"Depressed people tend to think everything will go wrong, and they’re usually right."

Well there's a happy thought this morning from Gary Carson, author of the Math and Poker blog.

Although he may sound rather cynical, I do appreciate the point he makes in today's post about calculating expected value.  Sometimes in poker EV is really pretty impossible to calculate since it is dependent upon the action of others and these actions are often based on factors too complex and/or random to model effectively - especially while sitting at the poker table using only your own brain.

This is one of many areas where political science and poker overlap.  How can we effectively model human behavior?  This is, in my opinion, the goal of both poker players and political scientists.  In political science we have the added advantage of not having to rely solely on our brain and under time constraints like one does at the felt.  However, this is still an issue that everyone in political science struggles with. 

Although there is a debate in the field of political science over the goals of "good" political science, I firmly believe one of the primary goals of good political science is prediction.  What good is knowing the characteristics of something if we cannot say that it is likley/unlikely to happen again given certain circumstances?  This is part of the larger methodological debate within the field - how can we reduce complex human behavior to a mathematical equation that tells us (almost?) everything we need to know?  How do you measure core concepts in political science like power and political sophistication quantitatively?

I could write at length about this subject, but suffice it to say that we can only do our best, and a lot of the time, our best is good enough.  (This is the rallying cry of quantitative political scientists everywhere - if you don't believe me, just sit in on one of the classes I'm in sometime.  I think it is a perfectly legitimate response to ask someone questioning your measures: "Well if it is problematic, what would be a better, feasible way to measure it?").  We can often make pretty accurate predictions about the political world around us even though human behavior is erratic.

I think in poker it's the same way.  A lot of times our best effort may be good enough.  Yes, we might not be able to accurately calculate EV, but if we actually take the time to think about the situations we find ourselves in, I think good poker players generally make the closest to right decision enough of the time to be profitable in the long run.  I don't disagree with Gary Carson that people (self included) often overestimate EV, especially when considering implied odds.  However, how big of a problem is that in the overall profitability of people's games?  I guess if you are constantly and significantly overestimating EV then this could be a big hole in your game.  However, I don't think most thoughtful poker players overestimate EV that significantly that many times for it to make such a large difference.  Additionally, in many lower limit games, you don't have to think about the implied odds too much because you know there are a large number of people who will pay you off 9 times out of 10. 

And because we are only able to use our own brain power at the felt, I'll (rhetorically) ask Gary Carson (and any poker players reading my blog) the same question that political scientists ask everywhere: "So, how could we feasibly do it better?"