Risk-taking

March 1, 2009
159 Views

I’m reading The Poker Face of Wall Street by Aaron Brown. The first half has been very good so I’ll write some thoughts on it when I’m finished.

Off-topic, one cool thing he mentioned was that in Texas Hold’em the odds are best when you have a hand that either beats everyone or loses to everyone based on the card that shows up in the river. I immediately recognized this as another example of the alignment principle I wrote about earlier in A Game at Hell’s Gate. It’s a more practical example though.

Anyway I was thinking about what creates risk and why risk taking is so common. In my computer science class we learned that a uniform random variable can be constructed from an infinite number of coin tosses, and that any distribution can be constructed from a uniform random variable by using the target distribution’s inverse CDF. In The Poker Face, Brown claims (I cannot find verification) that Claude Shannon “built a mechanical hand that could flip coins that landed reliably heads or tails, whichever he specified.” (p.10) Combining these three things seems to imply that there is no intrinsic chance in the universe. Alternatively you could make a similar argument that a complete unders


I’m reading The Poker Face of Wall Street by Aaron Brown. The first half has been very good so I’ll write some thoughts on it when I’m finished.

Off-topic, one cool thing he mentioned was that in Texas Hold’em the odds are best when you have a hand that either beats everyone or loses to everyone based on the card that shows up in the river. I immediately recognized this as another example of the alignment principle I wrote about earlier in A Game at Hell’s Gate. It’s a more practical example though.

Anyway I was thinking about what creates risk and why risk taking is so common. In my computer science class we learned that a uniform random variable can be constructed from an infinite number of coin tosses, and that any distribution can be constructed from a uniform random variable by using the target distribution’s inverse CDF. In The Poker Face, Brown claims (I cannot find verification) that Claude Shannon “built a mechanical hand that could flip coins that landed reliably heads or tails, whichever he specified.” (p.10) Combining these three things seems to imply that there is no intrinsic chance in the universe. Alternatively you could make a similar argument that a complete understanding of physics would make the future deterministic.

So then risk is a human invention. It is spawned whenever someone makes a decision based on imperfect information. Imperfect information could be the result of either something random, like the roll of a die or a deal from a shuffled deck, or a decision that had to be made hastily- like in an emergency situation.

The fact is is that humans cannot see deep into the future because the possibilities increase exponentially. Even a dramatically limited universe like a chess game overwhelms the best-trained brain. Similarly, mathematical proofs are such a challenge because often you cannot tell if a result will be “interesting” without seeing multiple steps into the proof, which branch exponentially with the number of steps into the future you project.

Therefore making decisions under uncertainty is inevitable. Being a risk taker is the skill of going as far down each branch of the imagined decision tree as possible, clipping off the rest, and rapidly replacing it with a probability that condenses the odds of success from then onward. Or rather it’s the willingness to act on this approximation with certainty.

Hope you had a happy/filling Thanksgiving. Thanks again Squanto