Thursday, July 31, 2008

Why aren't there more knuckleballers?

Economic theory posits that in the long-run no company can earn a profit in a competitive environment, because somebody else will come in and copy them. But Tim Wakefield has been a fairly successful pitcher for the Red Sox as a knuckeballer for the past 6 years, eating innings and more importantly earning a major league salary at the age of 41. He has beaten or tied the league average in adjusted ERA for the last eight years. But, there aren't many knuckleballers: besides Wakefield, only R. A. Dickey (who plays sparingly for Texas) appears to be currently pitching with this style. If there are any parents out their who want their kids to play in the pros, I have the following advice: teach them to kick footballs, help them to grow an afro, or tell them to throw a knuckleball.

Wednesday, July 30, 2008

Why is baseball's All-Star Game so non-random?

I've often wondered why the American League seems to win every year, and after this year (when they won again) Columbia's stats blog tallied up the numbers. Since 1965 to the present, the results have been:

NNNNNNNNNNNNNNNNNNANNANAAAAAANNNAAAAATAAAAAA

With "N" signifying a National league victory, "A" for the American league, and "T" the one year that Bud Selig famously called a tie. As Phil notes,

Predicting next year's winner to be the same as this year's winner would have correctly predicted 80% of the games in [this time period] ... and that's if we pretend the National League won the tie game in 2002. (If we pretend the American League won it, it's 84%).

Especially given how focal these games are, it's very confusing, because saying that one league is better doesn't explain away the trend. How do you explain away the lack of variance?

Monday, July 28, 2008

Underestimating variance in foreign countries

Interesting observation from Austin Tyler out of China: Shanghai is much more Western than Beijing. The difference is apparently "night and day."

This is counter-intuitive to me, because my notion of China is fairly homogeneous, but it's pretty clear now that I'm biased. This seems like a general phenomenon: we underestimate the variance of things that we don't know much about. So we think that people are either always angry or always happy or always whatever, when it is much more likely that there is more variance and less homogeneity at play.

Book ratings

As I have often noted to anyone who will listen, there is no "imdb of books." There are reasons for this, beyond the silly notion that nobody has thought of the idea. I can think of three explanations:

1) People don't read as many books as they watch movies. This makes sense, since it is both more time consuming and taxing than watching a movie.

2) There is a greater amount of diversity in books than movies. It is much (much) easier to publish a book than it is to finance and shoot a movie, which means that there is a larger set of books that you need to account for.

3) People who do read books will be less likely to rate them, because they are predominantly older and less technologically literate. Your great-aunt might be a voracious reader but not have the computer skills to let the world know what she thought of each book.

These are three pretty big obstacles, and each of them certainly get in the way of creating the "imdb of books." However, it is because of point #1 that such a site would be so popular. Reading a book is a large investment, both time-wise and money wise much larger than a movie. If there were some way to create a list anywhere close to the imdb's top 250, drawing from a large set of independent opinions, it would be highly successful.

Addendum: Ben Casnocha seems think that TV and presumably movies have less variance in opinions than movies. I disagree, I think it's more of a reluctance to invest (a buyer-side issue), based on a similar argument of Tyler Cowen's here.

Sunday, July 27, 2008

Rating systems are underrated

I write a lot about how movies are rated on imdb, and I'm going to continue to because I think the way the system works is fascinating. But before I do so, I have just one bone to pick.

You might not think that rating systems are a very big deal, but they're all around you. Amazon and The New York Times have book reviews. Yelp and your local paper do restaurant reviews. The US News and World Report does a very important and influential review on colleges. IGN has extremely influential video game reviews that I use to base my video game purchases on. You could even view democratic elections as one long, lumpy, and fairly inefficient rating system.

And now that the internet is able to bring millions of people with independent opinions into one place, they're going to matter more and more.

Saturday, July 26, 2008

The unintended consequences of political opinions

My friend Lincoln just got back from a trip to Benin, a small country in Western Africa. He didn't speak any French, so when he couldn't communicate with the locals he would just yell, "OBAMA!", and everyone would laugh and be merry.

