As you may know, I’ve been immersed in a mammoth study of Nevada casino slots’ hold percentages. It’s driven by the question: “Does raising hold percentage actually decrease play?”
That would seem to be the intuitive answer. If people get less value for their dollar (assuming that value=time on device), then they will be less ready to play.
The revenue statistics, however, don’t seem to say so. Here’s a chart that shows the total number of slots, the win, the hold percentage, and the total amount played, or handle, for the Las Vegas Strip 2000-2008.
As you can see, both handle and win steadily rise after the 2000-3 decrease. The hold percentage increased during the slump, and continued to go up as the handle and win grew.
Here’s another chart that may make more sense visually. I’ve had to multiply the hold % by 10,000 to get it to show up.
It doesn’t seem that rising hold percentage has anything to do with the handle. If it did, you’d expect to see handle falling as hold percentages go up.
This is where it gets tricky. You can spin it just about any way you like. Those arguing for loose slots can say that in 2008 casinos reaped what they had sown as players finally got fed up with tighter slots. But the visitor numbers don’t bear that out: people who came to Las Vegas still played, tight-slots advocates might argue.
There doesn’t seem to be a direct correlation, and the fact that revenues grew most impressively during years with big jumps in hold suggest that something else is at work here.
It’s worth mentioning that, thanks to higher hold percentages, that in 2008 total handle fell to 2004 numbers, but slot win fell only to 2005 numbers. That’s an argument that most executives–and shareholders–would probably find persuasive.
Comparing 2008 to 2000, we see more than $800 million more in revenue and over $6 billion more in play, with an installed base that’s shrunk by about 18 percent.
This is why the slot hold study is taking a while–it seems like every time I look at a source of data that will conclusively prove a link between hold and revenue (one way or another), I’m back to square one. After analyzing revenues, I thought that hold might be a fruitful avenue, but I’m still back where I started.
It doesn’t seem possible to prove a connection between hold and handle, and higher holds have led to consistently higher revenues. It’s just possible that the bean counters are right. On the other hand, handle has been falling faster than win, so those who demand looser machines may have a point. On the third hand, a decrease in hold percentage did not figure in the 2004-2007 boom. There just doesn’t seem to be a correlation.
In the paper, I explore some theories for the climbing hold percentages that may suggest the true reasons for the increase have less to do with the dictates of slot managers than changes in technology and player preference.