http://online.wsj.com/article/SB10001424052702304626104579123383535635644.html
The casino billboards lining America's roadways tantalize with the lure of riches. "Easy Street. It's Only a Play Away," screams one in Arizona. "$7.1 Million Every Day. We're a Payout Machine," reads another.
But how often do gamblers really win? What are the chances that a gambler will win on a single day or over a longer period? Don't bother to ask the casinos. Although they gather vast quantities of data about their customers for marketing purposes, including win and loss tallies for many regulars, casinos keep such information a closely-guarded secret.
Now, thanks to an unprecedented trove of public data detailing the behavior of thousands of Internet gamblers over a two-year period, The Wall Street Journal can provide some answers.
On any given day, the chances of emerging a winner aren't too bad—the gamblers won money on 30% of the days they wagered. But continuing to gamble is a bad bet. Just 11% of players ended up in the black over the full period, and most of those pocketed less than $150.
The skew was even more pronounced when it came to heavy gamblers. Of the top 10% of bettors—those placing the largest number of total wagers over the two years—about 95% ended up losing money, some dropping tens of thousands of dollars. Big losers of more than $5,000 among these heavy gamblers outnumbered big winners by a staggering 128 to 1.
The analysis comes from a database containing anonymous records of 4,222 Internet gamblers who wagered on at least four days on casino-style games of chance such as blackjack, roulette and slots. They played between 2005 and 2007 on websites run by a major European online gambling concern, Bwin.Party Digital Entertainment PLC.
Bwin made the information available to gambling-addiction researchers affiliated with Harvard Medical School, who posted much of the data on the Internet. Bwin says there's no reason to believe a more current sample of customers would show significant differences.
Although the online Bwin customers differed in some ways from those typically found in a U.S. casino, their win and loss patterns should be roughly similar because the games are similar, said Robert Hannum, a University of Denver specialist in gambling mathematics.
To check, the Journal asked Puneet Manchanda of the University of Michigan and Hee Mok Park of the University of Connecticut to analyze a private gambling database to which they have access, detailing two years of play by 18,000 holders of loyalty cards at a Native American casino in the northwestern U.S.
The researchers found similar patterns: Only 13.5% of gamblers ended up winning, versus 11% among Bwin customers, and the ratios of big losers to big winners were similarly large.
The Bwin data also offer a peek at the economics of the casino industry that only insiders normally glimpse. Among the findings is an extreme reliance on revenue from a small number of gamblers.
Of the 4,222 casino customers, just 2.8%—or 119 big losers—provided half of the casino's take, and 10.7% provided 80% of the take.
Such revenue concentration long has been quietly acknowledged in the casino industry, but the Bwin information may be the first to show it with hard public data.
The issue is a sensitive one in the industry because gambling critics often cite revenue concentration as one sign that casinos exploit gambling addicts, which executives dispute.
"Politically, we don't want to talk about it being more concentrated than other industries," said Andrew Klebanow, a marketing specialist who has consulted for dozens of casinos. He said the Bwin results are in line with his own estimates, based on confidential casino data, that many U.S. casinos get about 90% of their revenue from 10% of customers. (The two professors found that 9.3% of the gamblers at the Native American casino produced 80% of the group's revenue.)
Jim Kilby, a former professor who has written three books on casino management, said the scant number of winners among Bwin customers was surprising even to him, and should be educational to gamblers.
Although gamblers know the house has an edge, he said, "the average person doesn't understand the math" of the multiplier effect: "Casino games are nibbling machines, and the more nibbles you have, the bigger your losses."
The Bwin data clearly show that. The lightest gamblers—the 10% of customers who placed the fewest wagers over the two years—also had the highest winning percentage. About 17% of them ended up in the black—tough odds but still better than the dismal 5.4% winning percentage of the heaviest gamblers.
Among the whole group of 4,222 gamblers, just seven won more than $5,000 (€3,698) over the two years, while 217 lost more than $5,000. That's a 31-1 ratio of big losers to big winners.
Gambler No. 1357078, a Swiss man who was 56 years old when he opened his account, was a classic heavy gambler. He played an average of three days a week, typically placing more than 1,000 bets per day and averaging $9 per bet. He lost on 84% of the days he gambled, and over the two years gambled away more than $110,000.
Unless they cheat, about the only way gamblers can win at games of chance is to get lucky and then stop gambling.
That's how No. 1381787 emerged as the biggest overall winner. A 56-year-old Slovenian man, he typically placed only a few modest bets per day. Then he struck gold, twice winning more than $14,000 within 10 days. After suffering a partial setback, he stopped playing on the Bwin site, netting about $22,000.
A separate Bwin database covers poker play. Poker is partly a game of skill, and the outcomes reflect that. About one-third of the poker players classified as "most involved" by the Harvard researchers ended up winning money over time, while just 10% of the rest ended up in the black.
Despite the slim chance of winning shown by its own data, Bwin has been enticing gamblers on one of its sites to "play for huge rewards on classic casino games," adding "the odds are certainly in your favor!"
Joachim Haeusler, Bwin's responsible gaming manager, said the company provides entertainment and people shouldn't gamble "based on the idea to get rich, because they won't."
