Monday, November 29, 2004
Stevens: If you reduce demand, then you will reduce prices? Wouldn’t it increase prices?
Barnett: No, if you reduce demand, you reduce price.
Stevens: Are you sure?
Saturday, November 13, 2004
Finally, so many questions answered. This is a long article, but it's a fascinating look at a process you never see. When the media frenzy started about the delay and the source theft and all that, I figured we'd someday have this article to fill in all the weird stuff that happened for us.
Beautiful But Wrong: The Floating Hourglass Puzzle
I first found this full book on Rich Burridge's blog. Specifically, pages 130 to 138 are really fun to read. I'm sure I'll find some other good stuff in here too when I get a chance to look around.
Monopolies of the mind
I suppose it's nice to note that no less a publication than The Economist has now bashed US patent policy. I'm sure it's all going to start changing now.
If you're not in the DC area, you probably don't know or care that the mayor of DC is about to spend half a billion taxpayer dollars to get the fabulous Montreal Expos to move to DC and build a stadium. The idea is that this stadium will revitalize Southeast DC with all the economic benefits it will bring. Jacob Sullum points out what is obvious to any moron: that this is a really stupid idea.
Friday, November 12, 2004
This is a really fun piece, although much longer than your average article on the web. Although it's against the grain that characterizes this article, I have to quote my favorite bit:
As always, I don't fully agree with the criticisms (by her construction, it's because I'm too "in the system," but whatever), but it's amusing. Did she really just call engineers "attractive?"
I have not, I realize, painted a very attractive picture of economics. But these sins are widespread, I repeat, among non-economists, too—even that odd one, candid selfishness, which you can find Nature’s Economists articulating even when they aren’t trained in it. But I earnestly invite you to learn by further reading in the literature the offsetting merits of economists:
Economists are for one thing serious about the public interest, and are often the only people defending it with any sort of lucidity and persuasiveness against the special interests. The model of worldly philosophy was originated in crude form by the early pamphleteers and political arithmeticians (among them Daniel Defoe). Adam Smith a half century and more later brought it to perfection.
And if you like engineers you will like many economists. Engineers are attractive people, hard working (you have to be hard working to absorb all that engineering math), earnest and practical, bent always on Solving the Problem. True, they are often simpleminded. But simplicity gets the job done. Lots of economists are engineering types.
Or lawyer types. Like lawyers the economists are good arguers, which is good when you need a good argument ("How do you want it to come out?"). Economists can debate each other and yet not lose their tempers and not make irrelevant appeals to rank. Economists like lawyers are clear-minded, professionally. They are used to
getting to the point and staying there. The humor of economists, unhappily, is often cynical, as it is also among lawyers, seldom generous, but that’s true in many fields of the intellect.
But, above all, economics is about important matters. It would be remarkable if the economics-since-Marx that most non-economists would rather not read had nothing worthwhile in it. After all, thousands of apparently intelligent (they certainly think so) economists have labored away at it now for a century and a half.
I beseech you, dear reader, think it possible that economists, even Chicago-School economists, even Samuelsonian economists, have some important things to say about the economy.
Why Does the Government Patronize Us?
Recently crowned Nobel laureate Ed Prescott argues persuasively for mandatory private individual accounts for social security, with a minimal "free money" system for those who for one reason or another weren't able to save enough for their old age. My favorite bits:
It would be one thing if the government's Social Security system paid a decent return, but as the President's Commission reported, for a single male worker born in 2000 with average earnings, the real annual return on his currently-scheduled contributions to Social Security will be just 0.86%. And for a worker who earns the maximum amount taxed (then $80,400), the real annual return is a negative 0.72%. A bank would have to offer a pretty fancy toaster to get depositors at those rates of return.
Further, about two dozen countries have reformed their state-run retirement programs, including Chile, Sweden, Australia, Peru, the U.K., Kazakhstan, China, Croatia and Poland. If citizens in these countries can handle individual savings accounts, especially citizens in countries without a history of financial freedom, then U.S. citizens should be equally adept. At a time when the rest of the world is dropping the vestiges of state control, the United States should be leading the way and not lagging behind.