Which brings us to an obvious point: foreigners love Obama. In his speech in Berlin he was already urging European countries to send troops to Afghanistan. It's like, wait, aren't you still just a presidential hopeful? But the 200,000 plus Germans lapped his rhetoric up.

People love to use these anecdotes as support for the claim that Obama will be able to make foreign policy decisions that other countries will not. That may be true, I don't know.

Instead, I want to make the point people should be a little more cognizant of the actual consequences of their support for various candidates. Nas, it might not be helpful for you to go off and shout at Bill O'Reilly. Moveon.org, have you ever considered the possibility that your outlandish commercials might actually anger swing voters?

I want to pose a thought experiment. Which event would be more helpful to the Obama campaign: a student walking around a liberal arts college wearing an Obama t-shirt, or a student walking around a liberal arts college wearing a McCain t-shirt?

I think it's the McCain t-shirt, and my reasoning is simple. If you wore an Obama t-shirt it wouldn't be a surprise. Fairly or unfairly, you'll be labeled just another liberal at a liberal school. In all likelihood, most of the students there will be slightly bored of Obama.

But if you were to wear a McCain t-shirt, then people might pay attention. Those who sort of supported Obama but had no intention of voting might realize that there is a whole other constituency out there. If you don't believe that this sort of image-cognition process is possible, you should read up on priming.

I'm frankly undecided on what my presidential vote will be, but it sort of annoys me when liberals think that screaming their talking points is actually going to alter opinions.

Friday, July 25, 2008

The Top 5 "last words" of all time

5) "To the strongest!" - Alexander the Great, in response to his generals wondering which of them would gain control of the empire.

4) "I know you are here to kill me. Shoot, coward, you are only going to kill a man." - Che Guevera.

3) "Drink to me!" - Pablo Picasso. A reasonable request.

2) "Tell the governor he just lost my vote." - Christopher Emmett, today.

1) "Don't let it end like this. Tell them I said something." - Pancho Villa.

I was planning on doing 10 of these, but I read through Wikiquote's list, and there truly are not many funny ones. Maybe people thought they had something funny to say, but wanted to wait until the absolute last moment to say it, and waited too long. The other possibility is that people predominantly don't care for humor in their last moments, which is a depressing thought indeed.

Thursday, July 24, 2008

Angry voters on imdb

Nick thinks that imdb is flawed because so many people give 1's to movies. He views this as evidence that the votes are subjective, and therefore that the system is ruined. In his mind, people are gaming the system out of their personal beliefs, instead of voting based purely on the quality of the movie, and it "completely defeats the point."

I agree (partially) with his premise, but disagree with his conclusion. First, you need to look at why voters would game the system. Is it because they are involved in the movie business and are trying to make a profit based imdb scores? This is highly unlikely, given the large mass of votes and the relative dearth of technologically literate actors.

The next possibility is that they vote movies down because they disliked the movie, or disagreed with what it stood for. So they will give a movie that really isn't all that bad a much lower score than it "objectively" deserves. In essence, they are punishing the movie too much for some small flaws. To me, this scenario is much more likely.

Now, if you accept the second scenario, then we begin to witness the beauty of imdb. Even though these individual voters may be subjective, they are subjective in varying ways, which means that their opinions are independent of one another.

Let's say that you watch Pulp Fiction and you hate the scene with the gimp, because you think it's distasteful. So you give the movie a 1, even though somewhere in your head you recognize that the movie probably deserves a higher score.

If you were the only one rating the movie, this would spell trouble. But since you are just one voter of many, this vote only lowers the arithmetic mean score by some fractional amount.

Moreover, this is not entirely a bad thing. If lots of people hate a movie (and award it a 1), that's not a very good sign for the movie! It should not be in the top 250. This is why The Titanic has dropped, and good riddance I say.

As long as individuals are not attempting to game the system out of self-interest, their low scores represent a valid, independent opinion on the movie. Therefore, votes of 1 (on a scale of 1-10) don't represent a system failure. Instead, they allow users the ability to punish a movie if it is highly annoying or unpleasant.

