The Lucky Donut

This week, I’ve been working on processing the World Series of Poker results to generate some stats for discussion on a well known poker web site.

All the information is out there if you want to try this at home.  You can get the full player lists in a PDF file for every event, and the list of winners is on a web page.

The challenge is that the player details are prone to mysteriously change from one list to the other.

For example, Michael “The Grinder” Mizrachi appears to have moved from Miramar to Miami in between buying his ticket and picking up a $1.5 million second place prize.  A whole bunch of players also seemed to suddenly move to Las Vegas from all over the country as soon as they picked up their cheque.

I expect their occupation also changed overnight, to “professional poker player” (or “adult movie star”) in anticipation of an inevitable Full Tilt (or Ultimate Bet) sponsorship deal.

Even having to contend with these delusions of celebrity, I don’t think I’ve done a bad job of getting some talking points out of the data.  A handful of players will have their wins recorded against someone else with the same name (but that happens to the best of us!) but mostly it’s a decent set of figures.  I’ll link to the full thing once it’s live.

But for now, the item that caught my eye the most is actually one where the player’s names don’t even matter, so I’m pretty confident in these stats.

In aggregating the entry fees and winning, and then calculating a return on investment by country, we can see that players from the USA are, as a whole – gasp – long term losers.

It’s not really any surprise to see a loss here.  Americans make up about 80% of the field and it’s a negative expectation game.  There’s juice of 6%-10% taken from every buy-in, no prize pool guarantees, no sponsor money added.  So, having the largest number of results available by far, you’d expect to see Team USA’s ROI tending towards this range sooner than any other country.

But it’s much worse than that.  After 16 events, they’re down a cool five million (about $350 per entry), with a combined ROI of nearly 20% in the wrong direction.

Across all players, the ROI comes in at roughly -9% (it’s skewed towards the top end of the range by the massive field sizes in the cheaper events, which are taxed more heavily).

Here’s the soundbite: Americans lose twice as much money at poker than the global average.

Sample size, yada yada yada.  Yes, possibly – but most of these events don’t have seven-figure prizes to quickly turn around that $5m national deficit.

I just thought it was an interesting number, and I’ll be watching it to see if the trend continues.