Dr. Mike Orkin is a Professor of Statistics Emeritus at California State University, where he was a professor and chair of the statistics department for many years before becoming a consultant and a nationally known authority on probability and gambling games. Since then he has appeared in numerous forms of media ranging from CBS Evening News, NBC’s Dateline, a Google Tech Talk series as well as authored serval books.
Episode Description
Gambling is a tricky topic. It got Pete Rose kicked out of baseball, and more recently, made news when Michigan State University announced a betting partnership with Ceasers entertainment. As with everything new media and big data have only complicated the conversations around games of skill vs. games of chance. That’s the focus of this episode of Stats+Stories with guest Dr. Mike Orkin.
+Full Transcript
Rosemary Pennington
Gambling is often a tricky topic. Hey, it got Pete Rose thrown out of baseball, and recently made news when Michigan State University announced a betting partnership with Caesars Entertainment. As with everything, new media technologies have only complicated the conversations and regulations we have around games of skill or games of chance. That's the focus of this episode of stats and stories where we explore the statistics behind the stories and the stories behind the statistics. I'm Rosemary Pennington. Stats and Stories is a production of Miami University's Department of Statistics and Media, Journalism and Film, as well as the American Statistical Association. Joining me is regular panelist John Bailer Chair of Miami statistics department. Our guest today is Dr. Mike Orkin. Orkin is a professor of Mathematics at Berkeley City College, and Professor of statistics emeritus at California State University, where he was Professor and Chair of the statistics department for many years before becoming a consultant and nationally known authority on probability and gambling games. Since then, he's appeared in numerous forms of media ranging from CBS Evening News to NBC Dateline, to a Google Tech Talk series. He's also the author of several books, and in November, Orkin, authored an article for chance, examining the intersection of games of chance, games of skill and New Tech. Thank you so much for joining us today. Mike. It's so nice to have you here.
Mike Orkin
Happy to be here.
Rosemary Pennington
I'm just gonna ask you to explain what's the difference between a game of skill and a game of chance that seems sort of crucial to understanding your chance article?
Mike Orkin
Well, there, there are a couple of different definitions. One is the definition that sort of has been in place by various regulatory agencies who don't, in the past haven't really liked gambling games online or in other forums in their states or federally, the there's this rule that's been around for years called a material degree of chance, whatever that means. It's never been clearly defined. But if a game has a material degree of chance, then it's not good for me or John or just a private citizen to start a gambling operation unless you're in Nevada and get a gaming license there. Now, this rule is slowly losing some of its impact because of a couple things, one, because there's so much money in it in fantasy gambling and other types of wagering, mainly sports wagering, that's really what we're talking about right now. But it spills over into other things. And so the federal government even sort of relaxed this powerful law about internet, sports wagering and now it's sort of left it up to the state instead of making it a federal problem for people who want to introduce fantasy sports and other sports types. Type of wagering. The big players right now are the fantasy sports companies FanDuel and DraftKings. But since the federal government relaxed its restrictions. You see some of the big gambling operations from Nevada, where all this everything in this area in this space is legal, such as Caesars, as you mentioned in the intro, rosemary, getting into the act, and in many states, you can you can make legal sports bets about the outcome of games online, which was not possible up until a year or two ago.
John Bailer
You know, I thought that, you know, in reading through some of your materials, you had that some discussion of the idea of, of the scale, a gradient of chance to skill. So could you talk about some of the anchors of that scale and some of the sports that are there? I mean, I think it's easy to anchor it, it's really hard to think about where you order it within that.