Under a reformed system there will always be some individuals who, owing to disabilities or other reasons that prevent them from working, will not have sufficient savings in their old age. The solution is to include a means-tested supplement to ensure that those citizens receive a required payment -- just like they receive today. Nobody gets left behind under this new system, and most will move ahead. U.S. citizens deserve more than a minimum payment, and the U.S. economy deserves more than to have its savings, capital and labor weighed down by an increasingly costly tax-and-transfer system.
Jeers and Cheers for Sears
A creative approach to making money from troubled retail companies.
I don't understand why game developers don't build in ways to counteract this. You need to balance the small number of times when someone drops because their connection actually broke or they crashed (the former happened to me the other night), and the much more frequent "I'm losing" drop. One thing you could do is count every drop after, say, 5 minutes as a loss. You could even allow one or two unpenalized drops per month. Alternatively, you could set a "drop to game ratio" and if someone has dropped more than X% of their games, they are stigmatized with a "dropper" icon. I'm much more in favor of the former than the latter; the latter makes it hard for someone to get back into the game's good graces, because they won't be able to play games to work away the ratio.
Another thing that would help, I suspect, is having a way to "resign" a game, so that the other players can get a win on their record, without having to wait for their units to pound all your buildings into dust. I think some people drop when they're losing because it's boring to sit there helpless while your base is destroyed (it takes a couple of minutes), and they don't realize that by dropping they're depriving the other team of a win.
The first game I played was a 2 on 2 game. Our opponents got together and rushed me early. Since one of the players was playing as Orks, I got my hands on a heavy bolter the only way I could without an armory: by upgrading the nearest listening post (usually this is seen as a waste of requisition). That started beating back the orcs, but it was so early in the game that I still had almost no units to fight back with. All of a sudden, a pink eldar army mosies over from nowhere and starts beating them back.
"Thanks," I said to my teammate. "Hey, we're a team right?" he replied. Incredible. I've never had a teammate come to my rescue, even when I'm asking for help. He came by unasked. I told him that he was the first non-bozo teammate I'd ever been paired with.
After the great mass of their forces were broken, we established clear control of the center. Since my teammate and I controlled the center of the map, and had most of the strategic points, we were going to win by a control-area victory, so we just sat there and waited (not by coordination, we both just knew that we could sit and wait). That was going to be a boring 6 minutes, though, so I built a tactical squad consisting entirely of rocket launchers, and used some undercover scouts to look around the nearest guy's territory. He had some listening posts, so I decided to attack his economy and had my rocket squad destroy those in no time. However many units this guy had, he wasn't going to be building very many more, so me and my teammate just went in for total annihilation (he dropped, of course, but it was still fun).
The other game was a bit later. I came into a 1 on 1 game with some guy playing as orks on a map I was unfamiliar with. This map has basically two valleys going to a river that divides the territories, so I divided my troops and had them stand at each of the two guarding my territory. Since I knew I was playing against orks, I got heavy bolters into as many hands as possible as soon as possible.
Soon enough, he brought some masses over to one of the valleys I was guarding. He must have been surprised when half my army totally destroyed his first wave (a pathetic number of marines compared to his ork masses...it was the heavy bolters). Unfortunately, his next wave was much harder. He started building wartraks, and they started pounding my troops. Since they all had heavy bolters, I couldn't get any rocket launchers over to destroy the wartraks soon enough, and had to fight them off by moving the rest of my army from the other valley. Obviously, this was costly.
While this was all going on, I had an undercover scout squad spying in his territory: one of his valleys was totally empty, and led right to his base. When I saw this, I remembered a game I had played the other night. It was a 3 on 3 game, and one of my teammates put out a call for help, but I was far away. I told him I was on my way to help, but he said, "No, just attack his base. It'll pull him back." It worked.
So when this guy attacked me again, and I didn't have the firepower to beat him off, instead of moving some marine squads from the other side of the map, I rushed them to his base and attacked. They got chewed up pretty badly, but it did pull him back from my other troops.
