Skip to content
 

There are four ways to get fired from Caesars: (1) theft, (2) sexual harassment, (3) running an experiment without a control group, and (4) keeping a gambling addict away from the casino

Ever since I got this new sound system for my bike, I’ve been listening to a lot of podcasts. This American Life is really good. I know, I know, everybody knows that, but it’s true. The only segments I don’t like are the ones that are too “writerly,” when they read a short story aloud. They don’t work for me. Most of the time, though, the show is as great as everyone says it is.

Anyway, the other day I listened to program #466: Blackjack. It started with some items on card counting. That stuff is always fun. Then they get to the longer story, which is all about a moderately rich housewife from Iowa who, over a roughly ten-year period, lost her life savings, something like a million dollars, at Harrah’s casinos. Did you know they had casinos in Iowa and Indiana? I didn’t. Anyway, the lady was a gambling addict. That part’s pretty clear. You don’t lose your life savings at a casino by accident.

The scary part, though, was how the casino company craftily enabled her to lose more and more. They offered her freebies (of course), but also, after big losses they’d call her on the phone at home and sweet-talk her into coming in and losing more. Sarah Koenig reports:

These kinds of calls are standard, apparently. Angie Bachmann always gambled at Harrah’s Casinos. The company is now called Caesars Entertainment, since it bought out Caesars. It’s the largest gaming company in the world.

And Bachmann happened to start gambling at the same time that Harrah’s began to overhaul its marketing strategies. Harrah’s knew how to track each gambler’s habits through Total Rewards cards, that each gambler, including Bachmann, would use throughout the casino. And that told the company exactly how much money each player spent, on which games, and at what frequency. The company would then use that information to tell them exactly what kinds of perks and rewards would keep certain gamblers coming back and at exactly what juncture to offer those perks and rewards.

I was listening to this on my bike, and I thought: this sounds familiar. I remember Caesars being celebrated as an admirably modern, data-oriented business. When I got home I googled a bit and found this interview with the CEO of Caesars entertainment:

Gary Loveman, the CEO of Caesars Entertainment, says there are three ways to get fired from the hotel and casino company: theft, sexual harassment, and running an experiment without a control group.

Loveman continues:

We need to overcome hunch and intuition with empirical evidence. . . . We can start with a hunch or strong belief, but we act on it through experiment. We want evidence. We’ve gone from the introduction of experimentation as a technique to a culture of experimentation as a business discipline. We hire people predisposed to do this by temperament and by background. Organizationally, we’re committed—and I’m committed—to making sure we have the discipline to have the decisions we make informed by this evidence.

The interview is entirely positive and nowhere does it suggest that there might be social costs to making this business more efficient. In retrospect, though, perhaps this bit should have set off some warning bells:

They might test which is likelier to get customers to spend more: a free meal or a free night in a hotel.

“Get customers to spend more” = “Offer gambling addicts hundreds of thousands of dollars on credit to stay at the blackjack tables.”

Here’s a quantitative question, which I do not have the answer to: What percentage of casinos’ profits come from addicts who are in over their heads? (I’m not an expert on addiction, and here I’m not counting people who have a gambling habit that does not take over their lives.) I have no idea. But the data wonks at Caesars must have an idea.

To her credit, Koenig, the This American Life reporter, looks into the question of the casino’s policy on problem gamblers. She quotes an earlier interview with the Caesars CEO:

Gary Loveman: We do not wish to be in the business of serving addicted gamblers. . . . our objective is to try to identify addicted gamblers as best we can and encourage them to seek treatment and help. And to the degree they’re willing to identify themselves as addicted or troubled gamblers, not serve them in any fashion, not market to them, not lend them money . . .

The magic words are “to the degree they’re willing to identify themselves as addicted.” It’s not enough that people display a pattern of addiction, they have to self-identify. Koenig says:

I ran all this by Caesars Entertainment, the supposed difference between their policy on paper and what actually happens on the casino floor. In response, a spokesman wrote to me that diagnosing problem gambling is extremely difficult, even for trained clinicians.

Where does this all go?

I don’t want to get on some moral high horse here. Lots of people like to gamble, I enjoy the occasional poker game myself. Caesars might be doing some bad things—I think they are doing some bad things, it seems immoral to me to have an employee whose job it is to convince big money losers to come back to the casino to lose the rest of their life savings—but, hey, I’ve heard that Coca-Cola does some evil deeds too but I keep drinking the stuff.