Mike Orkin
Yeah, the chance versus skill spectrum has been around for a while. It was invented by regulators who don't want to don't want to legalize gaming. And I actually have done some consulting over the last few years. And as I mentioned in my chance article, I testified in a case on the East Coast about illegal poker machines that had nothing to do with sports betting and had nothing to do with the Internet. But they asked me in court the judge asked me or the I forget was the judge or the the attorney representing the state asked me to draw what he called the chance skills spectrum, the chance skill spectrum and you start at, let's say roulette, which is all chance and end at some actual event game like bowling or golf, not, not the betting on it, but the actual event. And of course, even it's kind of silly because the word is Blackjack, rank and compare it to poker. And it's just hard to really understand what it means plus, even an actual game forgetting about gambling, even if you have to equal players in a golf game, or a bowling game, or a football game to equal teams, then the result is somewhat due to chance. And so there's a chance in every type of game, whether you're betting or not betting, whatever. So that so-called chance skill spectrum is kind of silly in place. So the other definition of chance versus skill is a statistical definition, which I used in a number of my consulting projects, which is just you look at player data, and see if some players exceed what luck will allow. So you can be lucky in a game of chance, or even a game that has skill. But you can measure statistically by doing the hypothesis test and a p value, which statisticians know about whether a player and a game is given enough games, whether their win percentage has gone above what the range of luck is, and that's easy enough to measure. And so if you haven't, have enough data, you can tell whether there is skill in that definition.
John Bailer
So just make sure that we're it's real clear to a listener here, the player that you're referring to, is not a competitor in some sporting event, but it's actually the person placing the wager?
Mike Orkin
That's correct so I mentioned a minute ago, you know, in the chant skill spectrum, yeah, I mentioned players being actual competitors in a sport. But now I'm not talking about that. I'm talking about competitors in a gambling game, whether they're playing against the house, or in some of these situations, you're playing against other players. But still, the question is the same as your win percentage, this is distilling it down to something that's easy to understand, is your win percentage better than what luck would allow. In other words, you know, people have diagnostic tests. Like in medicine, you get a blood test, if, if your measure of some characteristic of your blood is beyond some range, then it's weird. And this, you can do the same thing statistically, with luck versus skill. If it goes beyond that range, your wind percentage does. It's weird. And the weirdness can be explained by skill, or cheating.
Rosemary Pennington
That word is cheating, I like to so I, I'm going to open up as I love Blackjack, and poker. I'm bad at both. And I don't play for real money, but I love them. And I grew up playing it with my dad. And so as I and I are just unlucky, right. And so in the piece you wrote about Mega Millions, you talk about how luck is a group phenomenon. And I wondered if you could sort of talk me through that, to help me understand how my bad luck is part of something else? Or am I just weird, you know, somewhere in that weird spectrum?
Mike Orkin
Well, no. First of all, what I was talking about Mega Millions is the multi state lottery that everyone probably knows about. And the chance of winning the Mega Millions lottery, if you buy a single ticket is about one in 303 million. Which is a very difficult thing to do if there's no skill involved in lotteries, you can pick your favorite numbers, but it doesn't really have anything to do with your chance of winning. And so, in fact, the chance is so low. If you buy 50 tickets a week, you'll win the Mega Millions jackpot on the average of once every 116,000 years.
John Bailer
But it's my year, Mike, it's
Mike Orkin
But why are there winners, because there are winners. So I can give you all these analogies about how unlikely it is to win the Mega Millions jackpot. But there are winners fairly often. And the reason is, because so many people buy tickets. And that's actually part of a phenomenon called the law of very, very large numbers, which was sort of coined by mosteller and die connoisseurs to the status of well known statisticians. A number of years ago in a paper, sometimes also called the law of truly large numbers, as opposed to it was a famous statistical result called the law of large numbers. And the law of truly large numbers says, informally given enough opportunity, any weird thing will happen just due to chance. And so that applies here with the megamillions lottery, which is a game of pure chance. The opportunity is so many people buy tickets. So if you have enough tickets sold, in fact, when the jackpot gets high, there may be a couple 100 million tickets bought. There will be winners now. Now I'll answer Rosemary's question, when you have a lottery winner hidden in the background the chances were about one in 303 million to win. So hidden in the background are about almost 303 million losers. You don't just win it on your own. If I bought a single ticket, and nobody else bought tickets, and I won, then I would probably be cheating, because there isn't enough opportunity for me alone if nobody else is buying tickets. Now Rosemary, you were playing games of skill, Blackjack, and Poker. Blackjack is fairly clearly measured as a game of skill due to the work of Ed Thorpe, mathematician, math professor at UC Irvine. I think he's retired now. He wrote a book called Beat the Dealer and it was published in 1962, based on computer simulations, which is what mathematicians and statisticians use now for problems that are too complicated to do analytically. The dealer gave a specific blackjack strategy. So you could actually win in the long run. Most people probably like you, Rosemary, who don't want to spend the time to memorize and use Thorpe's theory, or the theories that came out or strategy or the strategies that came after that will probably, if you keep playing, you'll probably lose against a skilled player. So you're not weird. You just didn't take the time to learn a card counting strategy I, I would guess,
Rosemary Pennington
We're just playing for buttons. We're just playing for buttons.