As fast as I could, I started building marine squads and then a machine cult, so I could bring some dreadnoughts over. Cleaning up the intrusion onto my territory took some time, but I finally cleaned him out and started building up forces. I got all the tech I could as fast as I could, because I had the requisition to spare and knew that his next wave was going to be killer, and his base was very near the point of contention, so he could get his troops to battle quickly. As part of this, I built several predators and terminator squads.
His next wave was huge and filled with vehicles, while most of my army was still stuck with heavy bolters (which don't do much to vehicles). However, his other valley was still empty. I ran two predators, fully decked out with four lascannons each to the edge of his base. I also brought my force commander over and ran straight into his base, and he called in orbital bombardment on his base. The few buildings that survived were being destroyed by the predators' lascannons, and he congratulated me on my victory.
Monday, November 08, 2004
I saw this essay on Spiked-Online linked to from Marginal Revolution. Every word is a joy to read.
Getting Over It
Interesting article by Malcolm Gladwell about human resilience.
People are bad at forecasting their emotions—at appreciating how well, under most circumstances, they will recover. Not long ago, for instance, Gilbert, Wilson, and two other researchers—Carey Morewedge and Jane Risen—asked passengers at a subway station in Cambridge, Massachusetts, how much regret they thought they would feel if they arrived on the platform just as a train was pulling away. Then they approached passengers who really had arrived just as their train was leaving, and asked them how they felt. They found that the predictions of how bad it would feel to have just barely missed a train were on average greater than reports of how it actually felt to watch the train pull away. We suffer from what Wilson and Gilbert call an impact bias: we always assume that our emotional states will last much longer than they do. We forget that other experiences will compete for our attention and emotions. We forget that our psychological immune system will kick in and take away the sting of adversity. “When I talk about our research, I say to people, ‘I’m not telling you that bad things don’t hurt,’” Gilbert says. “Of course they do. It would be perverse to say that having a child or a spouse die is not a big deal. All I’m
saying is that the reality doesn’t meet the expectation.”
Medicare Covers New Treatments With a Catch
Medicare is going to start covering new and expensive treatments in exchange for getting companies to pay for studies on which methods actually work. Hopefully this will provide hard data about the effectiveness of new treatments, which can eventually be used to stop paying for the ones that don't work.
The new initiatives began this year when Dr. McClellan, who is both an internist and an economist, decided that evaluating treatments for elderly Americans falls under Medicare's purview to pay for whatever is medically necessary. If the treatment does not work, he argues, then it really is not medically necessary, and Medicare should not be paying.
Over the past six months, he has been fundamentally transforming Medicare's scope by offering to pay for a number of new and expensive treatments and diagnostic tests - ranging from $30,000-a-dose cancer drugs to brain scans to diagnose Alzheimer's - but with a catch. Patients can only have them if they enter into studies that evaluate how well they work. And someone other than Medicare will have to pay for those studies.
In the past Medicare just paid the bills or said no to treatments. Now, with a flood of promising but astronomically expensive treatments and little data on how they work in the real world, Medicare has decided to use its 41 million beneficiaries to get some answers. And it is using the threat of refusing to pay unless patients are in a study as a cudgel to get companies or foundations or professional groups to pay for the research. The goal, Dr. McClellan said, is better and more cost-effective medicine.
And here we tie together the two previous articles: An article about drug prices by Malcolm Gladwell.
First of all, the short that opens this movie, Boundin', is unquestionably the best short Pixar has ever made. It's fun, has cute music, is told in rhyme, and is just plain adorable. Everyone's heart melted. I almost want the DVD just for Boundin'.
Now, the rest of the movie is so great, it's hardly worth talking about all the different aspects that combine to make it fantastic. Let me just describe what it was like, seeing the movie on opening night, and I think you'll get an idea of how great this movie is: I haven't seen an audience in such rapture since I saw The Matrix on opening night (in fact, the movie actually has some of the most dynamic, exciting action sequences I've seen since The Matrix). The laughing was so loud and continuous that I lost track of how many times I couldn't hear the next line. All around me people were leaning forward, propping their elbows on their knees and their heads in their hands, smiling at the screen.