The Caesars case (I keep wanting to write Caesar’s but apparently no, it’s Caesars, just like Starbucks) interested me because of the role of statistics. I’m used to thinking of probability and statistics as a positive social force (helping medical research or, in earlier days, helping the allies in World War 2), or mildly positive (for example, helping design measures to better evaluate employees), or maybe neutral (exotic financial instruments which serve no redeeming social value but presumably don’t do much harm) or moderately negative (“Moneyball”-style strategies such as going for slow sluggers who foul off endless pitches and walk a lot; it may win games but it makes for boring baseball). And then there are statisticians who do fishy analyses, for example trying to hide that some drug causes damage so it can stay on the market. But that’s a bit different because such a statistical analysis, no matter how crafty, is inherently a bad analysis, trying to obscure rather than learn.

The Caesars case seems different, in that there is a very direct tradeoff: the better the statistics and the better the science, the worse the human outcomes. These guys are directly optimizing their ability to ruin some people’s lives.

OK, maybe you could argue that it’s a good thing to take a million dollars from someone who doesn’t really need it (as seems to be the case for this Iowa woman; at least, it doesn’t sound like she is starving or homeless) and transferring it to middle-class casino employees. They could rename the casino Robin Hoods and see if anyone gets the point.

But I don’t see anyone making this point. Gary Loveman doesn’t say, Yeah, what of it? We’re fleecing the suckers, taking millions from rich addicts who don’t need the money anyway. Instead, he implausibly denies that they are systematically plucking people. What’s it like to be a statistician working in the keep-them-coming-back-for-more department of the Caesars marketing department, knowing that, the better you do your job, the worse you’re hurting people? Maybe someone holding such a position feels no worse than I do, every time I drink a Coke.

P.S. I’m annoyed that This American Life did a show called Red State Blue State and didn’t interview me! They also have this show with Malcolm “Igon value” Gladwell.

P.P.S. My sound system is an ipod plugged into the Goal Zero 90401 Rockout Speaker which goes for $25 on Amazon. It’s loud enough that I can hear music or podcasts in traffic.

78 Comments

  1. Joel says:

    What’s it like to be a statistician working in the keep-them-coming-back-for-more department of the Caesars marketing department, knowing that, the better you do your job, the worse you’re hurting people?

    Probably not too different than being a statistician in the keep-them-coming-back-for-more department of the perpetual political campaign.

    • Andrew says:

      Joel:

      I don’t think ordinary political campaigning, or political advertising, or (for that matter) advertising in general, is quite the same. In campaigning or advertising you’re pushing a “product” but not in general to addicts whose lives are being destroyed. From that perspective, it’s a bit creepy to me to see Caesars described as some sort of paradigm of statistics-centered business.

    • zbicyclist says:

      There are a lot of dark areas to predictive analytics.

  2. I’d wager most of their profits come from heavy gamblers just as most alcohol is consumed by alcoholics, so the right comparison would be “anytime I sip a glass of Pinot Noir” and not Coca-Cola.

    • Andreas Baumann says:

      “most alcohol is consumed by alcoholics”

      A highly dubious claim. Do you have some data to support that?

      • A bit glib in the word choice, yes, but not absurd:

        http://www.ncbi.nlm.nih.gov/pubmed/10096312

        In the two studies reported, “the top 2.5% of drinkers by volume account for 27% and 25% of the nation’s total self-reported alcohol consumption in the telephone and in-person surveys, respectively; the top 5% account for 42% and 39%; and the top 20% of drinkers account for 89% and 87%”.

        *

        I wonder how much the consumer credit industry resembles this: good for most of its users, destructive for a small minority, but most of its clients being of the latter kind.

        • The last sentence should have read:

          I wonder how much the consumer credit industry resembles this: good for most of its users, destructive for a small minority, but most of its business coming from the self-destructing clients.

