Secretly Incredibly Fascinating - Random Numbers

Episode Date: December 28, 2020

Alex Schmidt is joined by bestselling author Jason Pargin (‘John Dies At The End’ series, ‘Zoey Ashe’ series) for a look at why random numbers are secretly incredibly fascinating. Visit http://sifpod.fun/ for research sources, handy links, and this week's bonus episode.

Hey, it's me, Alex, and I have a tip for you about the Patreon page for this podcast. We are approaching our second ever group goal on Patreon. The first one was the mini membership drive. I don't know if you remember that, but I sent out stickers and I wrote cards and you guys blew out the goal for that. That was amazing. If we reach the next Patreon benchmark, there will be a large enough group of SifPod supporters for me to begin offering democracy.

Once a few more of you back the show, I will start taking episode topic suggestions from listeners on Patreon, and then we'll have a monthly vote on topics that are suggested that I think are particularly promising, and then the winner becomes an episode. So if you've been enjoying the podcast but haven't begun supporting it yet, have just been kind of cruising along listening to the main shows, this is a great time to start. So do that thing. Support the show by visiting sifpod.fun. That takes you to the Patreon page where you can get an immediate bundle of tons of supporter benefits, including a library of almost two dozen bonus shows, just right there.

And if enough of you step up and join in, all supporters will gain a whole new benefit. You'll gain a voice in what these podcasts are about. So let's do that. Let's make 2021 a democratic version of this podcast. And in the meantime, here's a new episode. Random numbers. Known for being random. Famous for what I just said. Nobody thinks much about them, so let's have some fun. Let's find out why random numbers are secretly incredibly fascinating. Hey there, folks. Welcome to a whole new podcast episode. A podcast all about why being alive is more interesting than people think it is.

My name is Alex Schmidt, and I'm not alone. Jason Pargin is my guest today, my former colleague, my old pal, one of my favorite authors, and so much more. I hope you know about his latest novel. It's entitled Zoe Punches the Future in the D**k. It's funny, fun, full of big ideas, a great use of that gift card you just got for getting books if you have one. And Jason is a full-time author.

More books are on the way. I'm always grateful to him for making time for guesting on podcasts like this one, because he has a lot of writing to do. That's what a full-time author does. Also, I've gathered all of our zip codes and used internet resources like native-land.ca to acknowledge that I recorded this on the traditional land of the Catawba, Eno, and Shikori peoples. To acknowledge Jason recorded this on the traditional land of the Shawnee, Eastern Cherokee, and Sa'atsoyaha peoples. And acknowledge that in all of our locations,

native people are very much still here. That feels worth doing on each episode. And today's episode is about random numbers, and we will explain why immediately. So please sit back or try to pick a number between 1 and 10 without picking 7. Either way, here's this episode of Secretly Incredibly Fascinating with Jason Pargin. I'll be back after we wrap up. Talk to you then. Jason, happy almost the new year. Thanks for doing this as always. Yeah, I've got to kick this one off by warning people. This is a weirder episode than what it may it may appear on the tin so i don't know if if in the middle of the holiday season you're prepared for something that that goes to kind of a weird place but i'm not going to say it takes the crown for the weirdest episode of your show versus what it appears to be from the title but i'm going to guess it's top five

yeah that's probably that's probably, that's probably worth saying because yeah, I'm so excited about this topic that, uh, that you originally suggested because it's, it's such a particular thing that I think people never think about. Yeah. And it's one of the weird fascinations I've had for years and something that I can never properly explain to anybody why it's interesting. So now it is your problem. And you mentioned you've been interested in this for years. I guess that's part of the answer to what your relationship to the topic is or opinion of it. But anything else there? Let's get into it.

Probably my most everyday encounter with random numbers or generating random numbers is in video games. Since I'm not a gambling man, video games fills that hole in my life. element of random number generation, whether you're aware of it or not, if you're playing a role-playing game and there's a particular enemy that you're trying to get a certain weapon and you know it's a rare drop and you know that somewhere hidden deep in the code of the game, only one out of every 150 of these monsters drops the sword, well, you are at the mercy of a random number generator. It has just determined that, you know, it's basically picking from a pool of one out of 150 instances. And you're trying to get that one. If you have played Animal Crossing, as I have, and you're trying to fish for one of the rare fish and you've got to get it before winter ends in the game because it's seasonal. Yeah. And you know, from looking it up, that only it only has a one in a thousand chance of spawning. You are basically at the mercy of somewhere deep in the code, a random number generating algorithm.

And the problem with me saying, well, it's just a random number, is that it turns out that trying to get a computer to generate a truly random number is incredibly difficult. And the reason it's difficult has to do with the nature of the universe and existence itself. I'm so excited about this stuff. Because I never thought about games that way either. I also, gambling makes me mostly sweaty. I really don't enjoy it that much. With gaming, like I've been messing around with a mobile game called Retro Bowl, where it's a little football game and they're little pixelated players. And every time

a guy fumbles, I'm like, this is the worst game in the world. Because it's clearly just that tackle led to a fumble in the random number generator. There's no other reason for a fumble. And it feels deeply unfair, even though it is realistic that guys would fumble once in a while. That's not unfair at all. In any mobile game like Candy Crush, you think of that as somewhat of a game of skill because you do have to make decisions. But you are at the mercy of a random number generator that determines which style of candy is going to drop in which row. And ultimately you can wind up with a map that's unwinnable because it just randomly gave you a bad fall.

But as with any, with any randomness, you can fool yourself into thinking the same thing that gamblers think, which is I'm on a hot streak or an animal cross. He's like, well, the last time I caught one of these fish, it was up here by the pier. I bet that's where they, they live. It's like, well, no, the game, they spawned, they spawned somewhere on the map. The game is, but you want to, you want to believe there's something more to it. You want to believe you're lucky or that you, you pick the right spot or you're on a cold streak or a hot streak and you want to project some meaning onto the randomness. Yeah. But it's

the same thing, you know, if, and if it sounds simple, you know, cause you think, well, like playing a bingo, like my grandma plays bingo at the church and all, they just got a little drum, they spin around and they pull the ball out with the letter and the number on it. And that's how they determine it. Trying to get a computer to do the equivalent of the drum full of ping pong balls is unbelievably complicated. Right. It should be easier for a computer to do what that stupid physical glass ball, not even glass, it's plastic. It should be easier for an amazing machine to do that. But we'll talk about why that is not an easy thing to do at all. I also am very excited that on every episode, our first fascinating thing about the topic is a quick

set of fascinating numbers and statistics. I feel like that's kind of recursive with this topic, but there's fun ones here. That's going to come in a segment called, with this topic, but there's fun ones here. That's going to come in a segment called Should Auld Acquaint Stats Be Forgot and Numbers Brought to Mind? Should Auld Acquaint Stats Be Forgot and Days of Auld Tansyne? And that name was submitted by Dan Starton. I especially appreciate the many mathematical terms there. And there's a new name for this segment every week. Submit to SifPod on Twitter or to SifPod at gmail.com.

Make them silly, wacky, and bad. But almost Happy New Year. We almost got there. It's great. You're only 20-some episodes into this show, and your stats segment names are now getting so convoluted and arcane. I have no idea where you're going to be on episode 250 because these, these got convoluted fast. Yeah. Yeah. People really stepped up. I think, I guess I'm hoping there's like a live band or fireworks or something down the line. Like, yeah, we're going to need to escalate somehow. Yeah. Well, I know for a fact you had to rehearse

that before you were on mic. I know you did. I have a worse answer, which is that I did not need to rehearse it because I'm pretty sure we've sung that song back in high school choir. And so then I did not need to prepare for just the worst reason. Yeah. In fact, the nerdy math version you probably sang in. Oh, yeah. Right. A collab with fourth period calculus. Yeah. So we have numbers here about the world of random numbers. The first set here is a set of numbers about Mega Millions and about Powerball, which are the two large U.S. lotteries. And there are similar lotteries in other countries where people listen to the show. But these are the odds for it.