One of the things I've always loved about Pixar movies is how clever their characters are. Instead of having an answer for every situation, they are pressed to be clever and resourceful. This holds true even of the superheroes in this movie. Instead of feeling boring and overpowered, they always seem pressed to the limits, and in danger of failing. They need their intelligence to win. Truly geek superheroes.
The movie also has a kind of retro-future feel that I really enjoyed. Pixar has for the first time moved humans into the leading roles, but they've done it their way. Instead of going the Polar Express route, which attempts extreme realism, only to wind up with jerky, highly disturbing copies of real people, The Incredibles uses highly stylized, cartoony-looking humans. Plus, the art direction is fantastic, filling the movie with some great visuals. It manages to spoof James Bond, Star Wars, and other superhero films while at the same time outdoing them. One of my favorite shots is Mr. Incredible standing in silhouette in front of a gigantic wall of pouring lava in the supervillain's lair.
Finally, we can officially say that Brad Bird is one of the best directors working in Hollywood today. He's only made two movies, The Iron Giant and The Incredibles, but they're both spectacularly good films that can do the unthinkable: hit every note, and appeal to every audience while not feeling forced or over-stuffed at all.
Friday, November 05, 2004
I went into this movie not having seen a trailer or even a still. In fact, I didn't really know much about it except for the one sentence plot summary. Frustratingly enough, the movie didn't have opening credits, so I couldn't tell who the director was, and thus whether I was supposed to like the movie or not. Fortunately, the closing credits told me that the director was Jonathan Glazer, of Sexy Beast, and I felt relieved that I hadn't just fallen in love with a Joel Schumacher film.
I was enchanted almost as soon as the movie started. The opening scene, long slow shots moving along a snowy Central Park, combine perfectly with the music to give the movie a sense of magic that lasted throughout the movie for me. I came out of the movie just assuming that everyone thought it was a great film, and feeling warm from the rare experience of a wonderfully executed film.
Ten years after her husband Sean's death, Nicole Kidman is finally remarrying. But a ten year old boy named Sean shows up, and tells her that he's her dead husband, reincarnated, and he doesn't want her to remarry. Obviously, the fact that this possibility is not easily dismissed throws her into an emotional turmoil, and that is the core of the movie.
As I said, the movie is perfectly scored, and matches well with Glazer's visual style. It really feels enchanted, and at times creepy, and it is powerful. Nicole Kidman puts in a good performance, and is supported well by the intense Cameron Bright (from Godsend), who plays the reincarnated boy. The boy is serious beyond his years, and makes the whole ridiculous premise work.
There's a much discussed bath scene, which is often summarized as "Nicole Kidman taking a bath with a 10 year old boy," suggesting a scene of pedophilic sensuality. It's pretty ridiculous, that's not what happens at all. Nicole Kidman is curled up in a bath, he lets himself in, takes off his clothes, and gets in the tub, curled up in a ball at the other end. They don't touch, and after a moment she tells him that she wants him to leave her alone. Which of course isn't appropriate behavior on Nicole Kidman's part ("Uh, get out of the bathroom you little perv..."), but it's certainly not smut either. Maybe she figured the kid had seen it all before.
The other thing that might strike you is that...Nicole Kidman looks kinda weird. Combined with her ultra-short dark brown haircut, and pulled-up eyebrows, she looks like a pixie from Peter Pan or something.
Finally, you might like or dislike the ending, but I thought it was perfect. Spoilers ahead, so select the text to read it. I always hate it when people claim that they "saw the ending coming a mile away," but I really had already started to theorize along very similar lines. That isn't a complaint, I just want to say that I think the ending turned out to be exactly what the movie needed to tie things up correctly, and I was hoping for my theory to be correct. It's crushing, but it feels right; seeing what Nicole Kidman goes through, you don't really want it to be her reincarnated husband, and have them stuck in an impossible love. Kids can do horrible things to people, especially without intending to. The boy doesn't seem malicious, he seems genuinely confused, and that makes it even more painful.