          • SRM says:

            Quite a lot, actually. Another NPR show, “Planet Money”, teamed up with “This American Life” and followed the mortgage crisis of 2008. One thing they discovered was the researcher for the credit card companies who pointed out that they got more money per capita from a person who sometimes missed a monthly payment or who kept trying to pay but couldn’t quite make the full balance and so built up late fees and interest than they ever made from the person with stellar credit. That lead to the companies giving credit lines to college students in the mid-to-late 1990s, something that *never* would have happened in earlier years. This lead directly to the offering of mortgage loans to people who couldn’t really afford them because it was better for the industry to have near-failures being drained of additional fees than to have mostly successful folks paying only the basic interest.
            The original show (link includes a full transcript):
            http://www.thisamericanlife.org/radio-archives/episode/355/the-giant-pool-of-money
            A good written summary:
            http://www.dailyfinance.com/2011/01/20/whos-to-blame-for-the-mortgage-mess-banks-not-homeowners/

            This practice fell into disrepute in the wake of the mortgage collapse. But here’s a 2012 story on the return to that practice… and it is even teaching people with bad credit how to go get such cards! Yeesh.
            http://money.cnn.com/2012/04/19/pf/credit-cards-bad-credit/index.htm

  3. Elizaveta Markova says:

    You’re just noticing that statistics can be used for evil? When one of our most respected journals is the former Annals of Eugenics?

  4. Bill Jefferys says:

    We have kept careful records of our gambling since we were married, and I can honestly state that our total winnings are three times what we gambled.

    When we were on our honeymoon in Germany, my wife stuck 20 pfennigs in a slot machine, won 60, and quit while she was ahead.

    :-)

  5. Mike Bailey says:

    I listen to podcasts on my bike too. So what’s the new sound system?

  6. Mike Bailey says:

    And btw the comment on eugenics is knee jerk and incorrect. The Nazis were evil because their eugenics was coercive and they murdered lots of people. However, eugenics is many things, including testing for genetic conditions such as Down syndrome. Is that evil? See: http://chronicle.com/article/The-Eugenic-Impulse/135612/ for a non-knee jerk consideration.

    • revo11 says:

      Eugenics is based on an assumption that there exists a time-invariant, universally-optimal genetic composition. I’m okay with not calling it “evil” and just calling it “stupid”.

      If you’re going to remove the value judgement aspect of eugenics and discuss relating genotypes to phenotypes, you’re no longer talking about eugenics, you’re talking about genetics. Genetics is not without its own ethical issues, but I think most scientists wouldn’t take issue with the field as a whole.

    • Elizaveta Markova says:

      I don’t think eugenics as Pearson understood it is inherently evil. I also acknowledge the non sequitur: statistics is associated with eugenics and eugenics with evil, but it does not follow that statistics was implicated in evil eugenics. It was, though, in fact.

  7. Anonymous says:

    And now quick reactions from Devil’s advocate:

    Gambling is a service in which you pay the House a commission. Yes, winning is important but people also enjoy the game, enough to compensate for the expected commission loss. The problem is not so much spectacular losses I suppose, but excessive consumption and associated commissions. Punters spend hours on slot machines.

    As for addiction, the whole country just went over the hill with people compulsively gambling their life savings on a housing lottery where, you guessed it, banks make a killing on commissions. Though here the expected return is perhaps positive I doubt this was the case at the top of the bubble.

    People compulsively consume a lot of things. Do we need a higher authority telling people no, you cannot buy that house, or invest in stock market, or buy any more stamps for your collection, or bet your life on an untrustworthy partner, or buy a diet plan, or an exercise machine advertised in a late night TV show that will not reduce your BMI, or that doughnut that may give you diabetes?

    Life is a gamble. There is an ideal in between Big Brother and laissez faire, and there is need for some regulation, but I do not see casinos being so special or evil.

    PS I doubt casinos in Vegas make their money from compulsive gamblers as presumably they are driven to pecuniary extinction, as it where.

    • Andrew says:

      Anon:

      I’m not saying that it’s evil to run a casino. I’m particularly talking about the methods they’ve developed at Caesars to lure back compulsive gamblers who are trying to stay away, as described in the This American Life story.

      • Anonymous says:

        But those are the same methods used by banks (e.g. Capital One) to market their mortgages, by Starbucks to market their sugar laden drinks, by some late night miracle diets, by Philip Morris to market cancer sticks, and so on for a range of industries serving customers’ willingness to trade off health, secure income whatever for something else.

        • Andrew says:

          Anon:

          Listen to that podcast. That lady wasn’t talking about tradeoffs. She was addicted and the statisticians at the casino were working hard to suck her back in. They weren’t just offering her discounts. They were calling her at home, repeatedly.

          Also, see the second-to-last paragraph of my post. The people in charge of this operation don’t seem to be proud of what they are doing either.