First number is 1 in 302,575,350. 1 out of over 302 million is the official odds of winning the Mega Millions jackpot. That comes from their website. Similar number here is one in 292,201,338. So more than 292 million versus one is your odds of winning a Powerball jackpot, according to CNBC and Slate. They're pretty similar. They're within about 10 million of each other, which in this case is not a huge difference. Right. And both of them take

advantage of the fact that the human brain cannot comprehend that number at all. Right. So you might as well be, you might as well say 40 bajillion. Like it's just, the lottery makes all that money based on the fact that human brains can't comprehend how slim their odds are. If you ask somebody, why do you buy a ticket every week? It's like, somebody's got to win. Like, okay. And a meteorite has to fall on somebody and kill them at some point, too. But you don't walk around expecting that.

You don't plan your life around that happening. But anyway, that's another subject. Right, right. And it is, I was sort of excited to say those long numbers without needing to like go back and clarify them much, because I know no one can process them. They're in the hundreds of millions and it's meaningless. Yeah. If you added like three more digits to the end, if you just kept saying digits, not a single person in the listenership would have said, man, that doesn't sound right. If you said 300 trillion instead of 300 million or 300 billion,

it's like, yeah, it's just, it's a big number. There's a point where if you close your eyes and try to picture 300 million apples, you're not, I guarantee you, you're not doing it. It's hard to picture like 150. I think, I think I could have borderline thrown in a letter and nobody would have blinked. Yeah. And both these systems also, I feel like the most interesting thing about them is not only that those odds were kind of similar within the hundreds of millions we're talking about, but also the Powerball and Mega Millions in the U.S. are marketed as separate games, but since 2009, the two games merged jurisdictions and have become almost exactly the same. According to Slate, they both sell $2 tickets. They both sell a $1 multiplier if you want that. They also both do a drawing of five white balls and then one special color ball for the numbers for the prize. And then also the top jackpots ever, the two biggest US lottery jackpots are a Powerball win

in 2016 that paid $1.586 billion, and then a Mega Millions win in 2018 that paid $1.537 billion. So again, massive numbers, but within tens of millions of each other. And the difference in the odds is because in Mega Millions, they're drawing the five balls from a set of 70 and the special ball from a set of 25. And then Powerball changes both those numbers by one. It's 69 balls for the five set, and then 26 balls for the one. So they have taken these two, I believe, separate games initially and just made them into almost exactly the same game. But people do like a Coke or Pepsi with which one they buy. Right, because they perceive one as being luckier than the other or whatever.

But both of them are just attacks on poor people. It's attacks on people who don't comprehend. They're preying on people either have a gambling addiction or just preying on people who want some kind of hope in their life yeah like you buy that ticket and you can spend the next few days hoping that you're going to be rich because you have no other avenue available to you and it's like it's kind of like the cruelest text someone could come up with again not the subject of this show today but that's my feelings on the lottery are i know i've known some gambling addicts in my life and it's uh that the state sponsors this is like a way to fund schools that it's always bothered me a lot yeah i i again gambling makes me sweaty. I've never had an interest in doing it, but it does seem to just prey on people. And then a lottery winnings related number here,

it's $344.6 million. $344.6 million is the amount won by a Powerball player who played the numbers from a fortune cookie opened by his granddaughter at a Vietnamese restaurant. That's coming from Vice. It was June 2019. A man named Charles W. Jackson Jr. in Cumberland County, North Carolina, won the largest ever lottery jackpot in the entire state of North Carolina's history by just taking the numbers from his granddaughter's fortune cookie, playing them, and winning. And the article says that this kind of thing happens a lot. Somebody won a million dollars in New Jersey in 2018. Somebody won $125,000 in Vancouver, British Columbia in 2018. There's also been a $10 million win in Florida in 2015. And then one case where the fortune cookie company, Wonton Foods Incorporated, printed

one fortune cookie slip that 110 different players played, and they all won between $100,000 and $500,000 in 2005, because the one slip lined up with the one that was played in New York. And this is some nice foreshadowing for what we're about to get into. Because if someone asked me, like they're going to go buy a lottery ticket and they asked me to suggest a number for them to play. And I said, play 000000000000, they would think I was being sarcastic because their answer would be, well, what are the odds that the lottery draw is going to be all zeros?

Because they can't comprehend that the odds are exactly the same as anything else they played. If you played 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, like it's, it's the exactly same. If you won the lottery and then you decided to keep playing the lottery and the next week you just played the same numbers, some friend would surely say to you, well, that's dumb. What are the odds? It's going to be the same number twice in a row. It's like, it's the same odds as anything. It's the same odds as it being any numbers the entire game is predicated on the fact that you can't process that so when you get a story like the fortune cookie thing where like there's a hundred different winners because you always get somebody

claiming that it's a sign there's got to be fraud or was arranged or something like that because they can't it's like well what are the odds that that fortune cookie would match up with and it's like there's no odds of any of this happening right when you hear lottery most of you listening to this probably think of that big machine where it's a bunch of ping pong balls blowing around with like compressed air and then one flies up to the top and that's how they pick it but in a bunch of u.S. states, and it's actually kind of hard to figure out which ones, I think it's 17, they don't have that machine. They use a piece of software that is a random number generator, the same kind of software algorithm

that determines whether or not you're going to catch the great white shark in Animal Crossing or get another sea bass. It's one of those. And there are states that have shut down their lottery game because they got results out of the random number generator that they started to think were not random. And so obviously people lose faith in the system and think, well, it's not truly random or it's been rigged or whatever, then that ruins the whole game because it has to be random for everyone to believe they have an equal chance of winning. But when you're using a piece of software, trying to figure out if what it's spitting

out is truly random or if there's a pattern that a smart person could discern is unbelievably difficult and makes people very, very angry to the point that they wound up just shutting down the games in some cases, because it is such a difficult puzzle to solve that they just didn't because the possibility that in fact it was, you know, predetermined what number it was going to spit out, or it's, there's some flaw deep inside there so that it repeats the number every x number of sequences it's such a complicated engineering problem that it is one of the big issues with running any kind of game either a casino game or a lottery lottery or anything else yeah and it fits into that next number here which uh which picked out, which is amazing about another machine doing the same kind of thing.

Yeah, the number is $270 million. That's how much money is earned in a single month by gambling machines on the Vegas Strip. When I say gambling machines, I'm talking about slot machines, video poker. poker. If you've not been to Vegas or if you've not been to a casino recently, if you're thinking of a slot machine as still this physical thing where you pull the lever and there's like three mechanical wheels that spin around and randomly hit on a button and a bunch of coins spit out, that's not how it works anymore. Slot machines are now just, it's a screen. It's just a video game. It's the same as the mobile game on your phone. All of them are simply random number generators with fancy graphics to make it look like something else. So there will be a graphic image of the little

number spinning around, the three or five or however many, and they have programmed it so that when it lands on a loser, it will look like you almost won because you can see like the number scroll up off the screen and you think, oh, the wheel stopped just one slot. That's there to trick you. In reality, the random number generator spat out loser, but it displayed, you know, cherry, cherry, and oh, I got the turd or whatever. I don't know what the symbols are on a slot machine that means you lost, but you saw that third cherry go right up off the screen. It's like, oh, I just missed it.

You did not just miss it. It is a random number generator. Two hundred, a quarter of a billion dollars a month they make. This is an estimate off of these software driven gambling machines that are simply using a piece of software to give you a randomized winner. that are simply using a piece of software to give you a randomized winner. And that's how much money it extracts from people who put money into these machines, basically waiting for a piece of software to tell them they won. Even though the casino is not going to allow itself to lose money.

