This blog takes its name from Raven, a female chimpanzee who, in 1999, “picked” a portfolio of 10 stocks by throwing darts at a board arrayed with the names of more than 100 Internet companies. MonkeyDex, an indexed fund based on Raven’s picks, wound up ranking among the top Internet funds that year with better than 200% growth. Since then, the phrase “dart-throwing chimp” has stuck in the lexicon of financial analysis as a knock on the “expert” analysts who were outperformed by Raven — and pretty much anyone else who makes a living by prognosticating.
I like the title because I think it’s funny, but it’s also meant to serve as a reminder of the need for humility whenever we try to forecast, especially with events or processes that depend on human behavior. Self-proclaimed experts have been trying to predict political events forever, and we’re still not nearly as good at it as we’d like (and sometimes pretend) to be. In fact, with the noteworthy exception of forecasting election outcomes (see here and here for two good U.S. examples), most political scientists these days don’t even try.
If we can’t forecast political events with precision, why bother? For one thing, prediction is an essential part of the scientific method. Forecasts are an important way to test our theories against evidence we couldn’t see when we developed them in order to see if those theories are actually useful.
Here, though, I’m more interested in the benefits of applied forecasting — that is, forecasts that might be used by people trying to make decisions about their future courses of action. In global politics, these “people” include leaders of national governments, diplomats, and military officers interested in advancing or protecting their countries’ interests. They also include advocacy groups, development practitioners, corporate executives, and, last but certainly not least, individuals in the societies likely to be directly affected by the relevant events. Nowadays, warnings can quickly be distributed to (and even generated by) all of these constituencies, but those warnings won’t be useful if they aren’t fairly accurate.
And that, in my view, is why we need to keep trying to do it better. People and organizations will go about their business whether we forecast or not, and their actions will be based, in part, on their expectations about future events. Many of the rare events of greatest concern to observers of international politics — wars, revolutions, coups, mass atrocities, and the like — often have a big impact not just on the societies directly involved, but also on the world around them. Efforts to prevent those events or to mitigate their impact will usually benefit from planning, and planning benefits from foresight. Under these circumstances, even modest improvements in the accuracy of that foresight have the potential to help produce better outcomes.