Baccarat, ETGs and AI: viewpoint from the world’s most successful casino cheat

In the second installment of AGB’s interview with Richard Marcus, the world’s most successful casino cheat breaks down why Roulette is the most cheated table game worldwide, the particular vulnerability of electronic table games (ETGs) – largely due to a lack of oversight, and how artificial intelligence (AI) is heavily weighted to help those who want to cheat.


It’s a good point to make: the whole world believes that baccarat is the most cheated game in the world by far, it’s not even close. Roulette is cheated across the world four or five times more than Baccarat.

Baccarat is, obviously, the most cheated game in Asia because, I guess 80-85 percent of the gaming tables are Baccarat.

Like I mentioned before, you read about the Baccarat scams – they’re always for large amounts of money where all the other cheating tends to be for smaller amounts of money. But if you add it up cumulatively, there’s more money being cheated at Roulette across the world than Baccarat.

Now a lot of people have trouble believing that, but when you go to Europe it’s the complete opposite – the casinos there’s 60-70 percent Roulette wheels and maybe they have a few Baccarat tables.

I want to go into, aside from table games – which are a huge component, we’ve seen a huge rise of ETGs and obviously slots have been a huge component of Vegas over the many years. How vulnerable are these electronic table games to any type of manipulation and then cheating?

Well, I could answer that question by asking you, or anybody else: Have you ever walked through a casino where the slots are and where the ETGs are?

Oh yeah.

Do you ever see anybody watching them, from the casino? Do you ever see any kind of game protection taking place? Are the machines more or less all by themselves? And, you know what, it tends to be the same way up in surveillance, they don’t watch them either. Especially the ETGs table games. Nobody watches them. Then there’s sophisticated methods to cheat them, all high-tech. One is called visual ballistics, where they use a prediction app that they bought online and they use a metronome that records sound and records tracking and all that.

There was a young kid, he was a Google programmer out of Australia, his name was Yujin Wu and he did just that. He was armed with a metronome and a prediction application, and he went across places like Greece, Turkey and Australia, then he went to the UK and he was winning lots of money. In four or five months he won, or he cheated, about half a million dollars. And now he was ready for the big time, he was going to the United States, and he was going to go do his thing there. But he made the mistake, he went to Quebec.

In Canada, he went to Quebec, and he went into a relatively small casino, Casino de Charlevoix in Quebec, and he used his equipment – he cheated them for $20-25,000. That’s a lot of money in a casino like that.

Then he went back two weeks later, did it again, took another $25,000 and then they got suspicious, but they had no idea what he was doing. So, they decided to wait for him the third time and as soon as he did come back the third time, again two weeks later, they were ready, and they were filming him. And they saw him tapping on his cell phone. And they succeeded in busting him, even though they had no clue what he was doing, they busted him.

When I heard about that, I realized: this kid, he was a genius, but he had no street sense. You don’t go a little casino in Charlevoix to make $20,000, when Caesars Palace and all these big giant casinos are in Las Vegas, where you take $100,000 out of there and it’s nothing.

I think he would have been the most devastating cheater in the history of legalized gambling had he had the brains, the street smarts. He would have made hundreds of millions of dollars. And, as a result of him, all the manufacturers of ETGs immediately went to work on strengthening their vulnerabilities with the machines.

And the slot machines, they’re also vulnerable because they’re not only vulnerable to cheating now, they’re also vulnerable to advantaged play, where guys can use high technology to actually know when there’s a very good chance, or much better chance than the odds would dictate, when the jackpot will hit. And nobody watches them.

They have to handle their equipment, sometimes they have a smart phone in their pocket in an unusual place, where if casino surveillance or game protection people just spent time walking around the slot areas, they would probably – once in a while – stumble onto somebody who’s just not looking normal.

Like if somebody’s holding two cell phones while they’re playing slots, there’s something going on.

The bottom line here is that casinos just do not pay attention to ETGs and slot machines.

That also brings into question what is happening with AI, I know that that’s going to be one of the components in the upcoming conference as well. Is technology in general, but specifically talking about AI, is that weighted more in the favor of people who are wanting to try cheat, or to the people who are trying to make sure they don’t?

Letter A: it helps the people who are trying to cheat. Now, why? That’s a very bold statement. Because AI in itself – and I will repeat this to you that no matter how good the technology is, the human factor is what makes it work. Someone like me, I would say “bring on all the AI you want, I don’t care”. Because the reason is: the AI gives you data, but the AI – as far as I know, right now – is not capable of tapping you on the shoulder, if you happen to a floor supervisor and you’re standing in your Baccarat pit, the AI is not going to tap you on the shoulder and point you over there and say “That guy, in the disguise, that’s Richard Marcus, he’s about to do a big cheating move on that Baccarat table”.

They’re not going to tell you that.

So, what AI does is it gives information. It’s data, data, data, information, information, information.

So, it comes down to: in general, as far as cheating is concerned, do the casino people know what to do? Now, they do know what to do when they get the data from customer service. That’s easy – it tells you what games people like, where you should have your tables placed, it helps all of that. But does it really help with cheating? It should. Does it? As of right now, not really.

Facial recognition, we can consider that AI as well. Facial recognition assists in identifying people that are cheaters, advantaged players, and what-have-you, but how accurate is it? If I walk into a casino through a chokepoint – a chokepoint is the few areas in a casino have to pass through – and I’m wearing a cap, and I put my head down, or as I’m walking through, I start rubbing my face or whatever, how accurate is it? What percentage of these undesirables are they going to detect and be able to alert surveillance (to see) if the people are in the database? I don’t know.