          • Anonymous says:

            I will listen to it but note that a lot of people are similarly targeted. People addicted to credit card debt receive dozens of mailings every week, and calls. Shoppers that buy junk food receive special coupons to buy more junk food,and there is evidence sugar and insulin can lead to addiction. And so on.

            One might say two wrongs don’t make a right, and I agree, but I would question the “wrongness” here. Taken to an extreme this line of thinking says smokers are victims of Philip Morris, obese people of Coca-Cola, all the way to criminals are the victims of society.

            That lady could have signed up to a program not to be called, receive mailings etc. The option is there. She chose not to.

          • Anonymous says:

            Listened to podcast. Agree with conclusion that the aggressive marketing was morally repugnant but not illegal. (Abortion is morally repugnant for many and not illegal in many circumstances, which is ok with me.) And Ceasars did have a program to allow her to sign out of all contact. She may not have been able to control her gambling whilst on the game floor, but presumably she could have signed out after a cold shower Monday morning, but I am no psychologist.

            Moreover, even if we accept the premiss that people may not be cognitively responsible for their own actions it does not immediately follow that their supplier should be the one responsible. What about her family, that we are told accompanied her on some of those trips, her local community, NGOs, the regulatory state?

            Finally, this is all relevant to your post because it is not clear that the quants were the ones telling salesmen how to pitch the sale, which is perhaps the most repugnant aspect (“come back to win”). For all we know they provided a database of high spending customers to which salesmen added comments and info.

          • Andrew says:

            Anon:

            You write that you think the aggressive marketing is morally repugnant and so do I, so we’re in agreement. The relevance to my post is that predictive analytics is often uncritically celebrated (as in the interview with the Caesars guy linked to above).

        • Andrew says:

          Anon:

          I agree that statistics are used to promote lots of bad ideas. The difference, as I see it here, is that nobody’s going to write an uncritical piece in a mainstream magazine on predictive analytics for cigarette companies or Nigerian spammers or whatever. I have no problem thinking of casinos as legitimate business enterprises that provide entertainment to people—but when they start targeting compulsive gamblers to come back for more, that’s another story.

          • John Mashey says:

            Cigarette mention raises a factual question:

            If someone is addicted to gambling, what’s the distribution of ages at which that started?
            I.e., is that an adult-responsibility issue, or is it like cigarettes, where a majority of adults addicted to nicotine had started by age 18 and where we have a reasonable idea of the distribution? (I.e., right skewed, peak ~ age 13-14 these days, earlier than the ~15-16 of 1984.)

    • Roger says:

      A lot of people were unhappy about the housing bubble, in case you haven’t notices. They are all in favor of laws, policies, and info to help prevent it from happening again. But at least buying a house provides a tangible benefit to the buyer. There is not so much benefit to someone playing slots for 15 hours.

      • Anonymous says:

        Roger has the point here – in that taking a loan twice the size of what you can afford still at least provides a roof over your head. Blowing life savings over 12 hours at the blackjack table, and taking out loans (courtesy of Ceasars) to go into debt gambling – is by any measures, without tangible benefit to the loser.

        Correct me if I’m wrong, but I remember in that episode the lady actually told the caller from Ceasars she had already blew her husband’s life insurance benefits away, that she can’t afford it anymore. When someone tells you they spent the entire 300K left from her now-deceased husband at your casino, Ceasar’s response should not be “Great! She’s a high-roller, how can we bring her back to our casino?” – but apparently some people feel that’s fine.

        It seems the problem is that statistics collected through rewards membership cannot fully distinguish a widow spending 100K of her savings away, or a wealthy woman blowing her ex-husband’s alimony away. And Ceasars doesn’t seem too eager to place some sort of alarm bell to their formula for obvious reasons. Money is money after all, right?

        • Anonymous says:

          The point about the home is a non sequitur: if you cannot pay the mortgage you gambled on you loose the house.

  8. Manoel Galdino says:

    I keep thinking this all the time. How much harm do I do in my job? How much predictive analytics is a force of evil? I hope most of time I do no harm and I surely think that applied statistician (even people like me, who work with statistics but don’t hold a degree in statistics) should have the same oath as physicians: First, do no harm.