So they have specifically programmed this for every x amount of dollars put into the machine to award x percentage back to the players and they simply adjust this and casinos actually compete against each other because they will actually edge their payouts up by one percent and then boast that they have like the loosest slots or whatever, the most generous slots around. And it's actually like this predetermined thing, which should take the fun out of it. But it doesn't because, again, this is a quarter billion dollar tax on people who can't process the fact that there's no such thing as being good at a slot machine. Or there's no such thing as I'm sitting, I've been sitting at this machine for six hours. It's due to pay out. It's like, man, it's just as likely that the moment you get up, that next person's going to sit down and they would win what you think of as your money. But in reality, it's just an algorithm that is determined every X number of polls. It's

going to, it's going to spit out a winner. Yeah. And it almost feels to me like another non-random element is how much we as a human population are into it. Individuals vary, but it seems like whether it's Nevada casinos or national and state lotteries, we as a population will play as much of it as we are allowed to play. There no other like buffer on it it's very strange well all you have to do is look at what video games do now with what they talk about loot boxes where you pay real life money in these multiplayer games and instead of paying for a gun or an outfit for your character you pay for a box a mystery box and it will pop up and you pay the same amount of money but when the box opens you

may get a rare gun you may get a completely lame skin for your character something's totally worthless but it is based on gambling mechanics and the gambling addiction that's based on getting you to put more money in thinking you're going to get lucky and get an item that in reality has no monetary value but it's based on that same personality flaw. Whereas if they just sold these straight up and it's like, it's $5 for this gun. They would not sell as many as they do by playing it off that thrill of the next one's going to hit. The next one's going to hit. Wow.

Yeah, I have heard of loot boxes and have not experienced them much. But I have heard of them being described as, oh, they found a loophole where teenagers can gamble now. And it sounds like it. Yeah. Yeah. And the reason that's not legally considered gambling is because you don't win anything. If one of the prizes, well, you want a bunch of cash, that would make it illegal. would make it illegal but the the fact that you're winning a completely valueless digital pair of pants for your character means it's completely legal but the fact that you can't win is what

makes it okay it it you it only takes money in so it would be like if you were going to a slot machine in vegas instead of giving you cash it just displayed a picture of cash that you could use to buy like a digital T-shirt that you can't wear, but it displays on your screen. Like if someone opened an actual casino that did that, the people would burn it down. But the video games have quietly slipped into doing that. And this is these are players that are 10, 11, 12 years old who don't even know what gambling is or don't even know what these impulses are. But it's that same thing that, you know, that you talk about like the Skinner box experiments where they found that if you put, you know, a rat in a cage and have a lever that if it hits it, it randomly spits out food instead of one lever spitting out the food every time. It's like randomized by giving out those randomized rewards the rat will just sit there and press that lever all day long thinking

the next one's going to be the lucky one man the entire loot box situation makes me feel very old-fashioned because the other the the other thing we've talked about that makes me feel that way is these video slot machines. The most recent time I was in Las Vegas was last year. And it was because my friend was getting married. And we were like, okay, what do you want to do because you're getting married? And he was like, I can't wait to play a specific brand of video slot machine. That is my favorite thing in all of Las Vegas. And I watched it trying to figure out what the fun part is. And we ended up going and playing roulette because I was like, at least that's tactile.

At least that's analog in a way that's like slightly more interesting to me. Great, like let's do that instead. Yeah, and we'll talk about that because the physical machines like roulette wheels, things like that, they have to be like very perfectly calibrated. You know, there's obviously if you buy a roulette wheel that is tilted to one side

so that people know it's always going to fall down to the left, like you've got a jacked up roulette wheel, like it's got to be perfectly balanced. The ball's got to be perfectly balanced. If you bring your own set of dice to a craps game, they usually get mad because they want you to use their dice because in order to get dice to give you a truly randomized result, they've got to be perfectly balanced, perfectly made. You know, if you're fishy about the way you throw them so they're not tumbling over, they'll they'll get mad at you. There's all these things, these physical things that give you a random result that in like in the example of the lottery machines with ping pong balls, like they have

experts who make sure those ping pong balls all weigh exactly the same, that they're all perfectly round. Cause if one of them is odd, if it doesn't weigh as much, it's like, well, technically that could make it more likely to zip to the top. They also switch the balls out so that you don't have the same numbers on the same balls every time. Like every, you know, all these things to ensure that it's a truly random result, even with something like that, is difficult. Because again, there's money at stake. And do you want to give people this next number here?

Because I feel like it is truly random and then they make the money off the game rules outside of it. Sure. If you have a deck of 52 playing cards and you shuffle it, ask yourself how many possible combinations are there? Because like a child will say, well, there's probably 52. And it's like, well, no, there's actually multipliers of multiple cards can be out of order. And if you say well it's probably i don't know like 10 000 the answer is an eight with 67 zeros behind it yeah there are more there are more possible combinations

in a deck of 52 playing cards than there are atoms on earth to put that into perspective we're going to link this this is a direct quote from the article that's got the source to put that into perspective, and we're going to link this. This is a direct quote from the article that's got the source to put that in perspective. If someone could rearrange a deck of cards, every second of the universe's total existence, the universe would end before you got one billionth of the way to finding a repeat. I love it. In other words, if you grab a deck of cards out of a drawer and you shuffle them and you got, you, you do have to shuffle them. Like you, like, you know, they, you don't just get them fresh out of the pack. Congratulations. You are holding a deck that has

never existed in the history of the universe. It is virtually impossible for anyone anywhere in the history of playing cards to have ever had a deck shuffled in the exact order you just created. It's so, like I'm sitting here looking visually at this number written out, which again is 8 times 10 to the 67th power, or 8 with 67 zeros after it. And the number is long in a way that makes me angry. There's too many commas, There's too many digits. I don't like it. But that's what a single deck of cards is in terms of its potential configurations. It's amazing. Yes. And this is why shuffling a deck of cards, that is a random number generator. It is one that you're doing with your hands. And it is one that is almost impossibly complex. with your hands, and it is one that is almost impossibly complex.

Yeah, and you can just go buy a deck of cards and do it at home, and you're kind of more powerful than a computer that way. Powerful being a very loose word for it, but it's amazing. Yeah. Last number here, I picked this one out because I love how deeply not random it is. The number here is 10,000, And 10,000 is the number of possible pin numbers in a four-digit system, personal identification number. Because I think not everybody knows this top of their mind, but once you think about it, it's pretty obvious. The choices are four zeros through four nines. And so there are 10,000 different options when you're picking a pin number. And if you read about pin numbers, they are a huge security vulnerability because we all choose our pin numbers and we tend to be pretty lazy about it or else pick something very easy to figure out. Yes.

And that's going to, again, foreshadow more weirdness. Because if you say, well, why would you say it's hard for a computer to come up with a random number i can do it right now six six four four you actually cannot come up with a random number if you look at the distribution of people's pin numbers you will find actually no there's a very small range of numbers that people pick because they don't they want one that they can remember. And any magician knows if you ask someone to pick a number between 1 and 10, there's only a few choices everyone picks. If you ask people to pick between 1 and 3, most people say 3.

Wow. Yeah. As for why, often you don't know. It's something in your brain. Yeah. It's actually much—if you're trying to guess a pin number, if you know the person, you've got a better than one in 10,000 chance of doing it probably. Yeah, because there's a few common mistakes people make. And a lot of this is coming from Snopes, also from electronicsweekly.com.