Mike Orkin
So really, there's no reason for me to try to get better. And one button is worth 10 bucks, right? Yeah.
Rosemary Pennington
You're listening to stats and stories. And today, we're talking games of skilling games of chance with Mike working.
John Bailer
So now rosemary, I think we've just we've just found out what you should put on your wish list for your favorite book vendor, you know, now you can get your get your gaming read on, as you think about your travels,
Rosemary Pennington
I'm going to get better now that I know this book is out there, I'm gonna get up.
Mike Orkin
Thorpe's book, which was published in 1962, is still in print. And if you go to Amazon and look up blackjack books, you'll find a bunch of them. Thorpe's is the original one, and it still may be a little more competent, his strategy may be a little more complicated than some more recent ones. But they're all pretty much the same. You keep track of the high cards and the low cards as they come out of the deck and make decisions accordingly, using the fact that the dealer in blackjack has to play by if you're in a casino, the dealer has to play by a fixed strategy, they have to do certain things in you don't. So you can take advantage of that flexibility. And sometimes, if the deck is not shuffled after every hand, sometimes you'll encounter a positive expectation for that configuration of cards, which means a winning strategy, but you have to be very patient, and you have to be very good at keeping track of things.
John Bailer
You know, I did like the various analogies that you were using within your megamillions piece, as you were talking about kind of just the rarity of success here. I thought that was really nice now. Now most of the types of gaming that you're talking about in particular in your chance piece is more related to things that are more frequent. You know, so you're within the lottery setting, you're really living in the world of pure chance. Whereas if you're getting into these other fantasy games, you're dealing in a play. Ideally, there's some skill involved. So could you talk about some of the fantasy games that you were investigating as part of this chance piece?
Mike Orkin
Right. So the one that I was talking about in the chance article was a specific project that had to do with what are called parlay bets. In a parlay bet means you bet on more than one thing at the same time. There's more than one event that you bet on and all the events have to come out winners as far as your bed goes, in order for you to win the wager. Now, there's still some difficulty in actually betting on the outcomes of games. Whether the team wins or loses because of this long time fear of games being thrown by people who have enough money to pay off the players. Now it's kind of silly to major in like the NFL or major league baseball because the players make so much money. They do not risk their career by having some underworld character offer the money to throw a game, it's not that easy anyway. But the type of the type of wager or proposition that I was looking at in this particular in the chance piece was a two way parlay bet, which means you bet on two events. And these, this project involves events happening within a game. So if it's a basketball game, you could, you could say that Steph Curry makes the next three point shot and something else. And so that kind of wager, which in this project you're wagering against, in this sort of control situation, but the idea is if you win, so if you don't know anything about about it, or if there's just no Skillman this, that the point spreads or whatever the odds are, are, are good, or however, you're betting this. If there are two events, and each has a 5050 chance, if you don't know anything, then the chances 1/4 that you'll win the bet. Now that's, that depends on what the bet is, of course. So what you do is, you look at the player data. And if a player gets higher than 25%, win percentage over time, there has to be enough data, and it has to be at the whim percentage is higher, high enough to rule out chance, which you can't do in a lottery or in roulette, or in craps or in any game of pure chance, but you can do in a game of skill. But that's the challenge to see if there are key players who get a higher win percentage than they should, or then they would just by being lucky. And if there are such players, in this case, analyzing these parlay bets, if there are such players, then that shows that there are skilled players, because someone is not going to get that I have a win percentage, just by being lucky. And you can do that statistically.
John Bailer
So could you expand a little bit on the the outcomes, you gave one example of a particular player hitting a three point shot, which is certainly would be less than a half probably in terms of what you might expect, but But what are when you did this exploration, what were two of the outcomes you might have been?