  9. A. Zarkov says:

    Andrew,

    Have you thought about switching from mathematics to the clergy? Actually in this case, I agree 100% with your condemnation of Caesar’s for enticing a compulsive gambler back to the casino with telephone calls and offers of freebies. Nevertheless I detect an obsession on your part in condemning what you think are the transgressions of others. While I’m not a Christian, I like “Let he who is without sin cast the first stone.” Recalling this piece of biblical wisdom often shuts me up. Alas today most people would cast the first stone. As Irving Kristol reminded us in his 1988 essay,”Why I left” (NYC for DC):

    “Washington, D.C, is a city in the gutter, wallowing in hypocrisy. It has become a bizarre sinkhole of character assassination and smirking self-righteousness. It will eagerly cast not only the first stone, but any other rocks it can lay its hands on.”

    If you want to cast aspersions why not go where the action is?

    • Andrew says:

      Zarkov:

      Don’t blame me, blame This American Life. They have a circulation many orders of magnitude higher than mine! In the meantime, no, I don’t think we should reserve ethical discussions for the clergy. Ethics is important for all of us, and sometimes we can learn from the study of small things. To quote another bit of spiritual wisdom: God is in every leaf of every tree.

      • Anonymous says:

        I agree with Andrew and think he provides a very useful service in this blog.

        We all have our sins and pecadilloes but that should not stop us from having a discussion about morality. Quite the contrary, we can speak from experience!

        “Oh Lord make me chaste and continent, but not yet.” — St Augustine

    • I’m not a Christian, either, but I think you badly misinterpreted that bit of scripture. Self-criticism, generosity of spirit, and forgiveness of others are not at all the same thing as refusing to make moral judgments about the actions of others.

      Not only that, but “leave the moralizing to the clergy” is also about as far from the message in the Gospels as is possible. So, all around, I think you’re very confused.

      I agree with what I think is probably a kind of visceral dislike of judgmentalism on your part — I share it. But I think of “judgmentalism” as being un-self-critical and unforgiving, not merely making moral judgments about the actions of others. It is often, I think, making a moral judgment about someone’s character, not just behavior, and in a very harsh and ungenerous manner. But, otherwise, I think it’s important to make ethical and moral judgments about the behavior of others, but to do so rationally, responsibly, and with a kind of generosity and constant self-questioning of one’s own behavior.

  10. Nick says:

    It seems that there are a few things that are concerning in this situation: 1) statistics is being used to take someone’s money, 2) statistics is used to manipulate a person’s behavior in ways that the person doesn’t even know, and possibly can’t control, 3) statistics is being used to take advantage of sick people. These are all interrelated – the manipulation takes place to get someone’s money, and someone’s susceptibility to gambling addiction is probably more of a spectrum than a simple “sick/not sick” dichotomy.

    But it seems like a lot of big data/predictive analytics applications are being done at places like Google or Zynga and if the concern is that we are manipulating people with statistics then I’m not sure the amount of money at issue is the problem – it’s the manipulation itself. A/B testing as done by Target to get you to spend thousands of dollars on junk that you don’t really need doesn’t seem much morally different than A/B testing to get you to pay thousands of dollars to watch video slots. And yes, “stuff you don’t really need” is already a moral judgment but you could also say, “stuff whose resale worth is minimal” such that the effect on one’s net worth is the same whether you spend it on virtual cows, blackjack games, or clothes. Or predictive analytics in elections – is that “manipulation” OK? Applications to crime reduction?

    As we develop more and more powerful tools to make people do what we want, I think that the statistician has to take responsibility for the actions of the people they are manipulating. When the statistician builds the model, how much of the ultimate behavior of the manipulated individual is the responsibility of the statistician and how much remains with the individual? To some extent this sort of thing must have been thought about by advertisers, but it’s a lot different to say “I think that people will buy more laundry detergent if we use a cute bear” than it is to say “If we use specimen 12.c.3 we expect to sell an additional 17,000 units next year” – assuming, of course, that you are right.

    • +1

      Although I agree with Andrew’s outrage, it doesn’t seem obviously worse than using business analytics to convince consumers to purchase things (travel, cars, etc) that they don’t need. In the absence of illness, gambling is just a service that you don’t need (or don’t need much of). Of course illness makes it more complicated, but addiction to gambling is only quantitatively (ie, not qualitatively) worse than addiction to shopping.

      Full disclosure: I love gambling, especially craps. Despite understanding frequencies of dice outcomes.

  11. Jeremy Miles says:

    I like to be aware of my environment when riding (or walking) so I don’t listen to anything – doesn’t it make you nervous?