But one mistake is people just do a lazy number so they don't have to do labor to remember it all the time. A German university study found that the most common four-digit pins in its sample were 1234, 0000, the next one was 2580, which sounds advanced until you remember it's just going down the keypad with your finger from top to bottom. Exactly. going down the keypad with your finger from top to bottom. Exactly. And then also 1111 and 5555 were the other things people tended to choose,

which are all just dumb. They're not very good pin numbers. The other common mistakes are somebody picking a relatively advanced pin number but using it for absolutely everything they ever do. So that's all it takes to get into somebody's stuff. And the other one is people using a pin number that's based on something easy to discover. Like if your pin number is your birthday, someone can find out your birthday relatively easily, at least compared to the task of cracking one of the 10,000 options for a pin number. It's much easier to do that. Yeah. And even the 10,000, that is so much infinitely smaller than the number of

playing cards thing that your brain just breaks trying to comprehend it. Yeah. It's really, it's just such a small amount. And then also the other thing this study has found, because like in my life, I have an iPhone. And when I got a new iPhone, they made me pick a six-digit pin number, which immediately felt like too much labor. I was very furious at Apple. But the same study found that a six-digit pin is not much safer than a four-digit pin because people just extend the laziness. And I won't tell you my pin number, but I did that.

You just pick something that is equally simple or easy to remember or easy to figure out for two more numbers. It's the same security flaw. Right. And the reason that phone numbers are seven digits is that is about the length that the average person can remember. You get an 8, 9, 10, and they can't. They've got to go. There's some people that have to go like, look at their social security card every time they're asked for that number, because it's like, no, it's too many. It's too many numbers. Around seven is about what the average person can be reliably memorized.

Wow. Yeah. I mean, it, it, I was going to say it makes sense, but it just simply tracks with what I've experienced. I don't otherwise know why that's the way we are. I guess, I was going to say it makes sense, but it just simply tracks with what I've experienced. I don't otherwise know why that's the way we are. I guess we just don't want to do the additional labor for this stuff. Right. And because you've got a whole bunch of passwords in your life, your PIN's not the only, you know, like so many, like the reason they use the last four digits of your social security number. It's because like, look, if you had to have a unique pin for every banking website or whatever you would just straight up write it down somewhere and at that point somebody they just have to find your post-it note now they have access to everything in your life so yeah they have to

keep it simple yeah it's also there's one more thing here about how little labor we want to do with pins which is uh Snopes debunked an interesting... They debunked it because it was an urban legend on the internet, but then also people heard the legend and tried to implement it. But in the early 2000s, people started to believe in urban legend where if you're at an ATM and you enter yourIN number backwards, that will withdraw money and summon the police. And the idea is it's like a silent alarm if you're getting robbed by someone at gunpoint at an ATM. So you say, oh, yeah, I'm complying. Here's some money. And meanwhile, the police come and catch the person and work it out.

But for many reasons, that's not a thing. Like, if nothing else, the police would just not get there fast enough. But then from there, people heard about this urban legend and state legislatures in Illinois, Georgia, and Kansas considered bills to implement it. Also a Chicago businessman named Joseph Zinger, like invented what the technology would be. He called it safety pin, which is cute, be, he called it Safety Pin, which is cute, patented it in 1998 and tried to make it a thing at banks and at places. But we are just not going to work hard enough to implement that. And also, it really doesn't work, especially if your pin number is the same forwards and backwards, for instance.

As many of them are. For people that chose 5555 or 1551, then the machine doesn't know if it's an emergency or not. Right. And tons of them are the same forwards, backwards, or just one number repeated because we don't want to do that labor. They also, as they were considering this, there was one person who interviewed Chuck Stones, who was a guy working for the Kansas Bankers Association. And among the reasons he didn't want banks to implement this is he said, quote, I'm not sure anyone here could remember their PIN numbers backward with a gun to their head, end quote. Because even in a regular situation, it would feel very bizarre to me to do my PIN number backwards. I like i was i was trying to like do some kind of sobriety test or something in a hard way and if you're also being robbed you probably

just won't effectively do this thing yeah although i also like that at one point people thought it was like an easter egg like this is an emergency service but only those in on it know it's like a video game cheat it's like well actually some of in on it know it. It's like a video game cheat. It's like, well, actually, you can summon the police with this secret pin. It's like, actually, I feel like they should have just openly said that. If that's, they don't keep the fact that you dial 911 into your phone to summon an ambulance. That's not like a cheat code. That's the thing they tell you.

Right. It's advertised. It would say on the ATM, it would have a sign on there. Hey, if someone is holding the gun to your head, just enter your pen backward and a cop will arrive. Yeah, it's not like even the secret menu at In-N-Out, you can Google it and find it. But like, no, only people who received this chain email in the early 2000s get to be safe at their bank. That's it. Otherwise, Chase keeps the secret.

Yeah. It's your little reward for unlocking all of the secrets of your banking experience. Like, welcome to the Wells Fargo inner sanctum. And then you get a robe and you get this thing. And yeah, sure. robe and you get this thing and yeah sure actually did you know that if you go to a hospital and ask for a dr hofstetter you get all of your treatment for free it's it's on their secret menu it's like no life is not life is not a video game it doesn't it doesn't work like that there's no there's no contra code for real life

off of that we are going to a short break followed by a whole new takeaway There's no contra code for real life. Off of that, we are going to a short break, followed by a whole new takeaway. I'm Jesse Thorne. I just don't want to leave a mess. This week on Bullseye, Dan Aykroyd talks to me about the Blues Brothers, Ghostbusters, and his very detailed plans about how he'll spend his afterlife. I think I'm going to roam in a few places, yes. I'm going to manifest and roam. All that and more on the next Bullseye from MaximumFun.org and NPR. Hello, teachers and faculty. This is Janet Varney. I'm here to remind you that listening to my podcast,

The JV Club with Janet Varney, is part of the curriculum for the school year. Learning about the teenage years of such guests as Alison Brie, Vicki Peterson, John Hodgman, and so many more is a valuable and enriching experience. One you have no choice but to embrace because yes, listening is mandatory. The JV Club with Janet Varney is available every Thursday on Maximum Fun or wherever you get your podcasts. Thank you. And remember, no running in the halls. Yeah, it's true.

There's two big takeaways for the episode. And this first one we've talked about a bit, but let's get into it. Takeaway number one. but let's get into it. Takeaway number one. It is extremely difficult for a machine to generate a random number. Because you would think a machine is powerful and it can do this, but it's just hard to convince a computer, which is designed to obey our instructions, to then also go be creative and invent random numbers all the time. It's very difficult. This is hard to wrap your head around because most of us computers are so complicated and software so convoluted

and so amazing what it can do that most of us think computers are magic right and that you you would think you don't think of it as a machine because a machine, you know, like if you look at a lamp in your room, that's a machine. It's got a switch. You flip it or turn it, and it closes a circuit. Just a little piece of metal closes the circuit between two wires. Lamp comes on. You flip it off. It opens a circuit.

Okay, it's not connecting to the wires. Lamp goes off. That's a machine that's very easy to understand. The idea that the computer you're using to listen to this or that your phone, that that is the same thing. It's just a series of switches is really hard to wrap your head around. Cause it's like, well, how can you get from a lamp to a machine that has like artificial intelligence or that can play, you know, a video game on it or something like that? This problem of generating random numbers is best understood that way. If you remember that your computer is just a machine,

because if you had a lamp and I gave you the lamp and said, okay, I need you to modify this in a way so that the lamp comes on at night and goes off in the morning you with the tools around your house and with enough time could probably do that because what you would do let's say you'd rig it up to a clock and when the hand comes around to whatever 6 p.m it hits that it causes that switch to flip over and and close the circuit and come on and when that other hand comes around it would trigger a switch and make it flip and go off so you could probably do that if i then said i need you to build me a lamp that goes on and off totally random in a way that cannot be predicted you probably would not know where to start. Yeah.

