Tuesday, November 02, 2004

Props to my polling place

The presidential election looks like it'll actually be pretty close. I got a nice sticker in return for voting. It depicts an eagle with the creatively-declined motto "e plurebus unum". I decided to be nice and not point this out to the poll workers. :-)


Blogger Mwal said...

Regarding Props, I voted yes on 1A, 59, 60, 61, 63, 64, 66, 67, 69, 72; no on 65, 68, 70, 71; and left 60A blank. (I also voted for the K guy, but my vote doesn't matter much in that election.)

11/02/2004 11:37:00 PM  
Blogger acg said...

The sticker I got just said "I voted" in English and Spanish, with a little flag. At least they spelled it right.

11/02/2004 11:59:00 PM  
Blogger Qian said...

Looks like it's over. Ohio's gone to Bush. Looks like the country has gone pretty solidly conservative now with the Repubs consolidating in the house and senate as well. They'll also be able to reuse a bunch of Bush signs and bumper stickers in four years when Jeb runs. I guess now is the time to take the long view of things and thank God for presidential term limits.

11/03/2004 01:15:00 AM  
Blogger Justin said...

Well, half the networks are holding off on calling Ohio. The Kerry campaign is arguing that there are 250,000 votes that haven't been counted yet.

Unless the rest of the midwest goes for Bush, I'm seeing litigation in our future. There are plenty of things that could be challenged in Ohio.

11/03/2004 01:52:00 AM  
Blogger Justin said...

And Edwards is about to announce that the election won't end tonight... If we wait to count the provisional Ohio ballots, it will be at least 10 days before we can even start counting them.

I'm curious whether the folks that were vocal about the popular vote count in 2000 will be equally vocal in 2004 should Bush maintain his lead in the popular vote while Kerry challenges the Ohio results.

11/03/2004 02:20:00 AM  
Blogger Eric said...

Looking at the by county results in Ohio on CNN, it seems to me that those results are quite unlikely to change. However, I'm not sure if those numbers include absentee ballots. Anyone know? Do we know how about many provisional ballots were cast? Oversees ballots?

11/03/2004 02:25:00 AM  
Blogger Eric said...

To Mwal:
I voted differently than you on some of the fiscal issues. But what were you thinking on 69? As I understand it that would (will) expand a DNA database by taking mandatory samples from people arrested for various crimes, even if they are not necessarily convicted. That last part seems fatal to me.

11/03/2004 02:37:00 AM  
Blogger Eric said...

I meant to say that we voted mostly similarly, and that I think I can see a reasonable case being made for your position on the fiscal issues where we differed. Somehow, in my rush to ask about the DNA database, I forgot to mention that part. :)

11/03/2004 02:40:00 AM  
Blogger Justin said...

Absentee ballots haven't been counted yet. Oversees absentees and provisional ballots won't be counted for 10 days.

Kerry's campain quoted a figure of 250,000 uncounted votes. The Secretary of State, on the other hand, was quoting an estimate of 150,000 provisional ballots.

11/03/2004 02:44:00 AM  
Blogger Eric said...

Regarding stickers... I voted last week as a permanent absentee voter. I think it went fine (not sure how I'd know if it didn't), but I didn't get a sticker. This is a significant problem, as people repeatedly ask "Have you voted?" So I think they should mail me a "I voted" sticker along with something saying that my vote was counted (or a notice saying why it wasn't and what I can do about it).

11/03/2004 02:46:00 AM  
Blogger Eric said...

THanks for the numbers Jocave. Cnn now shows bush ahead by ~139,000 in ohio. According to CNN, the county where Kerry has the biggest lead (Cuyah... with Cleveland) is 67% to 33%. So if we assume that ratio of ballots holds for the absentee/provisional ballots and that there are 250,000 ballots left (taking presumablely optimisitic democratic assumptions), then that would be a net 82500 vote gain for Kerry. Again, assuming 250000 ballots left to be counted, it would require that those ballots be >=78% for Kerry. I'd say that makes Ohio pretty likely to end up going for Bush.

BTW- According to an Ohio newpaper, last year approximately 90% of provisional ballots in Ohio were counted. So that's not likely to be a big factor.

11/03/2004 03:16:00 AM  
Blogger Mwal said...

Eric, props 60A, 69, and 72 were certainly the ones I was most doubtful about.

On 69, I was eventually swayed by the argument that this kind of DNA collection isn't all that different from the other information collected during many arrests (address, mug shot, biometric data, fingerprints, etc).

There is obviously a huge difference between DNA itself and fingerprints. DNA samples have a greater potential for abuse. But that has to be balanced against their power to free the innocent.[*] In my opinion, it's a hard decision to make. I'm still far from certain which is best, but in the end you have to either make your best guess, or else abstain.

Regarding this kind of voting in general: I think California would be better off with a higher threshold for voter initiative. At first I expected that I'd usually vote against almost all propositions because of this. But I've come to realize that, for better or worse, this is part of the government here. It's not ideal, but it's the way that a lot of business gets done, so I might as well participate. (But I still don't like it.)

