than being a cartoon character...

Beep Beep!



Wednesday, November 5, 2008

4 Senate seats and 10000 votes

There are 4 senate seats without a currently declared victor: Minnesota, Alaska, Oregon and Georgia. In MN, AK and OR, the currently leader is ahead by 690, 6187, and 3353 votes. In Georgia, the leader requires 50% of the votes to win outright without a run-off election, and Chambliss is 6250 (or 6251, who knows) votes short of the 50% mark.

In all four races, the Republican is ahead. Chambliss, without major interference, is likely to win the run-off election, as the Libertarian candidate votes will likely go to him. In Alaska, incumbent and current convict Ted Stevens is beating Begich. In Minnesota, currently Norm Coleman is ahead of comedian Al Franken, while in Oregon, Smith is edging out Merkley.

That's a total of 690+6187+3353+6250 = 16480 votes. This is somewhat of a reprise but in the Republican direction of the 2006 senate elections. There, 10078 votes was the total winning margin for the Democrats in the Montana and Virginia senate elections.

A lot of close elections: it's important to vote, you never know when your vote is needed.

Update. I thought CNN was reporting 99% or 100% in those races when I took the data. Several people I talked to did try to correct my reading of the data. Oh well. Currently I hear there are possibly 50-100 thousand additional votes in Alaska. Also, margins have changed. Minnesota: 477, Oregon:3932, Alaska unchanged at 3353 and Georgia is also unchanged needing 6250 votes. A total now of 14012 vote differential.

Update 2. Even as I wrote that, CNN updated data for Oregon and the democrat is now ahead by 6129 votes! Quite a reverse from being down 3932. That's based on 78% precincts reporting. I give. up. No more updates. Point still holds, but illustration would keep changing.

Barack Obama sent me an email!

I personally got the following email From Barack Obama
Subject: How this happened
XXXXXX --

I'm about to head to Grant Park to talk to everyone gathered there, but I wanted to write to you first.

We just made history.

And I don't want you to forget how we did it.

You made history every single day during this campaign -- every day you knocked on doors, made a donation, or talked to your family, friends, and neighbors about why you believe it's time for change.

I want to thank all of you who gave your time, talent, and passion to this campaign.

We have a lot of work to do to get our country back on track, and I'll be in touch soon about what comes next.

But I want to be very clear about one thing...

All of this happened because of you.

Thank you,

Barack
I appreciate the thought and the email. Even if it did go out to me and 3 million of Barack's closest personal friends. This is like the old FOB (friends of Bill) from the Clinton presidency, but on steroids.

The "I'll be in touch soon about what comes next" is disturbing, unsettled and vaguely ominous.

Interesting note 1: A DONATE button at the bottom of the email. Count on requests for money for the next 4 years.

Interesting note 2: I can't quite interpret all the header information regarding the dates, but there are four time stamps (I edited out most all other info) on the email:

From Barack Obama Wed Nov 5 04:34:13 2008
Received: from with SMTP; Tue, 04 Nov 2008 20:36:57 -0800
Received: Tue, 4 Nov 2008 23:34:13 -0500 (EST)
Date: Tue, 4 Nov 2008 23:34:13 -0500
I don't know where that Wed date came from, because I got the email Tuesday evening. But these guys are fast.

Fortune Cookie Fortunes

Work on ideas that are creative and can bring fine results.
Good advice for an academic statistician.
Your courage will reap rewards for you.
Sensible direction for an academic statistician.
You display the wonderful traits of charm and courtesy.
Useful prodding for an academic statistician.
Your lucky number for this week is the number three.
Huh?

How did they know that I would get that fortune cookie?
And get it yesterday in particular?
And what is it about this week that it is lucky?
Does the week begin yesterday when I got the fortune, or on Sunday as is tradition, or is this referring to the work week, starting Monday and ending Friday?
And just how shall I make use of this information?
Betting on craps?

Bill Kristol and the underendowed

Neo-conservative Bill Kristol, former chief of staff to Dan Quayle, was quite taken with Sarah Palin and was an early promoter of hers. He seems to have an affection for under-intellectualized politicians.

Tuesday, November 4, 2008