Because it's like, well, how the hell do you have a machine? Because the machine only does what I tell it to do. You flip the switch, X happens. You open the switch, Y happens. Like, that's it. It's got two states. So how do you get randomness? And what you would probably eventually do is like rig it up to a tree branch outside

so that when the wind blows, it opens and closes the switch based on when the wind is blowing. In other words, you would introduce something from the outside world to the system to introduce randomness to it. But the machine itself cannot do randomness. So if you're thinking about a computer and you're typing computer code and you say, okay, here's the part of the game where it's going to decide whether or not to give this guy a certain fish when he fishes in the video game. So we need it to pick a random value of, we've got 30 different types

of fish and here's the odds that each one will appear. So we need to give him a random number between one and 150. If you tell the computer, okay, give him a random number. How is a machine going to do that? It cannot roll a pair of dice. It doesn't have that mechanism inside of it. It cannot roll a pair of dice. It doesn't have that mechanism inside of it. I just love that this problem exists and then it becomes a thing that, like, while we're all just going about our day, top experts in mathematics and computer science will work full time on this problem and spend all their time generating better and better systems to do this thing that computers, despite their advanced power and capabilities, are almost perfectly designed to not do this unless you really program them correctly to be surprising in a way that is still ultimately not random in the end. It's amazing. Right. Because you typically do not want a lamp that comes on randomly.

Right, because you typically do not want a lamp that comes on randomly. The entire point of that mechanism is that it is predictable. The same input you put in turning the switch gives you the same output, light. If you have a keyboard, when you press the letter A, you want an A to appear on your word processor. You do not want a random letter to appear on your word processor. You do not want a random letter to appear on your word processor ever. But very early in computing, going back to the 1940s, they got into situations where they needed the computer to spit out a random number specifically for cryptography. Because if you're trying to encrypt a message, the best way to create a code that the enemy cannot crack is by introducing an element of randomness to it that they couldn't have known.

In other words, the code is not as simple as the number four is actually the letter A. So the letter A is a random number. number. But what they found out was, well, the easiest way to do is to have a human being just mash a keyboard and come up with a big random string of several hundred characters and let the computer just pick the next one on the list. But computers back then actually could not store a big long list of numbers like that. So even back then they had to try to invent an algorithm, which that would generate a random number anytime they asked it to and quickly found they could not. So instead, what they were doing and what your Nintendo Switch is probably doing or any simple piece of software is not a truly random number at all. It's a pseudo random number, as they say.

Yeah, that's exactly right. And we'll have a few sources for this. One source here is an article for the American Mathematical Society. It's by mathematics professor David Austin of Grand Valley State University. And his description of what we've been talking about includes stuff like, quote, computers simply execute a set of instructions whose output is determined by the input. Since we supply the computer with the instructions and supply the computer with the input, the output is determined by our choices. Then he goes on to say that we want the numbers to share many properties that we would expect a truly random sequence to enjoy. Such a procedure is often called a pseudo-random number generator, since the numbers

generated are not truly random, end quote, because the kinds of practices Jason is describing are not that different than, let's tie our computer to a tree branch and the way it moves is the way we get numbers, or a bunch of other things we have here for ways that people have basically tried to fool their computers into being spontaneous. I love it. It's the strangest thing. Now, we are going to do a poor job of explaining how this works, because you have to be a math genius to fully understand how it works. The actual algorithms for creating pseudo-random numbers, that is, numbers that would seem to be random

when you're from the user side that in fact are actually what they would call deterministic, meaning if you knew everything about how it worked, you could actually predict it. There have been different algorithms over time. They are named after the scientists and the geniuses who came up with them. They are represented by mathematical equations that look like hieroglyphics. Yeah.

I am going to explain this in my dumb caveman way. If there's anyone listening who is an actual engineer and understands this, this is probably going to be insulting, but please, I'm trying to explain it in a way that the average person could understand, and then it is being explained to you by a below average person in me. The early algorithms that they came up with

was that you could start with a number, what they call the seed, and then feed it into an equation. And I think it was something like, we're going to take the square root of that number out to a bunch of digits. And I think it's going to be 10 digits long. So the square root of this number, the seed number, is 4.238458. Then the computer was told to grab the middle four

digits of that long number, make that the seed and feed it through that same equation again, and do that a few times. And then you'll come up with an end number that bears no resemblance to the seed. The seed has to be random and no computer can come up with a random number. So that has to come from somewhere else. Either you can have the user input it, but the problem is if they put input in the number three, it will always give you the same answer because the computer cannot help, but do that.

Right. Yeah. So what they were doing early on was taking the time, the current time, because the computer has to track the time of day because its clock works according to it. And it would like take the current time out to a few decimal points. So you're getting to like the tenth of a second. So it was, you know, it was 1054 a.m. and 23.25 seconds and grab those last three digits and that's your seed

and that's where it gets the random number now to an outsider because they have no way of knowing what time of day the computer did that calculation it is random do you see but that is the only element of random and randomness that was actually, was the time that it happened to process it. So it had to grab something from the outside world, the time at which it happened to do the precise time it happened to do the calculation, to give you the random number. Right. so over time see the problem is if you're saying well why go through all of this you have to realize this was early on about code breaking the enemy has their own computers so if you get something where they can reverse engineer what you're doing and as time goes on you get into the 70s 80s the enemy has supercomputers

so if they can backwards go backwards and figure can go backwards and reverse engineer in a way that no human brain could, then you're kind of screwed. So you have to get weirder and weirder with how you come up with that seed. So for example, computer chips right now, there's multiple methods that we use. And one of them is to take your keystrokes and the exact time that you did the keystroke and have that time out to 10 digits. Yeah. So that you typed the key to start the random generator.

You type that last key at 0.23147 seconds after 1056 AM. Others that are more sophisticated, for instance, the computer that you're using, it is my understanding. Again, if there's engineers out there, please correct us in the comments or email Alex directly. He'll give you the address to the show. Not me. I don't have an email address off the grid. Yeah. Yeah.

Is my understanding that Intel chips will grab their random number from the temperature of the silicone. It will be the temperature out to many decimal points that the processor is operating at, which fluctuates randomly according to many different variables in the universe from the temperature of the room, what software is operating at the time and grabs its own hardware temperature what which it has to monitor anyway and from that we'll put it into its algorithm and now they combine multiple layers and different schools of algorithms to eventually get the random number at the end and then that goes into its encryption and all the things that it that it needs it for yeah there are others that do things like measure atmospheric noise or the radioactive decay of an atom um to get to to get what they what they call entropy it's a value that is truly random and to

get it they have to measure something in the universe that as far as we know, is completely impossible to predict. Yeah. And then every strategy that these computers are doing, which are, again, amazing, they're finding the temperature of themselves or the entropy of the universe, they can only do the strategies we program in. It's still not creative enough, in a sense. And it's still something that they're just, they're still just spitting out what we instructed them to spit out. We're just hoping that what they spit out is something that other people and other computers cannot reverse engineer. It's still not random or truly random. That's the only word. Yeah. And that right there is where you truly go down the rabbit hole right because the reason

they keep having to come up with new techniques for doing this is it every time they create one of these they will discover that if you crunch the numbers hard enough that it's like not giving you a truly random output you've actually got a slightly narrower range of outputs, which again makes it a little bit easier for a bad actor or someone from the outside to eventually reverse engineer it. This gets to what I had said at the top of the show about the nature of the universe, because the whole thing is they're trying to get data that is not deterministic, Because the whole thing is they're trying to get data that is not deterministic. That is not the proverbial lamp and the switch.

That is a deterministic machine. You flip the switch, the light comes on. In theory, if you had a powerful enough thinking machine, you could predict everything. For example, we thought the weather was random for a long time, but now we can predict the weather with pretty good accuracy. We can't predict that a year in advance, but we can predict that the next day's weather. I don't know if you've noticed this, but weather forecasting has gotten much, much better over the last couple of decades due to the power of computing. Yeah. And that is a system that has so many elements you would think of it as being random.