But on the other side of this, the advantage of AI to the cheats is that the cheats – they’re developing their technology right along with it, and they know how to use it. And the data that they get from their technology, they maximize it, as far as usage is concerned, to 100 percent. Whereas what percentage of the value of AI, when we talk about cheating, are casinos really getting? I’d say it’s very, very low.

I always like to say, the more technology advances, what happens? The human beings, they depend solely on that technology, and they forego training, they forego basic game protection procedure.

And by the way, the fact that procedure is not enforced enough in these casinos. Part of that is there’s not enough supervisors per table. So as the technology goes up, the human efficiency goes right down with it, or against it.

So, it’s no doubt that it favors the cheat.

We’ve seen that extend from beyond the casino floor into some of these larger hacks, like we saw happen in Vegas last year as well. The companies are more easily penetrated via third parties than they are through their own systems. So, as soon as they’ve outsourced any of their operations to a third-party system then that human factor comes into play. Is there any way – given how much psychology you’ve used over the years, how much you understand how you need to influence people in order to act a certain way – is there any way that these casinos or these integrated resorts can eliminate the possibility of risk from that third-party human factor?

Eliminate, no. Reduce drastically, yes. And it’s two words: Training and Education. What happened in Las Vegas, the scam that happened at the Circa casino in Las Vegas, which is called social engineering – where the third-party situation comes into play.

And the result of it is that the target person – who they’ve identified, these people that do these scams, these hackers, before they choose their target they do all kinds of research online, whatever they could find about their target person. So, the target person at the Circa was a young, female cashier in the cage, as far as I know. And they found everything out they could about her – you can find out so much online, everything is an open book, and they decided she was vulnerable to be convinced that: if somebody called her or texted her with the right manipulative language – and again the psychological effect is always there – they would be able to convince her that she had to get a million dollars from the cage, walk it out the door, meet somebody behind a McDonalds, and pass off that money, thinking she was saving the casino from some great tragedy.

It’s all psychological manipulation, but it’s the technology that allows them to do it. So, the only way casinos can get anywhere near close to eliminating that is through training, education, strict procedures. Simple things like one person should never be alone in vulnerable areas – like collecting money. It’s just about tightening procedure and the training and the education, and you can never overtrain.

Because, when you also have – when you’re training people who are working in casinos in important positions, right from the dealer to the person in the cage, to the director of surveillance, you have to overcome the fact that it’s not their money, not the employees’ money that’s at risk.

Most employees, they want to do well, they want to protect the casino’s assets. But they need direction. And globally it is not there.

Speaking of training, that is one of the major aspects of what you do now. But also, you have your conference that you founded coming up, it’s going to be March 5th in Las Vegas – Global Game Protection and Table Games Conference. What are some of the main aspects of that that you think people should be coming to try and learn about this industry, and how to protect themselves?

It’s the same two words: it’s training and education. We cover everything, in table games. We cover new table games, we cover analytics, we cover marketing, we cover player development, we cover the relationship that table games departments need to have with surveillance departments. And AI for both tracks. It’s a two-track conference – table games and game protection – which includes surveillance, compliance and security. We have 50 different speakers and they each speak about one aspect of it. It could be artificial intelligence, facial recognition, player development – how to get the right high-rollers into your casino, and on the surveillance side that they have to verify that these really are the right players that you’re getting into your casinos. Because a lot of the right players that you think are coming to your casino because they qualify for million-dollar credit lines: they’re the wrong players.

I want to ask you: Do you miss it? And do you know if any of your former partners are still operating?

Yes, at times I miss it because it was really so much fun. More than the money. We were making a lot of money, I started off making a lot of money right away, as soon as I got into this. It then became more of a challenge. And every time I invented a new move, especially the Savannah move, and when I actually saw it work. I conceived it in my brain, it starts off as a ridiculous thought and then I learned that the best casino cheating moves in history, including the Savannah, ridiculousness is the best component of it. A KISS move – Keep It Simple Stupid – that Savannah is an example of a KISS move. It’s so simple and so stupid, and while I was doing it, the casinos thought – like I mentioned before – they knew it was me, they knew I was cheating, they didn’t know how. Later, I found out, when I became a game protection consultant and I started working with the same people, or actually training the same people who were chasing me, trying to bust me, I found out from them that they were taking Roulette wheels apart in Las Vegas, looking for electronic devices – even at that time – thinking that I had some kind of control of where the ball would land.

And all I was doing, I was just betting $10,000 or $5,000 on a Roulette bet – even money or two-to-one bet – and if I won I got paid and if I lost I didn’t let them take the money. That was it.

The first time the Savannah worked and they paid me in these $1,000 and $5,000 chips – which have less wear-and-tear on them than the other chips because they’re not used that much, and when you jiggle them in your hand they have a different sound because of their newness, a different feel to it.

So when I first got paid on that Savannah, when it worked, the feeling of the chips in my hand, and when I was walking away from the table after I got paid, and I was jiggling them, it was the most intense high you can ever imagine, better than sex, I swear. It was that good. 

So yeah, I miss it, but I really enjoy what I’m doing now. So I don’t think about it much. It was by-far the best time of my life.

The other part of your question, are my partners still doing it? Actually now, no. But they did continue working for four or five years. Actually it was about four years because they knew I was writing a book about it. So, I agreed with them: “we’re all in our 40s now, you guys can’t do this forever”. And without me it was going to be a little harder cause I was the leader, and they did succeed on their own.

And I said “okay, I’m going to write the book, it’ll probably be published about four years later”, which it was, and that’s when they quit.

That’s amazing. Again, I want to keep you’re here cause there’s so much you could tell me, but I think we’re going to have to cut it off for now. We will be happy to have you again very soon. Thank you so much.

Richard Marcus, the world’s most famous casino cheater and now a casino table games protection consultant. Thank you for your time.

Thank you very much for having me, Kelsey.