Gambler's ruin

hooting and cheering was not in order.

Between races, Benter struggled to make his algorithm stay ahead of a statistical phenomenon known as Gambler's Ruin.

To explain Gambler's Ruin, imagine a game of heads or tails where you are making $1 bets on the outcome.

Now let's say you have $10 to play with and your opponent has an unlimited amount of money.

You might think that neither player is likely to walk away with any more than they started with, and mathematically this is true, but it does not take into account swings or streaks of one outcome or the other.

And the impact of all of that bad luck can be diminished by betting thousands and thousands of times.

Kelly's equations applied to the scale of betting made possible by computer modelling seemed to guarantee success.

This was what Benter desperately needed, a system to stay ahead of this gambler's ruin outcome.

You know, um, uh, I, I try to bring math into this book that you may not have seen in, in college or high school.

Um, you know, we, we, we delve into things like the gambler's ruin problem.

You know, say, um, say you wanna get, you wanna get to a concert, and the ticket, uh, you know, you're, you're in Las Vegas, and the ticket's gonna cost $100.

You have $60 on you And you say, "I...

We're just a couple of successful idiots using AI to simplify our lives and optimize our businesses.

Peter, true or false, the French worked out the math on gambler's ruin, and the British have been faithfully proving it at the bookies ever since.

100% true, although we invented the phrase bookie, so we get some credit for inventing our own downfall, don't we?

Yes.

There's a 300-year-old probability puzzle, and it's act- technically 370 years old.

It's a Substack piece, uh, from your online business, and it just published maybe a week and a half ago.

And it's on the probability puzzle called gambler's ruin.

And there was a line that I had to read twice, and it said: "Low-cost experiments let your skill advantage do its work.