That's why you can't do it to 100% because there's too many factors at play. But in theory, someday with a powerful enough machine, when you can truly factor in all the factors, all of the temperature of all the air everywhere, all of the weather, the wind patterns, the moisture levels everywhere, if you can measure it to that precise of an amount, and you have a sophisticated enough algorithm, you could get it down to, it is going to rain at 5 p.m. two weeks from now on Thursday. In other words, it is the oldest philosophical question in humanity, whether or not the universe is random or deterministic. Like, is your destiny set in stone? And as we have gotten more and more sophisticated computing, we pretty much have always found

you can predict something with a powerful enough computer. If you had an infinitely powerful computer, one that could track the trajectory and the state of every particle in the universe, in theory, as philosophers have been saying for centuries, it could predict the events that are going to happen everywhere a thousand years from now, including the actions of the people. Because their brains and their bodies are particles in that universe, which would mean that free will doesn't exist.

And I love that you can reach that question by trying to be a mathematician who's trying to generate lottery numbers. You know, like that can be your that can be step one of your journey to saying, oh, is there free will? Is the universe determined? It's amazing. Like, what a what a thing. It's so great that computers can't do this. I'm very excited. Yes.

But, and while this sounds like you've stumbled onto like an episode of the Joe Rogan podcast and it's like, like four hours into it and they're both incredibly stoned and they're like, man, if like God had a computer, he could predict, he could predict predict what you're going to do. The reason this is a better podcast is you and I got stoned much faster.

It doesn't take us four hours, right? We smoke much more rapidly. We're just good at it. Yeah. Well, the key is you start before you turn the microphone on. Yes, exactly. Be a pro. This is not a stoner's thought experiment

this is what computers are trying to do think about the everyday problems computers are trying to solve for example self-driving cars a self-driving car that cannot predict what's about to happen is worthless it is not. It cannot see an accident occurring and then say, oh, that's sad. It has to, it has to see the accident coming and, and predict it in advance enough to take some action. You do that, you know, you do it every day. Your brain is built to predict things. Um, So if you have a self-driving car, it's got to be able to match a sober human's ability to do that. And it's got to be able to look at all these variables from the weather to how the other drivers are driving, all of these

infinite factors, you know, the psychology of the other drivers, their mannerisms, and it's got to factor all of that in into, oh, we better slow down right now. Or, oh, we'd better, you know, this, this other person is clear. They don't see the stoplight. We've got to stop. And it has to account for other people's mistakes as to account for road conditions, all this stuff. You're trying to build a computer that is trying to take chaos and make predictions

out of it. One thing that has been controversial, for example, is computing that is trying to predict crime before it happens. There are people right now feverishly trying to do that based on analyzing all sorts of behavior and patterns and things like that to determine where patrols should be. Because again, policing, you know, the idea is to prevent the crime from happening, not to just get revenge on the person after they've murdered you.

Right. But that gets us into a very weird situation because there it gets into that whole thing of, well, by increasing patrols in this area, did you actually cause crime to go up or go down? Like, how did you influence the system? This is not a hypothetical. This is something people are making software they're trying to sell to cities right now. Yeah, it's actually less astounding than the Minority Report version, too. Like, it doesn't need to be three weird, pale people in a vat.

Like, it's just somebody who could program lottery numbers, too. Like it doesn't need to be three weird pale people in a vat. Like it's just somebody who could program lottery numbers, too. Yeah. Yeah. And it's them crunching a whole bunch of data on what type of person commits crime or what their behavior leading up to the crime looks like. And if you don't understand why you would want that, think about a regime that is worried about uprisings or protests or riots. So they want to be able to anticipate what the conditions are in a city or in a region leading up to a riot. Like how far away are we from some kind of breakdown? This is a technology that exists in the book I just wrote that Alex will talk about it after the show. But this is something that they all badly want to be able to monitor social media messages,

the movement of human beings, things like that, and say, hey, you are 48 hours away from violence breaking out. You need to go into curfew now. Think about the pandemic that we just had. You need to go into curfew now. Think about the pandemic that we just had. Think about how much it would have helped to be able to predict the fact that X number of people will not lock down. X percentage of people will not wear a mask.

Like how that would have changed policy if they had known sufficiently in advance by looking at past data, past behavior data, movement data, all of that to be able to say, hey, we predict that if you issue a lockdown on the X percentage, you're going to comply. Therefore, you need to instead do you need more strict measures. You need travel ban, something like that. We continually were shocked by people not complying with certain things. complying with certain things um and will they will be writing about this for decades right about you know how it got so out of control how the the virus spread so fast why people didn't stop going to bars and parties you know and all of that stuff but these are all things that could have been predicted if you had sufficient data because even though you know every teenager who decided to

sneak out of the house and go to an underground house party and spread it to all their friends, they thought they were making a spontaneous decision that with enough data, you absolutely could have predicted, hey, nationwide X number of teenagers are going to go do this. It is something that you can actually know in advance. The same as, you know, companies have to make these projections all the time when they're trying to project how many people are going to want a PlayStation 5 on release day. And so we only manufacture a tenth that many. You know, everybody who made that purchase decision thinks they made it as an individual.

But decades of market research told them, no, we're going to need to have 5 million of these. Yeah. And then it almost makes it seem like the human created seemingly random numbers are only random because we just haven't programmed effectively enough to know it. And then in general, random numbers are basically mythological. Like it's like it's it's like how centaurs don't exist. There are also not numbers that are random either. It's just something that we haven't sussed out the structure of yet. Yeah. And not just random numbers, but random events, random everything. Everything in your world that you think is random, your brain is in a constant battle to try to find the pattern in it, which brings us, I think, to takeaway number two. Yeah, it does. Let's go into the other takeaway. Takeaway number two.

There are a couple strange cases of people beating these supposedly random number systems for game shows and lotteries. And we have two stories here. I find them both amazing, especially because maybe you, the listener, heard, oh, there's a way to beat a game show or beat a lottery where I always win. Now I will be rich. And these folks did not get rich, but in ways I find very interesting. Yeah. And it's also a couple of great examples of where the game has tried to make itself appear random, but a smart enough person figured out otherwise. And the first story here, and I think you initially picked this one out because then I was amazed to learn about it, but it's the story of a man named Michael Larson.

And Michael Larson went on a TV game show called Press Your Luck in 1984. And Michael Larson broke the game because it had a big board that was supposedly random that was extremely not. And he had just figured out the human generated system that made this board go. So then he just very publicly beat the game as badly as it can be beaten. Yeah. And how much did he wind up winning? In the 1980s, CBS had a limit on how much someone could win on a daytime game show. They were only allowed to win $25,000, and then after that, they couldn't come back. And that's $25,000 in the money of that time. Michael Larson went on this game, Press Your Luck, and won $110,237, which is approaching $300,000 in today's money.