Thursday, November 04, 2004
In this paper, we present an information-based theory of strategic extremism. The key assumption in this theory is that awareness of a politician’s policies is higher among the politician’s supporters than among his opponent’s supporters. This asymmetry means that when a politician’s policies deviate from the median, he energizes his own supporters (who are more likely to be aware of this deviation)more than he energizes his opponent’s supporters (who are less likely to be aware of this deviation). That politicians have an edge in communicating with their own supporters is reflected in recent efforts by the Bush and Kerry campaigns. For example, during the Republican National Convention there was a "closed, invitation-only Bush campaign rally for Christian conservatives" at which "Senator Sam Brownback of Kansas called for a broad social conservative agenda notably different from the televised presentations at the Republican convention" (Kirkpatrick, 2004b). The Kerry campaign also appears to be targeting public appearances to reach loyal Democrats (Borsuk, 2004).
The paper also mentions another interesting result, which is that you should be able to tell whether certain political views are strategic or heartfelt. If an extreme viewpoint is strategic, then policy will be less extreme than the message, while the opposite is true for a heartfelt viewpoint:
A key difference between a model of strategic extremism and a model in which extremism reflects politicians’ preferences is that, when extremism is strategic, politicians’ policies will be more moderate than their messages. When extremism reflects leaders’ preferences, policies will be more extreme than political messages. We examine policies and platforms on tax policy and abortion over the last 25 years to examine these implications. The economic messages in platforms are extremely moderate, but there are big differences in mean tax rates between Democratic and Republican regimes. Conversely, political messages about abortion tend to extremes, but abortion rates are independent of the party in power. These results suggest that differences in economic policies between the parties reflect preferences of party leaders but that differences in abortion and other religion-related policies reflect political strategy.
Their paper also provides some numbers for the rise of religiousness as a political issue, and the decline of labor unions:
We end the paper by examining a key prediction of the model: religious determinants of political orientation will be maximized when about 50 percent of the population attends church regularly and that economic determinants of political orientation will be maximized when about 50 percent of the population is in labor unions.
Finally, we use our parameter estimates to see whether declining church attendance and declining union membership can explain the rising importance of religion in American politics. Unions have declined from being close to 35 percent of non-agricultural employment in the 1950s to being less than 15 percent of the labor force today (Freeman, 1998; Hirsch, MacPherson and Vroman, 2001). Church attendance has declined from 1968 when 57 percent of the population went church once per month or more to 2000 the point when 47 percent of the population went to church this often. Both of these declines should have caused politics to focus less on economics and more on religion.
I find the model of this paper very clever and useful (keep in mind I still haven't read very far in), and it captures the intuition in the median voter theorem in more detail. It still seems to suggest that a change in focus towards religion and morality is needed for the Democrats; either that, or a new group to target that maximizes the probability that members of that group will vote Democratic. If we knew what that group was, though, we'd have probably found it by now.
First of all, their job basically echos the job of the president but on a smaller scale. People might have valid reasons to be turned off from someone who has had a long history of political sausage-making in the senate. Governors, like the President, can simply accept or reject the bills that come out of their legisatures, letting them keep their hands clean and still take credit for the good things that happen under their watch. While governors and senators might both pander to their constituents, that behavior might be looked down upon more in a senator, who does such things with federal issues. Voters might be more forgiving of such vigorous representation at a state level, and see it as a desirable attribute.
Lastly, I think comparing the resumes of presidential candidates is really dumb. No one ever has enough experience to be the president, unless he already was president. I don't see why a governor of a given state's experience should be any less relevant than a senator from that same state, since they require the approval of the same set of voters. But I don't think people really care about their leaders' resumes anyways, they care about what their vision is.
However, if current trends continue, the Democrats might have slim pickings for capable presidential candidates. There was no net change in Democratic or Republican governorships this election, but if the Republicans pick up some more, the chances of finding someone in that group with national appeal get worse and worse.
I absolutely agree with Slate's rationale for posting exit poll numbers. However, once you get to that point, you don't have statisticians interpreting the numbers, but lay people.
As it happens, Kerry did carry Pennsylvania. In Florida he lost by two points instead of winning by two points. Far from being wildly off the mark, that variance is about par for the course, or even under par, for a mid-afternoon reading of an exit poll. Indeed, Tuesday's exit-poll numbers were no more off the mark than were those of four, eight, or 12 years ago.