Mike Orkin
Actually, when I mentioned something like Steph Curry making the next three point shot, there would be odds associated with that. So to try to make it be 5050. So the easiest explanation to probably most people would understand is the point spread system and football point spread system is a system where you add points to the underdog, to try to make the game a 5050 proposition. When you try to compute if someone is skilled, you would see if they're doing better than 5050 with the assumption that the point spread is fair, or accurate. Let's put it that way. So it was a similar thing with this project. And in this case, there are different versions to this some with which I've worked on different projects that are all involved the same kind of thing. And when it's a one point when it's player versus player, then once taking one side, once they enter, which was not the case here, then if neither players skill, then then it's a 5050 proposition also, that's another way to look at it.
Rosemary Pennington
Mike, I'm gonna ask Richard Campbell a question. So it's not related to the stats. But as I saw, I'm also a big football fan. I love NFL. And the thing that I've noticed the last several years is that there's been this increase in commercials for DraftKings and other kinds of gambling apps. Growing up I can't imagine having seen those kinds of commercials. Because it seemed like I remember Pete Rose being thrown out of baseball is like a really big sports memory for me. I'm from southern Ohio. My family were Reds fans, like it was a huge deal. And then now, you know, every Sunday, every football game like there's commercials and I wonder if this may not be something you could feel comfortable talking about. But it just feels like a Richard Campbell question, John. But I don't know. Do you think the fact that there are all these apps and all these things that have gamified gambling are these skills of chance like these various things? Do you think it's made it more socially acceptable to participate in, you know, loan sharks and like people who go to Vegas and smoke cigarettes and like, you know, play, you know, play craps? And like velvet suits. That was my idea. People who gambled as a child right?
Mike Orkin
Now, I have some good friends who are oddsmakers in Vegas. So none of them will wear velvet suits and they don't smoke.
Rosemary Pennington
Wow. But that's sort of like the general idea, right? They're kind of seedy people who are doing CD things that we think are socially unacceptable. And now, you know, every Sunday I see these commercials for DraftKings. And I wonder if you feel like technology is making it more accessible. Maybe it has made it more acceptable to like to participate in this?
Mike Orkin
Well, yeah, technology is part of it for sure. But there's another part of it too. And that's the government relaxing its regulation and online activities like FanDuel and DraftKings, which are not, in which you're not betting that they sort of got in, in a very clever way. Because they're not you when you when you do fantasy sports with FanDuel or DraftKings, you're not, you're not wagering on the outcome of any game, you're putting together your own team, and you hope your own team will amass enough points during the course of the season, so that you'll be the winner which is somehow more acceptable than betting on the outcome of a game. Now, but because the government relaxed the internet gambling laws FanDuel and DraftKings went through that little open door, and now dominate commercials and, and getting maybe not dominate, but they have spent millions and millions of dollars on commercials to try to make it more socially acceptable. So when you watch the FanDuel and DraftKings and now Caesars has Sportsbook as an ad to, you know, they want you to not think about the guys in the velvet suit smoking the cigar sitting or standing around the craps table and stuff like that. They don't want you to think of it that way. And so and by the way, Pete Rose would still not be able to bet on games if he were in the business, because if you're part of a professional sports organization, it is a strict taboo on gambling on games.
John Bailer
You know, I think it's been interesting to watch this evolution of fantasy leagues and fantasy sports, which is a consequence of this technology. But the interest that that generates in following more than just your local team is amazing. You know, all of a sudden, there are two teams playing that you have no connection to other than the fact that I have player x on this team and why I have the defense on the other team. So it's it changes kind of your interaction and your relationship with kind of league sport,
Mike Orkin
Where it's been a huge boon to professional sports, who used to say, oh, no, we don't want to do this because it blemishes the game or hurts the integrity of the game. But that's simply not true. And also now of course, as everyone can see, you have tons of money being poured into things like DraftKings and FanDuel, and lots and millions of people who are formerly not interested in betting in the sports space or making wagers. And so it's been good for the sport. And it also has sort of undermined the traditional bookie. An underworld character hangs out in the bars and you go and make your bed face. These people still exist, but they bet their profit margin has been cut into by the legality of betting. So it's actually helped the integrity of games rather than hurt the integrity of games to make some of these things legal.
John Bailer
Well, so Mike, Rosemary can still hook you up if you need that kind of connection.
Rosemary Pennington
That's right. I know all the seedy people writing you John.