And he also won a trip to the Bahamas and a trip to the island of Kauai in Hawaii, and he won a sailboat. So the actual value of the cash winnings is even higher. He just completely broke the game to the point where they had to split his one taping into two entire episodes of television because it just ran too long to fit in the standard time slot. I used to watch this show religiously and i i would have been like nine years old at the time and probably it maybe if i i swear i saw this episode i don't know if i'm fooling myself into it but press your luck is you know if you see like the screenshots of it if you land on the square it's like you win

like four hundred dollars like it's a very low stakes it's kind of it's you land on the square it's like you win like four hundred dollars like it's a very low stakes it's kind of it's kind of like the old days of jeopardy where you know the weather only only won like four thousand dollars it's like well that's that's less than what it cost him to fly there um so it's not the type of show where you it like changes your life it was it was a goofy show and then you've got these little what they call whammies and if you land on one you lose all your money so the thing is you keep pressing the stupid button and this little it's just this little light up thing goes around the board and seemingly random numbers light up and if you press your luck if it lands on you know five hundred dollars you can either keep the cash or you can hit the button and try again and you're you build up more and more but as you do it you have a great

and greater chance of hitting the whammy thing which wipes you out to zero that's the entire game uh and was fun it's the type of game a little kid can understand and this guy simply watched enough episodes that he figured out an exact way to never get the whammy so he just kept spinning forever for two straight days till he almost bankrupted the show and i i know that the the episode has to be on youtube somewhere i just can't imagine like i'm imagining how angry the producers had to have been back in the control room because at some point they had to have realized what had happened not yeah oh we've got a great winner here it's not like like on jeopardy where you know hypothetically if you knew somebody that went on and they just won for like the whole week and it's like thrilling because

they're winning through knowing things this is a this is like the dumbest game show in the world there's just a big red like a a toddler could play Press Your Luck. You just slap this big red button in front of you. There's no skill. There's no nothing. You just keep slapping it until you lose. So there had to be a point where this guy, he's wound up spinning 45 times in a row without hitting the whammy, which normally it's hard to do it like three times in a row.

I can't emphasize how absurd it is to spin 45 times in a row like you can't i can't emphasize how absurd it is to spin 45 times in a row so after like the 12th the producers had to have been like this guy has figured out something either our board is broken but you what can you do you can't interrupt the game you can't rush onto the set and tackle him and try to find, does he have a transmitter in his pocket? Is he doing something? Does he have a friend that's messing with it? All he did was realize that their random number generator was broken.

It actually had a pattern. And if you watch it closely enough, as this guy did, it could be beaten. That's really all it was. And I don't know if it's the entire episode, but I was able to find a YouTube clip of just his spin rounds. Because as Jason said, this game show, it had a trivia component. You do trivia to win spins on the board, and then the spins are where you get your money and you press your luck from there.

He was pretty bad at the trivia, but he just got enough. your luck from there. He was pretty bad at the trivia, but he just got enough. He got, you know, more than zero spins and then started winning all of the time in that money section. In the clip, I could find like, as he's passed, you know, or 10 spins or so, at that point, the live studio audience is like yelling stuff. And the host is pretty visibly, like he keeps doing the thing where he just walks away from his podium in shock and then walks back to it. You just see everyone melting down as this guy keeps hitting the multiple spaces on the board that have absolutely no chance of being a whammy.

The game mechanic is it's fun to see somebody hit a whammy and lose. He knows that several locations on the extremely not random board you can't lose on. So then he just keeps hitting them and keeps playing and keeps winning free spins to do more spins. We'll have links about exactly how this game works, but it's very fun how little strategy and how little what I would call code breaking was necessary to beat Press Your Luck. It was a really poorly constructed game. Yeah. The guy would spend the rest of his life, it's kind of a tragic and fascinating story, would spend the rest of his life like chasing that thrill again and would never get it

until he finally ran a Ponzi scheme, I think, and died like being hunted by the FBI. Yeah. I want to hit the board on then what happened to him after, because the board will have a Hollywood Reporter link where they interview Michael Brockman, who was the head of CBS daytime programming at the time. And he was describing the first episode of Press Your Luck. He said, quote, pilots are test vehicles and you can cut corners. And this show, what was expensive to create was the light pattern on the board. No one wanted to spend that much money on the pilot.

And then the pattern wasn't improved enough when the show went into production, end quote. So they had this board where all of the spaces on it would only show three different things. And Michael Larson bought a VCR. So he recorded episodes of the show and just noticed this really easily. And then from there, he does win all this money. But then from there, he tries to do a slightly harder contest where a local radio station in his part of Ohio was doing a contest where it was based on $1 bill serial numbers. So if you heard the serial number of a bill you had, you won. And so what he did is he withdrew almost 50 grand of his winnings in $1 bills, and then kept that at his house, failed to win the radio contest, and then he had the bad luck of somebody robbing his

home and saying, oh, there's 50 grand here. And, you know, between taxes and everything else, they kind of stole most of his press your luck winnings just by chance. And then from there, yes, he ran a Ponzi scheme and died before the combined SEC, IRS and FBI could catch him. There's also there's a site called Priceonomics that quotes an interview with his brother that said, quote, winning that game show was the start of Michael's downfall. It made him think he could trick anybody and do just about anything he pleased, end quote. Because this so shockingly non-random board was something he could break, and then he was chasing that dragon the rest of his life. Cannot emphasize enough, The house always wins in Vegas, in life with the house always wins. It's if you, if you somehow won some and got away, they will, they will pursue you to the end of your life, trying to get you to chase that thrill again. And they will get their money back some somehow. And then maybe not that specific game show, but in general, the system in general always wins. Yeah. And that wisdom was possessed by the person in our other story here,

because this is the story of somebody beating the lottery. And his name is Mohan Srivastava. And Mr. Srivastava was a geological statistician living in Toronto. And in 2003, he found a flaw in the system for Ontario scratch-off lottery tickets because somebody gave him some scratch-offs as a Christmas present and he won $3 Canadian. And then because he is a trained statistician with degrees from MIT and from Stanford, trained statistician with degrees from MIT and from Stanford, also has a job where he just looks for the exploits you can find in geology to go find oil and precious things to mine. He also got interested in the lottery tickets this way, and he just spotted a way to know whether these scratch offs were winners before you scratch them. That's all he's all he did. That like it was yet another non random number system that is dressed up to be random.

Right. But it's a case where if you are smart enough, you can do the math on not it's basically what it came down to was you could make more money by buying the cards than watch, than what it would cost you if you know what you're if you know what you're doing right yeah because it came down to the distribution of the tickets and the way they were because obviously if if they you know the lottery people have to know how much money they're giving away they've got a certain number of tickets that are that are winners because this is not the thing where you're picking the number these are scratch-offs and we'll

have linked a picture of this ticket it It was a specific province of Ontario lottery scratch-off ticket where the ticket shows a set of eight tic-tac-toe boards. So that's 72 numbers total. It's tic-tac-toe boards, but there's a number in each square. And Srivastava figured out that you could see whether the ticket was a winner based on how often the numbers displayed on it repeated within the ticket. And so you didn't need to hold it up to light or anything. There was no like scratching needed. And then he enjoyed the fun of figuring this out. He bought a set of lottery tickets and confirmed it worked.

And then his next move was not to like try to get rich real fast. He told the ontario lottery system and uh according to him quote eventually after a bit of telephone tag i decided to send them 20 tickets that i had separated into winners and losers i said go ahead and scratch them off and they did and realized that they really did need to talk to me end quote like he he figured out that there was just not enough money and breaking this thing for it to be worthwhile and then the lottery system was like slow to even bother taking him up on fixing it it's very strange yeah he said if he had made his

full-time job to do this he would make what he said like like $600 a day. Yeah. But just decided that wasn't worth it. But how much is that? $600 a day. And how much would that be in a month? It's a fair amount of money, right? Yeah, I think he is very well paid by mining and oil companies because he told an interviewer, quote, if it was my full-time job, I'd have to travel from store to store and spend 45 seconds cracking each card. Estimated I could make $600 a day that way.