[*] I don't know how accurate that particular article is, but the use of DNA evidence has certainly been revolutionary in criminal cases.

11/03/2004 03:20:00 AM  
Blogger Justin said...

I don't see a way that Kerry can take Ohio mathematically. They may be looking, though, at challenging the results to look for hanging chads and press claims against the Secretary of State on some voter registration issues as well as voter intimidation issues.

Note that last year, there were only something like 1000 provisional ballots cast, so the 90% trend may not scale well.

11/03/2004 03:36:00 AM  
Blogger Justin said...

It's been interesting to watch different sources take different approaches to calling Ohio depending on their political leanings. Fox News made the call hours before everyone else. The Networks made the call between 1 & 2. The Post and Times haven't made the call yet.

11/03/2004 03:47:00 AM  
Blogger Eric said...

For fun I went back to predictions from August and October.

In August Vincent said: "I'll go on record predicting that Bush will win AR, TN, NC, KY, and CO (196 total). Kerry will take WA, OR, NH, HI, at least 3 of the 4 votes from ME, and NM (195 + 1? total)."

All appear correct, excpet NM may still be in play.

Later, in October, Vincent said: "I stand by my initial map, with a couple of exceptions. Florida will go to Kerry, but New Mexico is a real tossup that has the potential to be the closest state for the second consecutive presidential election. Add the potential for a Kerry victory in formerly red Nevada, and I'll guess that Kerry's final EV count will be 311. While the outcome in at least one state will be within the margin of litigation, the nationwide result will not....

Kerry will win the popular vote and break 50% of votes cast (but just barely)."

FL and NV have been called for Bush. It appears Kerry will not win 50% of votes or the popular vote. Kerry will almost certainly get less than 311.

In August, Eric said: "Whoever wins two of OH, PA, and FL will win."

That appears very good.

I continued: "Ok, you want an prediction not conditioned on a posteriori information..." and went on to describe my theory of states bordering a major body of water with a few exceptions. For the sake of brevity, I'll skip ahead to... "If I had to be even more specific...

Bush: AZ (10), AR (6), CO (9), KY (8), TN (11), NV (5), NC (15), OH (20), MO (11) = 95

Kerry: DE (3), HI (4), IA (7), ME (3+1), MI (17), MN (10), NH (4), NM (5), OR (7), PA (21), WA (11), WV (5), WI (10) = 108"

I was wrong on WV. And we'll have to wait to be sure on OH, IA, WI, and NM.

But the best predictions was... Mwal, who said in October: "My prediction was: "I would bet at 53:47 odds that Kerry will win.""

A well designed statement that can't be refuted!

Mwal went on to say...

"Kerry gets 285 +/- 10 +/- 17 +/- 5 electoral votes. (Bush gets the rest.) Errors are: statistical, systematic, and electoral model, respectively.

Turnout is 120 +/- 9 million votes.

Kerry gets (49.5 +/- 0.7 +/- 0.7) percent of the popular vote."

So far so good. More carefully chosen predictions.

Unfortunately, Mwal made one mistake when he went on to say... "If Kerry wins, Bush will concede before 2004-Nov-03 23:59 UTC.

If Bush wins, then Kerry will concede by the aforementioned time if and only if a swing of more than 10,000 votes in a state would be required to change the result."

In fact, it would take a swing of more than 100,000 votes to change the result, and Kerry still hasn't conceded. Never underestimate politicians!

11/03/2004 03:51:00 AM  
Blogger Eric said...

Mwal... When in doubt and in debt, I generally vote no for increased spending. The three strikes change was the one I found the hardest.

I agree that it's impossible to make really great decisions on all the ballot measures (many of which I beleive are intentionally misleading or include deceptive enticements). I voted on issues where I wasn't sure rather than abstaining.

I agree DNA can be well used to clear the innocent. But I thought that usually that works by the defendant volunteering their DNA for comparison with DNA collected at the crime scene. That doesn't require there to be a database. Of course, proving you're innocent without finding the person who is guilty isn't as satisfactory as catching the perpetrator. I guess I just don't trust government to handle the information properly. I could very easily imagine police doing data mining on a large data base of DNA profiles and finding "matches" that are merely due to chance.

Also, there's concern that the information could get out (e.g. due to a computer break in as may have happened with SS numbers in CA recently) and how that could affect people's ability to get health insurance.

11/03/2004 04:00:00 AM  
Blogger Eric said...


In case you couldn't tell, I agree about Ohio. Kerry's only chance is in the courts.

BTW- That could get quite interesting if Renquist is unable to participate (but presumablely he would try very hard to). It's my understanding that in the case of a supreme court tie 4-4, then the lower court ruling stands. So the presidential election could conceivablely be determined by a lower court. I'm not sure about the chain of appeals on such issues. Do you know?

