Author Archives: kevin

About kevin

I write the posts

Planning around the planning fallacy

Posted on
The planning fallacy is our inability to correctly estimate how long things take:
Buehler et. al. (1995) asked their students for estimates of when they (the students) thought they would complete their personal academic projects. Specifically, the researchers asked for estimated times by which the students thought it was 50%, 75%, and 99% probable their personal projects would be done. Would you care to guess how many students finished on or before their estimated 50%, 75%, and 99% probability levels? * 13% of subjects finished their project by the time they had assigned a 50% probability level; * 19% finished by the time assigned a 75% probability level; * and only 45% (less than half!) finished by the time of their 99% probability level.
We are extremely bad planners; even if we've turned in a hundred essays at the last minute, we still think we're going to turn in this one on time. Every night I plan on not hitting the snooze button, and then morning rolls around and I crave ten more minutes of sleep. These instances are trivial but the planning fallacy imposes real costs: think about all of the government projects that balloon way over budget and past schedule, because the contractors and government overseers failed to plan well. I've found two ways to get around the planning problem. #1: Move up the deadline. Usually this involves texting one of my friends and telling them, "If I don't finish this essay by 3pm, then I owe you $20." Very effective, and because I'm pretty competitive I usually avoid paying out, although I don't do it often enough. When you consider how much it costs to go to school, and how much good grades help you get a job, it's not too ridiculous to use $20 as an incentive to get things done. #2: Leave yourself no wiggle room: If an essay's due at 3pm, then schedule a whole bunch of other work from 9 to 12. Deliberately ignore thinking about the essay until then. Then, write your essay in three hours. That way, instead of spending six hours working on the essay (and procrastinating), you get it done in three hours, and you get three hours of work done in the morning as well. If you hit the snooze in the morning, work late, so you have no wiggle room for sleep (Generally, I have to get up by 9 to get to breakfast). These aren't great alternatives but they're better than mindless procrastination, and spending three hours surfing YouTube before getting down to business. If this doesn't sound appealing, you can always plan on becoming a person who finishes their work ahead of time.

Liked what you read? I am available for hire.

Links for today

Posted on
I've been using Aardvark. Not sure how they plan to make money, but it's pretty cool to ask a question and get an answer back nearly immediately. Loneliness can be transferred from person to person. I've also been dismayed by the reaction to David Obey's war surtax, a proposed income tax which would cover a small percentage of the cost of going to war in Afghanistan. It's one thing to oppose it because it would increase taxes during a recession and another to oppose it on ideological grounds. As long as we can push the costs of war further down the road, we'll be biased in favor of going to war. If we're not willing to spend one percent of our income on the war in Afghanistan, then maybe we shouldn't be there at all. Unfortunately, the cost of war is not salient for many Americans. Here is a moving speech by Diane Savino, on the eve of the gay marriage vote in the New York Senate.

Liked what you read? I am available for hire.

The NFL oligarchy

Posted on
Every American sports league has a draft at the end of the season to allocate players graduating high school and college to teams. To ensure competitive balance, the teams with the worst records are allowed to pick first. This paper by Richard Thaler and Cade Massey suggests that despite having the top pick, those teams might be losing out:
A question of increasing interest to researchers in a variety of fields is whether the incentives and experience present in many ìreal worldî settings mitigate judgment and decision-making biases. To investigate this question, we analyze the decision making of National Football League teams during their annual player draft. This is a domain in which incentives are exceedingly high and the opportunities for learning rich. It is also a domain in which multiple psychological factors suggest teams may overvalue the "right to choose" in the draft; non-regressive predictions, overconfidence, the winner's curse and false consensus all suggest a bias in this direction. Using archival data on draft-day trades, player performance and compensation, we compare the market value of draft picks with the historical value of drafted players. We find that top draft picks are overvalued in a manner that is inconsistent with rational expectations and efficient markets and consistent with psychological research.
Unwittingly, when leagues allow the worst teams to pick first, they may induce those teams to overpay for a top player, and hurt their own cause going forward. I contrast that to the European system, where there's no draft, and the best teams systematically overvalue players. Liverpool, for example, paid 19 million pounds for Robbie Keane and then sold him back to the same team a year later for 12 million. Keane's not the only one: Liverpool overpaid for players like El Hadji Diouf, Djibril Cisse, and others. Which is better for competitive balance? Bad teams may have stingy owners, so inducing them to overpay for good players and open their wallets may lead them to become more competitive. However, the best teams overpay because they can; it costs more for them to improve their team at the margin, but even though they're overpaying, they're still improving their team. In either setup, draft or no-draft, as long as players are mis-valued, there are arbitrage opportunities - in the draft, trade away the first pick, or without a draft, develop an excellent farm system, and sell players to other teams for absurd amounts once they become good. This is another reason why we should scrap the draft.