That's not bad, but to be honest, I make more as a consultant, and I find consulting to be a lot more interesting than scratch lottery tickets, end quote. And he also said that he figured out the top prize in the game was 50 grand, but there were 4 million tickets in the game and only eight tickets with that top prize. And Ontario is enormous. It's as big as about one and a half Texases. So it'd be a lot of driving around and trying to track down these individual tickets. It just seemed like too much of a challenge. And not a fun day like just driving all day going to convenience stores and and using a nickel on tickets is not that fun right but this is why you hire somebody and you say for look i will pay you a hundred dollars a day and all you got to do

is go around and buy these tickets for me that's it's your only job it's some driving this will more than cover your mileage uh you know some college kid be happy to do this. They don't have to know that you're making six times that on your end is send them around. Like you got to you got to be smart. That's the thing. You don't work harder. You work smarter. And then after you've collected the money, then after the fact, you contact the lottery people and say, you may want to change this for next time because i i made 150 000 uh doing this but whatever right you you call them on your solid gold telephone and say i purchased this telephone with with winnings you should look into it yeah because it's completely legal yeah they set the rules. I mean, you didn't steal anything. They did not randomize their game sufficiently. And it's like, look, you make your money by tricking poor people into thinking they can get rich this way. So I don't feel bad about exploiting your system, your entire system as exploitation.

exploiting your system your entire system as exploitation yeah that's fascinating because he also according to the interviews with them and we'll have them linked Srivastava says that one reason he stopped is that he realized that if he personally beat the house on this he was also like Ontario had rigged the tickets so they would take in more money than they gave out. That's how lotteries work. But he realized that the more he won, the fewer winning tickets would be in the system for other people. So it's not like he's Robin Hood or something. He's just gobbling up all the winners and then other people lose this game more often.

So I don't know if that's the only reason he stopped, but it's kind of a nice positive reason to stop. Yeah. And this is where some people are yelling at their device for us to talk about the McDonald's Monopoly game that they had that HBO documentary about. I do not remember the name of the documentary, but I'm sure you can link to it in your sources. Yeah, totally. can link to it in your sources yeah totally um but it was a similar thing where it's this crazy story where again what you thought it was a random game and i don't know if some of you listening to this probably remember you would get a couple like two i think monopoly pieces on stuck to your cup or your fries or whatever and you had a board you could put them on and in theory

you could win you know the cars and houses and all sorts of stuff um and then it turned out it was it was rigged and there was like somebody with mafia ties this whole convoluted story where insiders made sure that they got because again it's this incredibly sophisticated system where they they know how many cars they're going to give away right right but they they also if you print out those winning the winning car piece and then the person doesn't play it and just throws their cup in the trash and forgets to even look so you've got to make sure it winds up in the hands of somebody but you've also once they've won there can't be another winner so in all of that complexity and trying to give out what should be

a random result there were people that basically rigged it and you and I had no chance of winning, it turned out. So it would have been the equivalent of this if this guy had, in fact, rigged it to get every winning ticket and not everybody else was just wasting their money. That actually happened with the McDonald's Monopoly game. In hindsight, I think that is the first gambling I ever participated in. Like I wasn't buying the food, my parents were buying it, but I was a little kid like peeling Connecticut Avenue off of my fries to put it on a little board. What a strange activity. Right. And it created all of the feelings of gambling because you could fill in like four of the five slots on your board or however it worked and think, oh my gosh, I'm really close to the car. Yeah.

But I got two out of the three light blues as a little kid. And I was like, oh, we got to eat McDonald's the rest of the month. Like, like, let's get over there again. Quick. Yeah. We're 66% of the way toward a car. But then you, you get 13 more tickets from eating at McDonald's every single day for two weeks. And it's like, oh, I've got

six more of the same, the second piece of the three. Why can't I get that third? It's like, well, they only printed up three of those and the third out of, out of the 8 billion items McDonald's sells in a week, you know, it's, you're now getting into like individual spoonfuls of water in the ocean that it creates the illusion that you're close when you are in fact you know at the mercy of of randomness and that i guess is the culminating point of all of this is that you have two challenges you have something that is random like that game that it is trying to make appear not random and there are other things that are not random that are trying to make themselves appear random and in both cases

it is usually with the goal of getting you to spend money and so that candy crush will will let you get three extra turns if you spend a couple of dollars or or whatever but whether or not those three extra turns actually give you the pieces you need to beat the board is is subject to a random number generator whereas they want you to feel like it's a skill thing it's like oh i made the wrong move and it's like actually there's not any move you could have made it did not give you the pieces you need yeah and maybe free will is an illusion. We don't know. And we kind of find out by thinking about lotteries and without pieces. Folks,

that is the main episode for this week. My thanks to Jason Pargin for joining me for the holidays and helping me decide to play the lottery the old-fashioned way, which is to not play at all. It's a pre-lottery world in my head. Anyway, I said that's the main episode because there is more secretly incredibly fascinating stuff available to you right now. If you support this show on Patreon.com, patrons get a bonus show every week where we explore one obviously incredibly fascinating story related to the main episode. This week's bonus topic is the United States-Vietnam War draft lottery in 1969. And then, from there, there's a whole nother bonus topic within the bonus topic. It's bonuses all the way down this week.

I'm really, really glad about that. So visit SIFpod.fun for that bonus show within a bonus show, for a library of nearly two dozen other bonus shows, and to back this entire podcast operation. And thank you for exploring random numbers with us. Here is one more run through the big takeaways. Takeaway number one, it is extremely difficult for a machine to generate a random number. And takeaway number two, there are strange cases of people beating these supposedly random number

systems for game shows and for lotteries. And there's lots more mini takeaways in there about lotteries and pin numbers and all the other digits that run our lives. Those are the takeaways. Also, please follow my guest. Jason Pargin is at JohnDiesAtTheEN on Twitter. That's the title, John Dies at the End minus a letter because of character limits. His new book is entitled Zoe Punches the Future in the D**k. That's written under the soon-to-be retired pen name David Wong. Find it at your local bookseller or in the episode links. Many research sources this week. Here are some key ones. A great article in Slate by Nick Green outlining the hilarious lack of difference between America's Mega Millions and

Powerball lotteries. Also an amazing article from the American Mathematical Society written by Grand Valley State University professor David Austin on the difficulties in random number generation. I know everything I just said sounds very dense, but it is actually very readable and very fun. You should check it out. I'm also linking pieces from The Hollywood Reporter, Priceonomics, NPR, Wired, and a few more sources, exploring the two parallel yet very different experiences of Michael Larson on the game show Press Your Luck and Mohan Srivastava breaking the Ontario scratch-off lottery system.

Find those and many more sources in this episode's links at sifpod.fun. And beyond all that, our theme music is Unbroken Unshaven by The Budos Band. Our show logo is by artist Burton Durand. Special thanks to Chris Souza for audio mastering on this episode. Extra, extra special thanks go to our patrons. I hope you love this week's double bonus show. And, and a little bit of an extended thank you to all our listeners, because this comes out on the final Monday of 2020. The next episode will be in 2021, which I feel is a science fiction year. That is only for sci-fi stories and C-Lab 2021 cartoons, and that's it.

Anyway, everybody's experience of 2020 has been different. Obviously, there are highlights and lowlights, and nationally and internationally, there are a ton of lowlights and a lot of danger and worry and just negative things. So I want to say I'm grateful to you for listening to this podcast and making it a thing. It was created this fall, and it's really improved my 2020 in ways you cannot know. I'm also amazed that some of you showed me your Spotify year-end thingy, if you listen on Spotify, and the fraction of the audience that listens on Spotify had a subset of people who had this podcast in their top listens of 2020. It's only existed since August. I'm amazed that that happened and fully didn't expect it. So thank you for that. That's just really cool,

really gratifying. There have been other highlights too. The show got written up by the AV Club. It met its mini membership drive. It's been just a real joy meeting a lot of you and getting to know you remotely by doing this. And the best part is we keep on going. Like now this is a thing and we do it in 2021. And I am so excited about what is already lined up for the new year. You'll find out very soon. So thank you for being such an amazing part of my 2020.

And I hope it's been a good part of yours. I'm thrilled to say we will be back next week, which is next year, with more secretly incredibly fascinating. So how about that? Talk to you then.