That's why the people who bought and paid for this intelligence kept it to themselves. Bill Wheatley, an NBC executive who understands these things, told the New York Times Wednesday that early afternoon numbers from an exit poll are "junk." Slate and the other Web sites on which these numbers appeared yesterday afternoon have every right in the world to get them however they can and publish them. But it's hard to pin the blame for the dissemination of these numbers on those who tried so hard to keep them secret.
Mind Poison, your Slate digest.
Wednesday, November 03, 2004
Two articles I really liked on this topic are this one, by William Saletan, and this other one, by Timothy Noah. Chris Suellentrop also has worthwhile thoughts.
Timothy Noah points out that the democrats have problems on all sides. He says that they can't move rightward, can't move leftward, and can't sit still (but you'll have to read his article to get the reasons). Sitting still is not a good idea, and it's motivated only by wishful thinking. I get the feeling that there was some of that this time around: Al Gore really won 2000, they think, so we just have to get our rightful result this time. It didn't work out that way. The world has changed.
Similarly, moving leftward won't work. Despite the great hopes for the youth voter turnout this year, I always held private doubts about whether that would materialize. If youth voters can't get up the self-interest and discipline to stop their grandparents from screwing them with debt, I can't imagine what would. Having lived on or very near a college campus for most of the past 6 years, I've seen the attempts of the highly motivated student organizations to do something, anything. Even with local politics that seriously affected Isla Vistans' ability to party, nothing happened. Riots in Seattle, Eminem videos, war that many think threatens a new draft (an exceedingly unlikely development, but nonetheless one which many young voters honestly believe is possible)...nothing seems to turn these voters out.
No, despite turnout attempts, treating turnout as exogenous in this situation seems to always be a more reasonable assumption to work with. For most of our lives, there have been those that vote as a bloc, and those that don't, and realistically, you have to work within that. This leads me to my criticism of Noah's piece.
Noah says that moving to the right is not an option, because moving to the right has no endpoint, and only pushes the Republicans further rightward beyond reason. I believe this is wrong. The median voter theorem states that politicians will more or less move to the position of the median voter. Any voter to the left of the most left politician still votes for him, and any voter to the right of the most right politician still votes for the rightmost polician. Thus, there is no gain to moving away from the center, because it allows the other guy to hug your position but stay just a little to the other end of you, grabbing more of the votes. Obviously, this is a major simplification, but I think this construction captures the basic observed features of politics remarkably well (ie, third parties never win because they're too far to the sides and only catch that tail of the distribution that is at least as extreme as they are, while major parties both resemble each other in the center and tend to win only by luck in close elections, and so forth).
If the Republicans are winning elections by moving to the right, it is because the distribution of voters is changing so that the median voter falls farther to the right. In this case, the best response actually is to move farther to the right, but not quite as far as the other party. This is where Saletan's article comes in. Saletan makes some good suggestions that I agree with, and I think his proposal for a grand vision makes sense, and I think the reason has to do with what's causing the shift away from Democrats.
The political spectrum is not a single line across which people fall, it has a more complex structure than that, especially with the electoral college. The parties sustain themselves by drawing together strong, disciplined voter blocs (this leads to such strange alliances as highly religious social conservatives and socially liberal libertarians voting for Republicans, and rich, highly educated coastal whites voting with middle class labor in the midwest for Democrats). For Democrats, their base groups are being rearranged, as Republicans learn to appeal to minorities better (it was only a matter of time before Republicans figured out that minorities, which are often very religious and entrepreneurial, would make great Republicans), and the long-term US decline in manufacturing weakens the labor unions. Painful as it may be for the party faithful, in order to get back in the game, Democrats have to move on from their current positioning, and find some new alliances, either by finding new voting interest groups, or taking some from Republicans. Saletan's proposal is about as plausible an idea as I've seen for that so far, but it's gonna take more than that.
In immediate, practical terms, Saletan suggests Edwards for 2008. Though I harshed on him earlier, I don't think that's implausible. What he's going to have to do is spend the next four years learning a lot about policy so that he doesn't have to dodge easy questions that he should know the answers to. And he needs to get better at debating, so he doesn't freak out so easily. It can be done.