Liked what you read? I am available for hire.

How can we make students write shit that’s worth reading?

Posted on
Responding to Steven Pressfield's blog writing lesson ("Rule #1: Nobody wants to read your shit") Ben Casnocha notes:
In school anything you write or do will be read and graded by a teacher paid to do so. In the real world nobody wants to read your shit, and you have to earn their attention every single day. Last year in a post titled You Have to Make People Give a Shit, I extolled blogging as a way to learn this value.
Teachers can't just give a 0 to someone because their writing is boring, even though in real life, that's what happens when you send in a crappy cover letter, and even though the goal of most educational writing is to prepare for real life writing. This means that students don't really have an incentive to be interesting. When we're told our writing needs a rewrite, often we just slap a few changes on the old piece, maybe Thesaurus a few words, and send it back in. It's tough to grasp the concept that no one wants to read your writing, especially when you know the teacher's going to read it, and you've spent the last ten hours writing it; it feels like your baby, even though your feelings about it are better than anyone else's. And it shows; former college admissions official Michelle Hernandez admitted that 90% of student essays that she read didn't make an impact. It's probably true that teachers give better grades to interesting papers, because after reading 20 boring ones, it must be nice to read a paper that reads smoothly. But this isn't an explicit part of the grading rubric. How can we make students' writing more interesting? Ben recommends blogging, as do I, except it's a lot of work for many students. The two things that will make anyone's writing better - read more and write more - require a large time investment. Here are simple changes to the class incentive structure that could work. Students are too accustomed to writing for an audience of one; get them to practice writing for a wider audience. Turn papers into competitions, where the winning piece gets published on the school's website, or read aloud to the class, or published in the local newspaper. This could be pretty easy; if you have kids write every day, just set up a class website and post "only interesting pieces" there. Unlike parents and most people in academia, I tend to regard competition as a good thing. One winner out of 20 is about right; think about how many applications employers get vs. how many applicants they end up hiring. Another alternative is to make a small part of every writing assignment - say, 5 points out of 100 - based on its interestingness. This is a simple exercise; the teacher simply flips through every paper and discards the ones that don't grab her right away. The ones in the discard pile get a 0 for interestingness. It's tough to peel away the layers of hackneyed phrases and tricks that accompany bad writing, to get to the interesting substance. Only recently have I started looking at whole paragraphs and pieces I've written and deciding that they need to be entirely rewritten. At the margin, if students spend more time writing (and more time writing interesting things) they will be better off, as will our society.

Liked what you read? I am available for hire.

How to mislead with graphs: Verizon edition

Posted on

You’ve probably seen the new Verizon commercials demonstrating that their 3G network is so much bigger than AT&T’s 3G network:

Nathan at FlowingData has a good post up about the differences. Obviously the Verizon 3G map covers more surface area, but that doesn’t translate to better coverage, because I spend most of my time in Claremont and the Bay Area, and little time in North Dakota. Here’s a map showing the population density in the USA:

I would guess that the AT&T map covers all of the major cities, where almost all of the people who use 3G live. Rural people aren’t going to have a very high demand for 3G coverage, so it doesn’t make much sense to cover them, or for 3G users to judge a company’s service based on how much rural area it covers. Verizon’s coverage probably is better than AT&T’s, but their ad campaign is not a good way to demonstrate that.