    But I do have a several hundred long backlog of TAL to catch up on, so I guess there’s pros and cons.

  12. Anonymous says:

    Statistics itself is neutral. Outcome depend on who is using it.

  13. Anonymous says:

    Sounds like universities peddling some masters programs with negative net present values on average, considering loan costs, opportunity costs, etc.
    Indeed, some law schools have been sued. Though I doubt these are driven by statistical analysis, but I’m suspicious of some online universities and professional schools..

    • Anonymous says:

      As a program coordinator for a masters program (not online, either) that, while cheap, is still unlikely to pay off much, I wonder this same thing.
      I counsel most applicants to my program (if they ever get in touch with me in advance) that there really isn’t much value to the degree, but some still come. Now, I myself have considered cooking classes and took a bartending course, so if the learning is worth our (cheap) tuition to them, that’s fine by me. And we do help the occasional student figure out that a PhD is right or wrong for them, which IS good to know.
      Yet, it still bugs me that our program continues to exist……

  14. t3 says:

    Your Coke example is weak and probably involves people in Guatemala that Coke as a company had little or nothing to do with.

    Coke would be much more like Caesars if Coke had sales and marketing teams that tried to put sugar drinks in schools all over the world to try get people addicted while they’re young and in a place where they have limited choices.

    • Michael Feltes says:

      Have you been in an American high school recently? Coca-Cola and Pepsico pay school districts considerable sums of money for exclusive contracts to put vending machines in cafeterias and provide scoreboards advertising their brands.

  15. Jonathan (a different one) says:

    I haven’t heard the podcast, but in my (perhaps outmoded) knowledge of casino analytics, they know what was wagered, but they don’t know (or care) what was won and lost. (The exception is for the largest gamblers, who can cut a side deal to have a portion of losses rebated. There they need to know.) There is clearly a statistical connection, but it isn’t necessarily the same thing. In general, casinos “reward” you for wagering, and let the net profits fall where they may.

    Thus, there are two possibilities: first, those who gamble the most may simply be those who enjoy the experience the most, either because they simply enjoy gambling despite their losses, or, improbably, because they are ahead, or nearly enough so. Second, there may be those who, despite a (present discounted value) lack of enjoyment, can’t stop. I don’t know how analytics would distinguish between the two.

    Finally, the addicted gambler is almost surely going to gamble somewhere. Much of the enticement may simply be an attempt to get the chump … er… customer to gamble at one’s own particular casino. Thus, the gambler’s money was already lost, the only question was to whom.

    • aidian says:

      they know what was wagered, but they don’t know (or care) what was won and lost.

      My understanding is that now, at least in the casinos that don’t pay out at the table/slot, they can track wins and losses.

      I don’t know how analytics would distinguish between the two.

      Assuming we consider gambling a potentially destructive activity and are willing to regulate it as such (which we currently do, at least in theory), then I’d be comfortable with a presumption that someone who lost at least x amount over y time has a problem and baring evidence to the contrary shouldn’t be allowed on the floor.

      the addicted gambler is almost surely going to gamble somewhere

      But, they were totally going to buy that crack somewhere. If I didn’t sell it to them someone else will. So why shouldn’t I make the money?

      • Anonymous says:

        I know people that have squandered their inheritance on whores and gambling, and have no regrets. Not addicted, not crazy, just choice. How can you tell a legitimate gambler from an addicted one by the magnitude of the loss?

        Some people give up everything they own to join a monastery on the gamble (for them certain) that there is an afterlife. Are they nuts?

        • Andrew says:

          Anon:

          The existence of other difficult cases does not negate an ethical problem. I do not think it is ethical to (a) lure compulsive gamblers back to the casino with strategically-timed phone calls and then (b) deny that you’re doing it (as I wrote above, Gary Loveman doesn’t say, Yeah, what of it? We’re fleecing the suckers, taking millions from rich addicts who don’t need the money anyway. Instead, he implausibly denies that they are systematically plucking people).

          • K? O'Rourke says:

            Its interesting and visible but surely not as bad as doing sloppy, misleading and perhaps even fraudulent research that claims vaccines are too dangerous to take.

            OK, it is hard to rule out “didn’t know better or truly understand the risk of being mislead” but I am aware of numerous cases where statisticians knew what they were doing was misleading but did for personal advantage.