Superficially, he's good: young, good-looking, doesn't have a long, embarassing senate voting record, is Southern and comes from the middle class. However, we'll have to see if he can overcome the stigma of failure and if he can overcome the intellectual, leftist urges of the party faithful in the primaries in 2008 (not to mention another potential candidate with more red state appeal). The game in 2008 is entirely about turning one or more red states blue. I haven't seen enough of Obama in action to form an opinion of him, but I'll have to be convinced that he could turn a southern or midwest state to the Democrats. Similarly, I have my doubts that Hillary Clinton could do it either. This is that talking with sincerity about moral values thing that Saletan talks about; I agree that it's essential, but she doesn't have it.
In fact, let's just stay away from senators in blue states altogether next time.
Update: Just wanted to add this link to another interesting article, A Functional Party No More, by Reason's Tim Cavanaugh.
Robert Wright also has some thoughts on Slate, Why Americans Hate Democrats.
Musil breaks down some of the minority voting numbers.
I walked home from work just now and noticed that they'd cut the tree down today.
I mentioned this before, but anyone who understands statistics a little bit should understand why this is nonsense (you'd think a political advisor would understand that much about statistics). I linked yesterday to MysteryPollster's explanation of exit polling, but if you didn't read it, you should. They try to sample 100 people at about 1500 precincts across the country throughout the day. In a tight race, with sample sizes that small, it would not be unlikely that you see results that suggest the opposite of what the actual distribution is.
If you've got say 5 million people voting, and their underlying distribution is that 51 percent support Bush, when you sample 100 of them, there's no reason you shouldn't expect to see that 51 percent of your sample support Kerry. Taking this sample, your maximum likelihood estimator would lead you to believe that 51 percent of the 5 million people support Kerry.
Now, they don't forecast a state with 100 voters. If the precincts are distributed across the states equally (probably not), then say there's 30 per state, or 3000 people. That's a much better-looking sample size. But keep in mind that these are not ideal sampling conditions; MysteryPollster brought up several problems that make exit polls not statistically ideal. Furthermore, that 3000 people isn't 3000 randomly selected, it's 3000 from 30 randomly selected precincts. Some of the precincts will be heavily democratic and some heavily republican, and some neither. By the luck of the draw, your 30 random precinct draws in that state will probably sample one or the other too much (keep in mind that their "corrective" weighting is based on elections in years past, and voter attitudes and demographics, could have changed greatly in that time, to say nothing of redistricting). Voting patterns might also be different throughout the day, and also, reading Drudge and seeing a close race that your guy is losing might get your ass out to the polls at the last second.
In the end, just about every early exit poll number I can find was within the margin of error (supposedly 4%), and well within it at that. Kerry supposedly had 51% in Ohio and he had 49%, supposedly 52% in Michigan and he had 51%, 53% in Pennsylvania when he only had 51%, 50% in Iowa when he only had 49%, 51% in Wisconsin (which was correct), 52% in Minnesota when he had 51%, and so forth. The farthest from the exit polls were New Hampshire and Florida, and I imagine if you go back through the polling data in previous elections, there are plenty of exit polls in very close states that were not correct within the margin of error. Only it wasn't a big deal because they were in states that didn't matter, or they predicted the same as the actual result anyways.
So the mid-day exit poll numbers showed a small lead in the wrong direction in the closest races. Calm down, Dick.
Most of you probably know that I really hate flying. It's a totally irrational fear, since I am quite well-informed about how safe it is. For some reason, I found this article strangely reassuring.
New York City Picks New Streetlight Design
I saw this on Fark. New York City just had a competition to design a new streetlamp for the city's inventory, with the eventual goal of replacing most of the city's lights. I actually really like the winning design. It looks cool, and I like the way they put the grooves up the stem of the lamp so that signs and signals can be attached without having to wrap those ugly metal bands around the whole post. Most interesting to me was the fact that they're designed to be powered by LEDs. Given the power usage of conventional streetlamps, and the fact that they need to be on at all times during the night, the power and cost-savings of converting those to LEDs could be substantial. UCSB's Materials department probably has at least one more Nobel prize headed its way some day for this invention.