Bottom Line: The fact that some voters give movies a 1 does not screw up the whole system. One vote has a small effect, and moreover, subjective individuals can sum to an objective population.

Tuesday, July 22, 2008

Tuesday Statisticz: Music that makes countries rich

Click here to read the whole thing.

You may recall that I've been planning to complete this for about two months now. It seems to me that this is the sort of list that every band would publicly want to be on the bottom of (for street cred), while secretly coveting the top stop, because that's where the money is. Also, if it ever became widely popular, I'd expect Nirvana's sales to shoot through the roof through an interesting application of the conformity theory.

Sunday, July 20, 2008

How far will The Dark Knight drop?

So I've promised not to discuss too many specific issues, but I've heard too much about the Dark Knight to not comment. Right now it's 9.7 on its specific page and 9.5 on the top 250 page (see here for why) with 46,000 votes. This is good enough for number one on the overall list, by a fairly high 0.5 margin. As wild as it may sound, none of this is highly unexpected. Even without the Dark Knight Chris Nolan movies average an unweighted 8.02, and that number will jump more if you weigh it. It's also worth noting that each of his three heavily funded movies are staples in the top 250. Christian Bale also does well on imdb, since 2002's Equilibrium his 9 movies have average an impressive 7.7 unweighted score. And when you factor in Heath Ledger's death and the fact that this is his last movie, the potential is explosive. The question now is how much the movie will drop, and at what rate it will do so. I'm curious to see, and I may keep track of its decay. Right now my prediction is that the score will even out to the midpoint of the top 250 score (currently 9.5) and the score of the top 1000 voters (currently 7.9). That would put the movie at 8.7, somewhere between #15 and #22 on the overall list. But yes, it will drop. Wall E dropped, No Country for Old Men dropped, There Will Be Blood dropped. They always drop. You will see.

Saturday, July 19, 2008

Against blogging about blogging about blogging

I love blogging, that much should be patently obvious. And I'll put up with blogging about blogging, because sometimes it's necessary. But what I don't like is blogging about blogging about blogging. It's like, if people want to blog about blogging, let them do it! You don't have go off and blog about blogging about blogging.

Then again, I don't mind blogging about blogging about blogging about blogging. If I did, that'd be pretty hypocritical. And now I'm dizzy.

The 10 most badass beards of all time

I just watched Tyler Cowen discuss world issues with Will Wilkinson on Blogging Heads. It was surreal because I had followed his writing for over a year and had yet to hear his voice or see what he looks like. Anyway, that video inspired this post. The 10 most badass beards of all-time, with links to pictures.

Honorable mentions: Zeus (too strict!), Karl Marx, (eliminated because he was a communist1)

10) Dumbledore. Would be higher but the whole movie thing has started to ruin the books for me. I'm turning into a huge snob, I know.

9) The dwarf in the Lord of the Rings. Technically his name is "Gimli" the dwarf (or should I call him a little person of Middle-Earth?) but who cares. All we remember is that he had a big red beard, he grunted more than he talked, and he carried around a huge battle axe. That's a recipe for success. Although, I must deduct points because he is a mythological character, not a real human. So sorry.

8) Socrates. Many an ancient Greek grew freakishly long beards, but Socrates stands out because he killed himself by drinking hemlock and because he appears in Bill and Ted's Excellent Adventure.

7) Teddy KGB. Played by John Malcovich in Rounders, the Eastern European poker player had a fairly badass beard that must have been intimidating at the table. Much better than the so-called "Jesus" of the real poker world. He also gets a boost because he undoubtedly had some Oreo crumbs lodged in there.

6) Robert E. Lee.

5) Noah. Recreated in Evan Almighty by Steve Carrell but it just isn't the same. I keep waiting for Steve to jump out and say "this business isn't about paper, it's about people!"

4) Teen Wolf. The athletic ability is a plus, as is the weird dribbling style, but I must deduce some points because he had hair all over his body, not exclusively on his face.