            For instance, quoting a reference that was known to be wrong to get a “paper quickly by a journal reviewer” who wanted such a reference, advising medical researchers to use step-wise regression in observational studies to quickly get publications with minimal effort, deciding not to raise and address multiplicity issues with clinical researchers as other statisticians at the university don’t and it would make researchers less likely to work with you, providing misleading arguments in favour analysis methods for detecting harms in a safety analysis because they have much less power, refusing to release data from a study that has perhaps mistakenly changed how a certain life saving treatment is being used (securely to others ethically authorised to evaluate it given credible suspicions about the study have been documented) as it unclear if you can be forced to do this.

            There is lots of damage that researchers (including statisticians) do all the time for personal gain – just not as visible or “good story telling”.

          • Anonymous says:

            I agree completely difficult cases do not negate an ethical problem, but it turns what first appears like a slam dunk into a moral dilemma: to help some you restrict the freedom of others.

            Again, how do you tell a “compulsive gambler” from a “rational gambler” who has decided to loose everything. What is the right test? Do we need to put them all through an MRI scan? Its just not so simple to me.

            Besides, I actually think Gary Loveman was being sincere. From what I understand Vegas makes a lot of money as a family/convention destination. It is not good for them as a brand to be associated with exploiting people who appear compulsive. But the front-line sales people have incentives, and no set of incentives is perfect. Listening to the podcast I think some of them were going beyond the call of duty in amorally reprehensible way. Clearly they need to do better, and in this sense podcasts like TAL are good.

            But also listen to the original Money podcast quoted in TAL. There Loveman complains that the real compulsive gamblers avoid the Loyalty Cards precisely bc they want to go undetected. So it is very hard for management to keep tabs on all problem gamblers. But somehow that statement did not make it past the TAL editors, although I think it is relevant.

            And also, the whole podcast is based on the testament of one woman and a disgruntled ex employee: the law of small numbers?

          • Andrew says:

            Yes, I agree that it’s complicated. As a statistician, I was unhappy to see statistics blindly celebrated in the casino area without discussion of the morality. I agree that other moral lapses by statisticians can be worse.

        • Anonymous says:

          Sales have an incentive to meet quota and reach targets – but when a woman tells you she already emptied her savings, an you still entice her to spend more at your casino, I think most of us can agree you’re pushing the bound.

          Sure, you can go find an exception and scream about those who may lie, but what is it about casinos think some think is worth so much – that we should let them get away with ruining lives?

          P.S. Go do read up on how casino floors are ran, and you’ll realize they are all trained to spot and tag all customers. If they can recognize black-listed players walking in for the first time, I have no doubt they can recognize an addict coming back for the fiftieth time.

          • Anonymous says:

            So barmen should refuse drinks to drunks, fast food employees refuse fries to the obese, game console manufacturers refuse games to kids addicted and missing school, and so on

            This woman was smart enough to sue but not smart enough go sign out of mailings, calls etc….

            Btw before casinos Vegas was a dessert. Thousands make a living there.

  16. Anonymous says:

    Andrew:

    You are posting too fast. As a compulsive reader of your blog I am procrastinating way too much. Please limit your posts to one per day.

    • Anonymous says:

      What with your aggressive RSS feed and cross posting in sister blogs I might just have to sue you for lost work. I am not responsible for procrastinating.

  17. David says:

    “or maybe neutral (exotic financial instruments which serve no redeeming social value but presumably don’t do much harm)”

    That is the most ridiculously naive things I have ever seen. Exotic financial instruments are designed the way they are to be in the banks favor. The entire industry feeds off of the greed of unsophisticated investors who are willing to bet on instruments they don’t fully understand.

    • Emily Kennedy says:

      I like Andrew’s post, but I was definitely caught on this detail as well. I recommend Freefall by Joseph A. Stiglitz. It’s hard to read that book and say that exotic financial instruments are neutral.

  18. Anonymous says:

    “he better the statistics and the better the science, the worse the human outcomes. These guys are directly optimizing their ability to ruin some people’s lives.”

    Braithwaite has a great book on this:

    http://books.google.ca/books/about/Markets_in_Vice_Markets_in_Virtue.html?id=zrZxC6DnQhMC&redir_esc=y

    Here’s a sample gem: “this book is about how competition drives the more efficient production bads as well as goods.” He talks about obesity and tax avoidance as well as gambling.

