The post suggests more, though doesn't illustrate it all (because I couldn't think of a simple way to do so). The key is that x and y can be statistically related yet still have zero correlation (because that is just one specific measure of relatedness). Interpretations of causation are not data driven. They come from theory.
The role of confounded factors came up in the examples because that is just an easy way to illustrate the points.
I get my blog content via RSS. All that fancy formatting and layout never reaches my eyes. That plus consideration of the average half-life of a blog post makes additional effort...well...kind of pointless. (Unless the point is to make art, which is different.)
I confess to knowing very little about the limitations of URLs so perhaps this is s dumb question: why not go all the way down to the shortest possible? Can they just use the minimum unique string of characters? For example http://xyz7g. What are the limits?
(a) It is a highly stylized example to illustrate game-theoretic concepts. In a sense I am asking the reader to imagine that THE GAME is about the runner getting to second or not holding all else about the game constant. That's not really baseball, but it is a baseball-like situation.
(b) The situation described is both vague (was the probability referenced by the color commentator for this runner? For this team? For this situation?) and specific (a specific runner/pitcher pair, there are zero outs not two).
(c) You are right about the goal of baseball. I had originally written it with the goal being scoring. But then it is a vastly more complex game.
I'm honored to be cited here. I heard about Y Combinator on EconTalk. I had no idea it was a place to find links to interesting posts and news.
Anyway, this post is applicable to investing and bubbles. How far from common knowledge is assumption about rationality? If it is sufficiently far then it is rational to buy/sell at a price very different from the one that would obtain with common knowledge rationality.
The role of confounded factors came up in the examples because that is just an easy way to illustrate the points.