Mike Orkin
Well, actually, I was living in the Berkeley area. And there used to be a restaurant in Berkeley called China's station, which is down by the train tracks and South Berkeley and there was a sports bar there. So a lot of people went and bet there this was before any of it was legal. And there was this one guy, the bookie, the bookie, the main bookie, they sort of controlled this work, but it was known as the plumber. And so I thought oh my god, you know, what if this guy flushes people down the toilet or pay sticks their head in the front, I called the plumber. But I used to go down there with some friends and we'd every now and then make a bet on a game and just an interesting scene at this restaurant and at the sports bar next, adjoining the restaurant. And so one night after I got to know this guy a little bit, I asked him, Why do they call you the plumber? And he said, Because I am a plumber. I'm a plumber during the day and I come here at night and I'm a bookie not nearly as threatening to know the backstory is.
John Bailer
So Mike, I wasn't much less ominous. Yeah, much less. I got a quick answer. I want to follow up on what you had talked about in a talk that you had given about some data mining strategies to possibly identify some wagers that might be beneficial. So you know, if your team lost last week in there playing another team that didn't cover the point spread then bet on them to lose you know that or not to cover the spread of that idea of exploring this incredible space of options and alternatives and then maybe seeing if there's nuggets there. You know Has that proven to be useful? Is that something that people have benefited from?
Mike Orkin
Well, it's that sort of a, there's not a clear answer to that. I'm a friend. And I developed some software like that a number of years ago, for pro football and pro basketball, we did the exactly the same data mining types of things that you just mentioned. So you would look and see if a team and it was all about covering the point spread, if a team covers the spread, if they lost their last two games and are playing at home, stuff like that. So there may have been a few things up. So the software would actually look at WIN percentages as in concert with enough games to look at z scores for the win percentages, assuming a 5050 chance that will end that's a statistical way to measure how strong something is. And so, this offer would automatically go through the sift through the data, truly data mining. At night, we even had a part of it called the night shift, which would look all night at your or whenever you want to, and find high Z score results, and would find some amazing things. But the problem is when you do things like that, you'll also uncover lots of random results that don't have any in terms of betting. And so most of them, so we did a little study, and most of these results are sort of random, but there are some factors people look at like whether a team is due to win or whether a team is hot, and when again, sort of opposite strategies and what the weather is like. So, this friend of mine, named Roxy Roxborough, used to be the head oddsmaker in Las Vegas for many years. He said, It's a great name. If you look him up, you'll see he was a very well known character. So they would keep track of everything because they set the point spreads. So there was this guy, this gambler, who would call them from time to time asking about certain things. So one week, he called them. So this is one week, he called them and asked, this is on a Sunday morning just before the games, and he called Roxy who's in Nevada in Las Vegas and said, What's the weather like in Chicago? Because I'm thinking of betting on the bears, and rocks. He said, where are you? And the guy said, I'm in Chicago. And he said, Why don't you open the window? Because he's in Las Vegas. But anyway, some people pay a lot of attention to these sorts of side factors.
Rosemary Pennington
Man, if you are a Browns fan waiting for the Browns to win because they do you're going to wait a long time. I speak as a lifelong Browns fan, right?
Mike Orkin
So there's this there's this well known phenomenon called the gambler's fallacy, which states that roughly speaking tense, the law of averages says things will average out in the long run. If you see a lot of let's say losses by the Browns, then just random stuff, then they're due to win. But of course, that's completely false. And that's why it's called the gambler's fallacy. So, I mean, there may be some factors in skill games, like in football or basketball that do influence what's going to happen. But I haven't, I used to know a lot of people who were sort of technical gamblers on sports in the Las Vegas world. And some of them won a little bit, but I don't know anybody who made a killing, that they mainly made money by selling tips to games they were touts, which means they sell their picks for the games or tips on who would win. That's how they made their money.
Rosemary Pennington
Stats and Stories is a partnership between Miami University’s Departments of Statistics, and Media, Journalism and Film, and the American Statistical Association. You can follow us on Twitter, Apple podcasts, or other places you can find podcasts. If you’d like to share your thoughts on the program send your email to statsandstories@miamioh.edu or check us out at statsandstories.net, and be sure to listen for future editions of Stats and Stories, where we discuss the statistics behind the stories and the stories behind the statistics.