  19. […] liked Andrew Gelman’s recent post (hat tip Suresh Naidu) about predatory modeling going on in casinos, specifically Caesars in Iowa. […]

  20. xi'an says:

    And can you also hear the cars with this new soundsystem?! Seems highly dangerous to me…

    • Andrew says:

      X:

      Yes, the cars are very loud! If we ever have fleets of electric cars on our streets I will be scared but this hasn’t happened yet.

  21. BD says:

    I’m reminded of the line from Rounders: “It’s immoral to let a sucker keep his money.” That’s basically what you want the Loveman defense to be, which would be pretty candid at least.

  22. […] You made the comment about how difficult it is to spot problem gamblers. I and a couple of colleagues [Paul Delfabbro and Daniel Kingjust] published this review of all the research done on spotting problem gamblers in online and offline gaming venues (attached) that I covered in one of my recent blogs. […]

  23. […] Gambling andrewgelman.com/her … ”There are four ways to get fired from Caesars…” on @StatModeling‘s blog […]

  24. David Duffy says:

    http://www.pc.gov.au/projects/inquiry/gambling-2009/report

    specifically

    http://www.pc.gov.au/__data/assets/pdf_file/0010/95707/24-appendixb.pdf

    show that 2-3% of all gambling machine users provide 35-40% of total spending.

  25. […] being at NIPS, which was held in a casino megaplex, Andrew Gelman’s post on casinos had more […]

  26. GC says:

    Andrew, I’d recommend reading all of Chapter 9 in Charles Duhigg’s book, on which that podcast segment was based. “Angie Bachmann” didn’t have much money when she started gambling. She went bankrupt in 2001, owing the casino $20,000. (I’d count that as “self-identification”, but I don’t run a casino.) This is after her parents had stopped loaning her money to pay gambling debts. Three years later, her parents died and left her $1 million. She then moved to Tennessee, where gambling is (or was) illegal, to try to stop.

    One a trip to her prior home town, she made the mistake of visiting the casino again. (It’s like an alcoholic who takes “just one sip”.) The manager recognized her and chatted her up. A million, you say? That’s when the heavy marketing started (free trips to Mississippi, etc). The casino sucked her dry again, knowing full well that she had gone broke before due to uncontrolled gambling. They even extended her a line of credit up to $125,000 after she had told them she had almost no money left, and knowing her complete history. Of course, they sued her when she couldn’t win enough to pay that back.

    Oh, and this time she gambled away a line of credit on her house, too, so she probably did end up homeless. I don’t know what the final terms of the settlement in the lawsuit were. Duhigg does say that she lost her countersuit.

    Bachmann knew she had a problem, but lacked the self-control necessary to overcome the casino’s ability to figure out what people’s hot buttons are.

  27. […] “This is the end result of all the bright lights, and the comp trips, and all the champagne, and free hotel suites, and all the broads and all the booze. It’s all been arranged just for us to get *your money*.” […]

  28. The Dark Avenger says:

    So barmen should refuse drinks to drunks

    In many jurisdictions, bartenders aren’t allowed to serve additional drinks to customers who show obvious signs of intoxication. That is not the case in Nevada, FWIW.

  29. TAL’s story about the gambling church members added a great and less-stereotypical human dimension to the Kevin Spacey Hollywoodised image of blackjack I had in my head.

  30. The catch with Indiana’s casinos is that they can’t be on *land*. So Evansville (river town) gets a casino on the Ohio River, which was intended. But casino owners are smarter in their own particular area than generalist state legislators (who hold other jobs as well). Like George Armor hiring ships to break ice across Lake Superior in 1909 or whatever, the value of building a magical money machine far exceeds the cost of hiring some diggers to make a fake lake for your casino “boat” to sit on. So in the middle of French Lick (historically a resort town–Al Capone is said to have vacationed there) now there sits a boat in a moat.

    http://en.wikipedia.org/wiki/List_of_casinos_in_Indiana#Casino_legalization_efforts

    Iowa doesn’t surprise me at all: it’s practically in the lap of Omaha, Sioux City, and bordered by Missouri and again not so far from Chicago.

  31. Via Richard Florida, avid hater of casinos, an economic impact study of the “social benefit” casinos brought to Atlantic City:

    “The typical customer of an urban casino is neither a tourist nor a deep-pocketed whale, but a local of modest means” http://su.pr/7kYuwx

    13 years after Atlantic City’s 1st casino…½ the population received public assistance…social problems…worse than ever http://su.pr/1daQa5