Las Vegas vs the Law of Large Numbers

Nassim Taleb likes to use Las Vegas as an example of a mathematical odds-based business which exists in economic theory much more often than it exists in real life. If you run a bunch of casinos with hundreds of thousands of punters coming and and betting hundreds of millions of dollars, then you can predict with some accuracy the amount of money you're going to make at the end of the quarter: it's called the law of large numbers. Except just look at what's happened at Las Vegas Sands Corp:

The company had a very unlucky summer: Its “table games win percentage,” which is the ratio of revenue from table games to chip purchases for those games, was just 14.7%. That was down from 23.4% in the year-earlier quarter, and an average of 22.9% over the prior 15 quarters — when it never dipped lower than 16.1%...
Its take from table games reached a peak of 36.9% on chips purchased in the fourth quarter of last year, and have declined each quarter since.

Now of course there's more than just luck involved in these numbers. The more time you spend at a table, and the higher your bets at that table, the greater the percentage of your chips that you're going to lose there. And a casino can lose its touch when it comes to persuading punters to stay long and bet large. Investors will be hoping that Sands is merely unlucky, rather than incompetent.