Liked what you read? I am available for hire.

Links for today

Posted on
Sign of the apocalypse: Test prep for 3 year olds. Website: Mystery Google, which returns the query that the person before you searched for. Some people are using it to send messages. Wonky: How accurate can NFL prediction algorithms get? Answer: As the distribution of wins and losses is indistinguishable from a schedule in which 52% of the games are determined by sheer luck, the best anyone can do is predict 76% of games correctly. Video: Deer finds water, gets very excited.

Liked what you read? I am available for hire.

Jerking: the new dance move I might attempt at 2 am on Saturday

Posted on
The NYT has a front-page profile of "jerking," the new dance move that originated in LA and is now 'sweeping the nation.' Here's the requisite video: The photos are great but as usual, the NYT's attempts at chronicling youth fashion are hilarious. No one explains cool better than a dance historian:
Seen in formal terms, said Sally Sommer, a dance historian who teaches at Florida State University, jerking may merely be a cousin to the “lambada or the twist.” It is certainly, Ms. Sommer said, less physically demanding than krumping or vogueing or the other highly skilled and innovative urban forms of dance. But the lambada was a fad. The twist was a fad. And jerking, its adherents say, has a cultural resonance that goes beyond the Reject and the Tippy Toe.
They go on to quote a music video director, the music editor of LA Weekly and one of the stars of the jerking movement. Maybe they should realize the irony inherent in trying to describe the new cool movement to a nationwide audience.

Liked what you read? I am available for hire.

Will our best works of art survive for 2500 years?

Posted on
CM: Well, I don't know what of our culture is going to survive, or if we survive. If you look at the Greek plays, they're really good. And there's just a handful of them. Well, how good would they be if there were 2,500 of them? But that's the future looking back at us. Anything you can think of, there's going to be millions of them. Just the sheer number of things will devalue them. I don't care whether it's art, literature, poetry or drama, whatever. The sheer volume of it will wash it out. I mean, if you had thousands of Greek plays to read, would they be that good? I don't think so. CM: This is just entry level to what's coming. Just the appalling volume of artifacts will erase all meaning that they could ever possibly have. But we probably won't get that far anyway.
There are two questions here: 1) If someone writes the next "Oedipus Rex," will we notice? and 2) Will our generation's best work survive? Unlike Mr. McCarthy, I think the answer to both questions is yes. Popular works of art follow a power law; the best known pieces are viewed exponentially more often than the least popular ones. People want to share and discuss things they've seen, and (some of them) want to try and view the best works of art our culture is producing. Many people have their ears to the ground, filtering out the best material from the noise. The chances of generating 7 excellent works out of 10,000 are much higher than the chances of generating 3 excellent works out of 100; even if we only recognize four of the seven, we're still better off. We'll have more high-quality works of art and continue to do so as the population keeps expanding and the cost of spreading information is low. We tell stories as a way of preserving our culture; the book of Genesis was passed down orally for thousands of years before it was written down by anyone. I don't think that our stories will vanish or get lost in the haze. The works of art we're generating today have a much greater chance of surviving than the works of the Greeks, because they're stored in so many different places. Sure, there's a reputation effect; relatives of famous artists, and people who have already done something good will get their work noticed by a lot of people, even if it's not very good. But the probability you'll tell a friend about some movie you saw increases with its quality. Works of art that withstand the test of time must be very good, and remain relevant even when our lifestyles have changed beyond recognition. We also have crude rating systems; reviews and Top 100 lists will give future historians a place to start. Perhaps we won't recognize "Oedipus Rex" right when it's published, but as long as it's accessible by future generations, someone will find it. I'm encouraged by the rise of some artists long after their career has finished, such as Nick Drake or James Joyce.

Liked what you read? I am available for hire.