Statistics for gambling

Posted to Statistics  |  Tags: ,  |  Nathan Yau

Statistics isn’t just for finding out how our world works and how companies can improve their business. No. It’s also for useful stuff, like, you know, gambling. Great interview with Edward Thorp [pdf], who’s best known for bringing card counting in blackjack to the masses.

I received my PhD in mathematics and then went out into the university world to teach. As it happened, I’d always had an interest in applications from all of my science play in my high school years. One idea I’d had during those days was the physical predicting of roulette. That idea had stuck with me, so as I was getting my PhD, I was working on that problem, just on the side for fun. That gave me an outlook toward gambling games that later paid off in the market. Although conventional wisdom held that you couldn’t beat these games, the outlook was that that wisdom was not necessarily true and, in fact, was probably wrong. Gambling games, which were perceived to be efficient — in the financial-world sense of the word — might not be. In fact, I was convinced that wasn’t the case in roulette. So I came to this orientation that the conventional wisdom wasn’t right. That led me not only to build a wearable computer for roulette in conjunction with Claude Shannon of the Massachusetts Institute of Technology, but also to investigate card-counting in blackjack. I happened to see an article on blackjack strategy published in a statistical journal that was fairly close to even. After I used it just for fun, I came back and figured out a way to construct a winning strategy for the game.

[Edward Thorpe via @pkedrosky]

1 Comment

  • If you find Ed Thorpe and Claude Shannon interesting (as you should!) the book Fortune’s Formula by William Poundstone is a great read.


A Day in the Life of Americans

I wanted to see how daily patterns emerge at the individual level and how a person’s entire day plays out. So I simulated 1,000 of them.

Divorce Rates for Different Groups

We know when people usually get married. We know who never marries. Finally, it’s time to look at the other side: divorce and remarriage.

One Dataset, Visualized 25 Ways

“Let the data speak” they say. But what happens when the data rambles on and on?

10 Best Data Visualization Projects of 2015

These are my picks for the best of 2015. As usual, they could easily appear in a different order on a different day, and there are projects not on the list that were also excellent.