The Idea Trap
This article by Bryan Caplan argues that there are two equilibria for policy: good and bad. Good policies reinforce themselves by creating growth and thus lead to reasoned debate and more good policies, while bad policies reinforce themselves by creating misery and desperation, which leads people to open themselves to nonsensical policies. There's no rigor here, and I don't think I buy his arguments, but there does seem to be at least a kernel of truth in what he's saying from a psychological standpoint...I'm sure we can all think of similar examples from history or life.
I think this paper is ultimately too simplistic, though; there are many countries that have bettered their lot by simply realizing that their policies were bad and copying those of countries with better outcomes. He says that that's due to luck, and in a sense it is (so are all good things), but it doesn't give credit to people who make changes through rational thought. Or to put it in his framework, if things get bad enough, people who dogmatically believe in certain bad policies might open themselves up to "crazy" policies like those that seem to be working elsewhere, which undermines his point. Also, I reject his simplistic notion of good and bad policies. Different people in different places would consider different things to be "good" policy.
In general, attempts to categorically summarize the situation of diverse sick and developing economies always fails. There are too many variables to boil it down to a simple framework like this, and even Jeffrey Sachs, for all his successes, has had his failures. So always be suspicious of such simplicity.
Of course, I don't blame the Kerry camp for not doing anything rash. You never know, and this is very, very close. It's perfectly valid to want to be sure before making an announcement.
But let's not kid ourselves, folks. Even assuming that somehow, those 250,000 uncounted Ohio votes actually broke 2:1 for Kerry (which is what it would take), any road to Kerry victory leads through the courts, and we know where that road ends.
Tuesday, November 02, 2004
Here's Drudge's update on the early exit polls:
17:12:21 ET // UPDATE: Exit poll mania spread through media and campaign circles Tuesday evening after exit data from big media sources claimed Kerry competitive in key states.... FL Kerry +1 PA Kerry+2-4 OH Kerry+1 WI Kerry+4 MI Kerry+2 NH Kerry +4"Competitive?" He's winning.
Of course, don't take that to mean that I think the exit polls mean Kerry is winning. If you've got a narrowly divided election (say, fractions of a percent), then your (very sparse) exit polls are quite likely to show a lead for one candidate or the other just due to sampling error. But if there's something that those polls he cites show, it's that Kerry is winning. Certainly it is better for your guy to be up 1% in these polls than down.
Why did I think Edwards was the best? Well, I think if you've seen him give a stump speech, you'd know why. He's eloquent, and speaks powerfully. He looks young and energetic up there. It's exactly what you liked about Bill Clinton.
However, I never watched any of the Democratic primary debates. I'm sure what I would have seen if I had was that John Edwards is an awful debater. He's terrible at it, he falls apart completely. He looks nervous, and behaves strangely. Watching him debate Cheney wasn't a pretty sight. If John Kerry wins tonight, it will almost certainly have been due in large part to his strong and steady debate performances against Bush. Edwards debating Bush would have been a whole different game than a faceoff with Cheney, but I still can't imagine Edwards coming out on top of that one.
In fact, given his poor performance, I am led to agree that picking Edwards hasn't done much for Kerry. The Democratic governor of Iowa, Tom Vilsack, made the short list of potential VP choices, and probably could have delivered Iowa for Kerry. Edwards doesn't have a hope of delivering a single Southern state, so now Kerry-Edwards has had to fight for a state that might have simply been theirs.
The only other realistic Democratic nominee was Howard Dean, and while I was sympathetic to many of his positions (especially his straight-shooting position that we need to raise the retirement age for social security), he struck me as a bit too unpredictable and politically tone deaf (ie, he was laughable when talking about religion) to be a good choice to face Bush. Plus, the Deaniacs alienated the hell out of me with their whole cult thing, and I'm supposed to be on their side.
So in the end, I think John Kerry was the best choice the Democrats had to choose from. And what a sorry statement that is.
Monday, November 01, 2004
That is so...weird.