Also, I've been looking at CNN and listening to NPR, but ocassionally see/hear about other sources calling races. I agree that there's an apparant correlation between news source bias and when they call things for the opposite candidate.

11/03/2004 04:07:00 AM  
Blogger Justin said...

If we reach the SCOTUS, things get really interesting.

On a 4-4 tie, the lower court ruling stands but no precedent is set. However, if Rehnquist is too ill to hear the case, he would presumably be ill enough to retire (there is some evidence that they couldn't get all the cancer, in which case he probably has no more than 6-9 months to live). Were he to retire, Bush would be able to make a recess appointment (no Senate confirmation). We could, therefore, be in a situation where Bush is unilaterally appointing a justice who breaks the tie to declare Bush the winner of the election so that Bush can nominate this new justice to a spot on the court. I hope it doesn't get that far.

11/03/2004 04:19:00 AM  
Blogger Mwal said...

Re: predictions:

2004-Nov-03 23:59 UTC hasn't happened yet, so there's still a chance that the prediction could still turn out to be true. But that would be by accident... I now see that I greatly underestimated the threshold needed for the Kerry campaign to contest a result.

Re: DNA:

Before forming an opinion on the proposition I really should've checked with Julie or someone else who'd know these things. But I don't think forensic DNA analysis records anything particularly useful for predicting a person's health. I'd be more worried about people who have access to the physical sample, which would presumably be permanently stored. I'm not all that concerned about health insurance providers, since they usually require a person to release his or her medical records before issuing a policy, which actually tells a lot more than DNA. But there may be other people who'd be interested in using DNA for some insidious purpose.

What it means for DNA to match is pretty technical and statistical evidence especially may not be evaluated correctly in the courtroom. This is a real problem which must be considered, but to me it is not necessarily a showstopper. (and note that fingerprints have similar problems.)

11/03/2004 04:59:00 AM  
Blogger Vincent said...

I think the two great questions about this election are 1) What happened to all the young voters? and 2) Why were the exit polls so horribly wrong? I think we need to see the dust settle a bit more in Ohio and Florida before we can come up with firm answers to those questions.

11/03/2004 12:36:00 PM  
Blogger Eric said...

Jocave: Right, about the lower court ruling standing. But which lower court would it be? In some cases things go through several courts before reaching scotus, but in other cases things can jump one or more of those levels. So I'm not sure which court it would be. It might depend on what exactly Kerry sued over.

I didn't realize that Bush could make a temporary appointment that didn't require Senate confirmation. That would be a very nasty way to end things.

Mwal: You may be right about them only saving limited info about the DNA. That distinction wasn't clear to me. As for the threat of data mining leading to false convictions, I think that both prosecutors and jurys recognize that DNA has the potential to be much more foolproof than fingerprints. That makes it more powerful and also more damaging if abused. Oh well. Hopefully, none of us will get arrested.

Oh, whoops, I missed the date. It looks like your final prediction of Kerry giving up may in fact come true. Sorry. Either way, your predictions prove the best.

11/03/2004 12:52:00 PM  
Blogger Justin said...

Where did the young voters go?

As a percentage of all voters, there was no increase in young voters, but turnout was up significantly, so young people at least increased their gross turnout, they just weren't more motivated than everyone else. Frankly, I'm suprised they did that well.

And exit polls?

That one really bothers me. Phone polls have to try to call a representative sample of people. They have to figure out who is a likely voter. They have to worry about leading questions. All sources of error. As I've said before, I'm not suprised that the polls are wrong.

Exit polls, on the other hand, should be hard to foul up. You put a guy with a clipboard outside the polling place. You tell him to stop every 5th voter and ask who he voted for. You add up the numbers. Not rocket science, yet the exit polls were systematically 3 or 4 points too generous to Kerry.

11/03/2004 01:44:00 PM  
Blogger Eric said...

There are still potential sources of bias. Women were more likely than men to vote for Kerry. I wouldn't be suprised if they were also more likely to stop and talk to an exit poller. In the other direction, old people were more likely to vote for Bush, and I wouldn't be suprised if they were more likely to stop and talk to an exit poller. You can imagine lots of things like that... income, working vs stay-at-home wives, etc. Still, I would have thought that lots of those would tend to cancel more than it sounds like they did.

There's also the possibility of the exit pollers being biased in who they select to ask. Given how poorly they did, you have to wonder about that.

11/03/2004 02:19:00 PM  
Blogger Justin said...

But it is really hard for me to imagine what demographic changes would cause the exit polls to go goofy in 2004 but not in 2000. Other than Florida, the exit polls were pretty darn accurate in 2000. Barring utter incompetence on the part of the pollers, I have a hard time figuring out why any group would be more or less inclined to answer questions this year than they were 4 years ago.

11/03/2004 03:31:00 PM  
Blogger Eric said...

Well, the group that did the exit polling changed since 2000. The old group had flaws, but also decades of experience. Perhaps the new exit pollers are just learning the ropes.

11/03/2004 03:42:00 PM  

Post a Comment

<< Home