3) The Captain in Das Boot. Aka "Lt. Henrich Lehmann-Willenbrock", but again, who cares.

2) Tyler Cowen. Seriously, I was blown away by this beard. A tint of grey can do wonders.

1) Abe Lincoln. Duh. I will note that the most badass part of Abe Lincoln's life was when he was in a debate and he pulled the height card, saying, "Look at my opponent, he's four score and seven inches shorter than me... how could he possibly make a better president?" And I only made up 88.5% of that story.

1 Oh really, was he actually a socialist, is there some sort of difference? Thanks for enlightening me with your wealth of knowledge. You see, this is the real reason the world hates lawyers: because history majors are the worst, and history majors go to law school in droves.

Friday, July 18, 2008

Do not study human cognitive biases, part II

You know that song from Zoolander that goes, "relax, don't do it..."? I think it's the song that Mugate plays when he wants Derek to kill the prime minister of Malaysia.

That's how I feel about people studying human cognitive biases. The next time you feel like clicking here, or especially here, just say to yourself, "no, I will sit on my sofa and sip some green tea instead."

Case in point. One of my old basketball coaches would often tell us after games we won that he knew we were going to win. Even when we were down and things looked bad, he had no doubt. I think his exact words were, "the whole time, I just knew that we would win that game."

A friendly reaction would to roll your eyes and smile. But once you've read about human cognitive biases, an alert goes off, like the buzzer in Ben Bernanke's head when he spots an opportunity to bail out another financial institution. It screams: "HINDSIGHT BIAS!" And it won't go away.

It's the sort of thing that bothers you all night when you're trying to sleep. You want to go to PubMed to print out an article and anonymously slip it into his briefcase. Do you have any idea how dorky that is? Do you want to be that person?

I'll answer for you: no, you don't. This is your second warning. In ten seconds this blog post will self-destruct.

Thursday, July 17, 2008

Self-Experimentation, Ideas

I've been interested in self-experimentation recently, inspired mainly by Seth Roberts (who kept track of his sleep), Ramit Sethi (who tracks his spending), and Ben Franklin. I've told some of my friends, and they told me that I was a nerd, and I told them to tell me something over which I don't already cry myself to sleep every night.

But now I want to add something else. Since I plan to live forever or die trying, I've thought that maybe I could track some facets monthly as well as daily. So I'd have a day every month that I list some crucial measurements. So far I've got weight, tentatively mile time, and how long I can hold my breath for. Anything else? My goals are: easy to measure, take 0-10 minutes, and interesting. I tried subjective happiness, but I gave myself 5/5 three weeks in a row and decided it was silly.

It's my life, but you can be a part of it.

Monday, July 14, 2008

Do not study human cognitive biases, part 1

This is a public service announcement. My dad told me yesterday that if I went in the water I wouldn't be eaten by a shark. What he said exactly was, "You won't be attacked by a shark." This was after my mom told us both that there had been a shark warning a few days earlier at the beach. Additionally, the seals were fairly close in, which means that the sharks may have drifted toward the shore. I went in briefly (only to body surf), but that is beside the point. The point is how annoyed it makes me when people tell me the probability of certain events happening in absolute terms, like "won't." One of my friends used to tell me to go ridiculously fast on the road, and when I told him I didn't want to get a speeding ticket, he told me, "oh come on, you won't." I might, god dammit. What people need to do is to check their priors for selection and availability biases. Most do not. Once you realize their fallacy, this will annoy you, and when you (almost inevitably) project your anger, it will annoy the people you talk to. This is a vicious cycle, and you do not want to get in the throes. Do not study human cognitive biases. Do not click on these links. Read an adventure novel. Go oogle pictures of Scarlett Johansson. You will not thank me later, because you will never know what I am talking about. You do not want to know. I am warning you out of pure altruism. Good night, and good luck.

Sunday, July 13, 2008

Quote(s) of the Day

"But polls suggest that Americans are happy with a certain amount of flip-flopping: Mr Bush has all but destroyed the market in stubborn consistency." - The Economist

"Never remember to forget yourself in the bustle of things." - In a three year old notebook of mine, I forget who said it.

