Sabermetrics for international relations?

Posted By Daniel W. Drezner Thursday, June 26, 2008 - 3:43 AM Share

I received the following e-mail from a reader earlier this week:

A crazy question I have to ask: would there be any possibility of developing a useful "sabermetrics for geopolitics"? I know such a correlation isn't obvious or perhaps even possible. I realize too, that SABR has been in baseball for over 20 years and is only now discovering the vagaries of fielding...and that geopolitics is orders of magnitude more complex. Moreover, what to analyze? how? what methods? what data..and on and on. I also realize that geopolitics and economics are already full of statistics, so "more" is simply not the answer in and of itself. That being said, is there more out there that can be done in how we organize and think about it? I have to think there is. Can it be studied and analyzed in ways that, like Billy Beane did, discover inefficiencies in way most people, governments and organizations look at things that can subsequently be looked at in a different way and/or exploited?

A lot if political scientists and international relations scholars would probably sigh pretty loudly at this question. After all, an awful lot of the field already uses a lot of quantitative measures to conduct statistical analysis of politics. Indeed, so much of the field uses these techniques that we've already experienced the backlash. As for foreign policy analysis, Bruce Bueno de Mesquita has not been shy about using formal and quantitative methods to make predictions in world politics. On the other hand.... Baseball Prospectus' Nate Silver recently made a splash when it turned out he was the driving force behind FiveThirtyEight, a blog about predicting the 2008 election. What Silver does is different from a lot of standard political science, in that he's not really interested in theory all that much. He's looking for the best empirical fit, rather than the most elegant and generalizable theoretical model. Because political science places a higher value on the latter, the kind of stuff that Silver does is -- unfortunately -- seen as less interesting within the academy. So maybe there is a niche to fill. Two problems, however. First, in a recent Q&A, Silver noted correctly that, "You can explain a higher percentage of baseball with statistics than you can of politics." Second, sabermetrics can't work unless it has statistics to measure -- and in international relations, useful large-N data is even more scarce than in domestic politics. Just because my imagination is limited, however, doesn't mean that a sabermetric approach can't add value. Indeed, the fact that I can't think of how merely suggests that I've been too professionalized. I'll leave it to the readers -- are there "metatechniques" developed by sabermetricians that could be applicable to international interactions?

EXPLORE:ACADEMIA, UNCATEGORIZED, INTERNATIONAL RELATIONS, POLITICAL SCIENCE, SABERMETRICS

(3)

HIDE COMMENTS LOGIN OR REGISTER REPORT ABUSE

PROFPTJ

7:51 PM ET

June 30, 2008

I'm quite skeptical of such

I'm quite skeptical of such "political sabremetrics" largely because politics is simply not organized like baseball. Baseball features a repetition of basic actions in a relatively closed and consistent environment, so the statistics are meaningful. Also, the actions on the field don't redefine the field or the basic rules of the game; that happens in politics all the time, and indeed the most interesting political moments are often precisely the moments when such redefinition happens. So in the end I'm a great fan of using the right conceptual tools for the right job(s), and I see very few baseball-like areas of political life where sabremetric tools would be appropriate for the explanatory job at hand.

HOT FUN IN THE SUMMERTIME…. « THE EVERYDAY IDEALIST

3:44 PM ET

July 6, 2008

[...] (of baseball) in [...]

PAUL WINSTON

12:32 PM ET

July 7, 2008

Popular opinion could be

Popular opinion could be considered along the same lines as sabermetrics, empirical statistics (both obviously gathered subjectively at times). Your sabermetrics solution might just be the theory of Constructive Sovereignty.

Constructive Sovereignty
Constructive Sovereignty is an emerging theory pioneered by John Maszka intended to address globalization's increasing onslaught against state sovereignty. The theory maintains that states are not the primary actors, their constituents are. Therefore, their preferences are not fixed. Since states merely represent the preferences of their constituents, they will only adhere to and ultimately embed those international norms that their respective constituencies will accept. Rather than push for larger and more powerful international organizations that will impose global norms from the outside in, the theory of Constructive Sovereignty posits that ultimately change must come from the inside out. That is to say, from each state's own constituency. As each state's constituents become more and more international, they will become more receptive to international norms and they will voice their acceptance of these norms both politically and (especially) as consumers.

It is therefore a central pillar of the theory that privatization is not only the driving force behind globalization, but also that private enterprise possesses the incentive to implement those international norms reflected in the preferences of consumers (profit). Private enterprise is also the primary consumer of proprietary data used to measure the preferences of consumers, and as such remains the most up-to-date source of changing consumer preferences. As private enterprise meets the increasingly international demands of consumers, it will itself become more international in scope. The cycle is self-perpetuating. In this way international norms are embedded and viewed with legitimacy by each state's constituency, while state sovereignty is maintained and respected.

The theory of Constructive Sovereignty ties in nicely with Maszka's model for combating terrororism. Maszka asserts that terrorism, regardless of its causes, is ultimately only possible with sufficient popular support. Consequently, only once we find a way to eliminate popular support for terrorism will we be able to eliminate terrorism itself. While some minimal definition is necessary to identify terrorism in a uniform manner, knowing what causes terrorism and collecting data on individual acts of terrorism is not as important as knowing how to stop it. When putting out a fire, while it is important to know what type of fire it is before attempting to put it out (e.g. applying water to an oil fire will have the same effect as using a flame thrower), firefighters understand that the key to putting out any fire is to remove its source of oxygen. This knowledge affords them a standard plan of action that varies only in detail (what kind of fire is it, and what is needed to remove the source of oxygen). Likewise, terrorism depends on popular support to sustain itself. Without popular support, the majority of funding, recruits and overall acceptance will disappear. Therefore, similar to putting out a fire, the primary goal for eliminating terrorism is to eliminate the sources of popular support. Measuring popular support for terrorism also affords us a method of measuring and predicting 1) the potential for terrorism in any given society, 2) the direction that acts of terrorism tend to be moving in (e.g. westward, eastward, or remaining static), and 3) broad trends in the support for terrorism, such as whether popular support is increasing among moderates, Westerners, etc...