JC Bradbury of Sabernomics (and a sports economist at the University of the South and this fall, Kennesaw State University) has posted a comment entitled So, You Want to Study the Economics of Sports. The post was written in response to a comment I offered on The Field of Sports Economics and some advice offered by Tyler Cowen at Marginal Revolution.
Let me repeat what JC states.
“It’s not that sports economics will soon be joining traditional fields—industrial organization, public economics, monetary economics, etc.—as equals, but economists are finding that sports need to be studied for two reasons: 1) The amazing amount of data available to study human behavior in a tightly controlled setting and 2) The lack of study of a human behavior where many interesting questions remain unanswered. The recent freakonomicizing (JC, is that really a word?) of the discipline has made the latter reason a more acceptable reason to study sports.”
JC and I both wonder about the same question. If you decide to study the economics of sports what kind of career will you have?
Like JC Bradbury, let me repeat the advice offered by Tyler Cowen at Marginal Revolution offered to people thinking about getting a Ph.D. in economics.
“Two core groups of people are well-suited to be economists:
1. You math GRE score is over 800, you are totally focused, you love working long hours on your own, and you have good enough letters of recommendation to get into a Top Six or perhaps Top Ten graduate school. Note that white Americans from this category have been partially preempted by competition from foreigners.
2. You could be happy as an academic without much of a research career. Working at a teaching school is a rewarding life, albeit a poor one relative to your investment in human capital.There is a third category, although you will fall into it (or not) ex post:
3. You do not fit either #1 or #2. Yet you have climbed out of the cracks rather than falling into them. You do something different, and still have managed to make your way doing research, albeit of a different kind. You will always feel like an outsider in the profession and perhaps you will be under-rewarded. But you will have a great deal of fun and in the long run perhaps a great deal of influence.”
When I read this I became a bit depressed. My overall GRE score topped 2,000, but that was on the strength of my analytical skills (a part of the test I hear they have since eliminated), not my math skills. Plus it is too late for me to attend a top school. Would it count if I just drove through one of those campuses?
I don’t want to be too critical of Cowen. He did favorably mention our book at Marginal Revolution. Still, I think aspiring economists need a view of our field from someone who is not a math star and did not attend a top school.
Let me start by throwing out a few numbers. There are more than 1,400 colleges and universities in the United States. Of course the vast majority of these are not top schools. So for most people entering graduate school, their future career will not be spent at places like Harvard or MIT. Can your life still be worthwhile?
I believe Paul Krugman once observed that there are people in his circle (Krugman graduated from MIT and teaches at Princeton) who are angry because they know they will never win a Nobel Prize. Which of course is just silly. I think at the top schools, though, that is the view. Your work must be recognized by people at top schools or it doesn’t have much value.
I think Tyler is influenced by this worldview, so consequently he separated the world into three camps:
#1: Highly mathematical people who the top people love.
#2: Pure teachers who might have a rewarding life but the rewards do not justify getting a Ph.D.
#3: Quirky people who do some stuff that occasionally gets noticed. But the stuff they do never gets enough recognition.
Notice his ranking is all based on a utility function (that’s economic talk) with only one element: how much do your peers in economics at top schools respect your work.
If that is all you consider in evaluating your career, then he is right, you are most likely going to be unhappy getting a Ph.D. in economics. Only one or two people get a Nobel Prize each year. Everyone else is a loser.
Of course there is more to life than respect of people at top institutions. First of all, the life of a teacher beats the hell out of 99% of jobs out there. So Cowen appears to underestimate option #2.
Let me add that I find teaching to be a terrifically rewarding occupation. As teachers we are trying to move information from our heads into somebody else’s head. That is not simple or easy. But it is definitely rewarding. And if people in typical 9 to 5 jobs saw how college professors worked and lived, I think they would be more than a bit envious.
Like JC Bradbury, though, I am more than just a teacher. Each of us has a clearly defined research agenda applying sports data to the study of economics (or economics to the study of sport).
Is this a worthwhile research agenda? I think much can be learned from human behavior studying how people make decisions in sports. Although sports may not be (and probably should not be) serious to fans, it is serious to the people in the industry. Examining how people in the sport industry utilize information can clearly tell us something about how people utilize information throughout the economy. So what we do has value. And I think JC would agree, studying sports is also a great deal of fun.
So here is my advice. If you like economics – and who doesn’t – and you want a rewarding life, follow your interests. The life of a college professor – even if you can’t spell MIT – is damn good. Of course, you might help us all out if you kept that a secret. I would hate to see an over-supply of economists drive down my wages.
— DJ
Wages of Wins Journal 
ChrisH
July 17, 2006
is it possible to score more than 800 on the math part of the GRE? I remember that being the maximum score but this is back when they had that analytical section you mentioned as dropped.
dberri
July 17, 2006
ChrisH,
I am not sure. I know I didn’t come close to 800 on the math part of the GRE. My analytical score was very high because I am great at solving silly puzzles. Students have told me that this part of the GRE has been eliminated. So I guess if I was trying to get into graduate school today I would have problems. How could I prove now that I was good at solving silly problems (such as determining the value of Allen Iverson)?
ChrisH
July 18, 2006
Just tell them Junkyard Dog is/was a top player and I’m sure everything out of your mouth from then on, would be accepted as scripture.
Now where is that post about the Bulls being better than originally thought?
Qball
July 20, 2006
The math portion of the GRE is a joke…it is like that they don’t expect most people to have learned any math at all in college. I would have expect economists to have a much better background in math than that. By the way, I’m pretty sure the max score in math is just 800.
Now my main problem with social sciences, in general, is the lack of control in the obtaining of statistics. In fact, I occasionally get into friendly arguments with social “scientists” about how scientific their field are regardless of the sophistication of statistical techniques. While that sometimes get viewed as a physical scientist looking down upon the social scientists, I hold the same skepticism toward their main stream social science results as I have toward results posted in this blog and sabremetric sites.
I guess my question is this: as someone who dedicates his/her life into sports economics, do you truly expect that any one equation or model that just capture everything or is this deemed impossible. What would be the “holy grail of sports economics”?
dberri
July 20, 2006
Qball,
I am not sure what it would mean to have an equation that captures “everything.” What exactly do you want the model to do?
Previously I have tried to be very clear about what I think we do and what we do not do. The model we present in our book connects the statistics tabulated by the NBA to wins. From this we can ascertain the value of each statistic. This approach allows us to estimate the value, in terms of wins, of each player’s actions in the past. So we are saying how productive the player has been. We discuss this “why” question, but that is a separate inquiry from the estimate of Wins Produced.
Now people have come back and said, “this stat is over-valued and this stat is undervalued.” This is fine as an hypothesis, but you have to do better. Can you show that if we value statistics as you suggest that explanatory power and predictive power increases? If you can, great. Then we will do it your way. But you have to show some evidence behind your claim. You can’t just say, “hey this makes sense to me.”