"There's nothing like visiting a foreign country like China to get an appreciation of what it's like to live under an authoritarian regime. I was reminded of this when I arrived home and found that the TSA had rifled through my baggage." - Alex Tabarrok on MR.

Friday, July 11, 2008

Against couch pillows

"If they took all the pillows away, and there was no more in the world, do you think that would affect you adversely, do you think you would notice it? Do you think it would bother you, like throughout the day, and oh there's no more pillows on this couch. I know at night, it would have an effect at night. What? A world without pillows. Twilight zone episode, there are no pillows. There's cushions, to sit on, but no pillows."

That's Larry David, in an early episode of Curb Your Enthusiasm, and I could not agree more. The only time couch pillows are good is when you spill some tomato sauce and you're too lazy to clean it up so you just put a pillow over the spill until somebody else deals with it.

Sunday, July 6, 2008

Jeffrey Sachs Lecture

Mr Sachs came to speak at the Vassar chapel on April 24th and it was clear that he's a great guy. He wore fairly tight jeans and a casual shirt, he squinted his eyes when he smiled, and he was doing the talk pro bono. These are fairly old, but here's my notes from his talk, because I'm about to throw away the binder they were in:
  • My friend Brian on why so many people showed up: "If it were just economics nobody would care, but it's ending poverty."
  • Bush sucked, in the future you should "have beer with your friends, but vote for somebody who can solve your problems."
  • John McCain apparently believes that the transcendent problem of our time is Islamic extremism, which garners scattered chuckles in the audience.
  • "I'm an economist so I chase the money. What is it costing us to be short-sighted?"
  • Finally, his 10 policy recommendations for the next president:
  1. Get out of Iraq.
  2. End the (ever controversial) Bush tax cuts for the rich, ostensibly totaling $250 billion.
  3. Create a new national institute of sustainable energy with $30 billion of it.
  4. Come back to the 1992 UN framework on climate change, which we signed. (Here. I add that it does not legally bind us to action, and that Clinton had no intention of signing Kyoto either.)
  5. Join other areas of international law: one for biodiversity, UN conventions on the law of the sea, and the Geneva conventions (for which the crowd went wild).
  6. Understand how to help poor countries. Stop subsidizing ethanol production.
  7. Invite leaders of African and Middle Eastern countries to a "dry lands" initiative to fight drought.
  8. ?
  9. Get eyes and ears back into the US government.
  10. When we don't act, people die. Stick to the UN millennium development goals: stop poverty, hunger, and disease.
And yes it was long-winded. I agree with numbers 1, 6, and sort of 7, I don't understand 4, 9, or 10, I disagree with 3 and 5, and I disagree with 2 with the caveat that we stop spending, but that's a whole other issue. If I was trying to be funny I would say that the only one I liked was number 8, which he apparently forgot about. Maybe you have to buy his book to get access.

Bottom Line: Government has done some terrible things but if we could just get the right people in office everything would change. And whatever you do, don't vote for McCain.

Thursday, July 3, 2008

Organizational musings

It's been a solid two years of writing here, and I'm already enjoying one of the major purposes, which is re-reading my old posts. Two years ago my goals for the summer were to read 6 books and make hella skrilla. What a loser I was. Anyway, some organizational notes:

1) I'm making Tuesday Statisticz a monthly instead of a weekly addition. This way I'll be able to deliver you the highest quality content. Hopefully they will offer more in the form of experimentation, natural or artificial, to deal with the correlation vs. causation issues I've been having.

2) My neurobiology type posts are being moved to the new BrainsLab.net blog, and my (dwindling) economics posts will be moved to the nascent Vassar Investment site. On this blog I'll try to focus on more broadly relevant stuff. Like movie rating systems.

3) I may be writing an e-book on the imdb top 250 once I finish watching them all, so if any of you have any feedback on the list, how it's made, and especially specific thoughts on movies, please e-mail me.