Lou DeGregorio has been well known in the casino industry for over forty years. He worked at Paul-Son Gaming Supplies for twenty-three years, eventually becoming Sr. Vice President of Sales, and was instrumental in Paul-Son becoming the leading supplier of live gaming equipment in the world. He introduced the first “carnival” game, Caribbean, Stud, to casinos, and his company D.P. Stud introduced No Commission Baccarat, Progressive Pai-Gow Poker, and several other games to the industry. Now Lou is focused on Shark Trap, a combination automatic shuffler and system to catch all varieties of marked cards, which he says will revolutionize table games. There are three versions of the shuffler for single deck games: one for poker, one for blackjack, and one for carnival games. The shuffler will fit in the palm of your hand. Pop a deck in, and the first hand is ready to be dealt in 3 seconds. No more need to alternate red and blue decks. Say someone is eating at the table, and a small amount of soy sauce, for example, ends up on the backs of some cards. If it turns out that it is showing up only on, say, tens and aces, and in the same general area, the pit and surveillance get alerted. If it is random, but the cards have reached a point where the deck needs to be changed, that would also get alerted. Lou tells me, “We believe the Shark Trap system will be a game changer, and literally change the way casinos operate and generate revenue from their table games.”
Shark Trap will be at G2E in booth 1937.
>>>>>>>>>>
Q: Could you give us an overview of the Shark Trap system?
LDG: The Shark Trap system is at its core a security platform for table games that comprises a network of super shufflers.
Q: Before we get into the details, what was the genesis of this system?
LDG: A few years back, we had the idea that technology had advanced to the point where it should be possible for shufflers to detect all forms of marked cards in real time, so we began exploring the possibility. After only a few months of research, we were convinced that a scientifically robust and reliable system could be developed, and that it could be offered to the industry at prices comparable to the normal leasing costs of today’s shufflers. Once we made the commitment to move forward, several layers of protection and sophistication within the platform began to evolve.
Q: The idea that a system can detect all forms of marked cards seems like an impossible accomplishment; how did you approach the problem?
LDG: You’re right, it’s an extremely difficult problem to solve because the field of marked cards is vast. But we were able to quantify the problem in a way that lent itself to a scientific solution, and it started with completing the most comprehensive study on marked cards ever conducted. We plan to publish the study someday, but for now, let me just say that in addition to talking with several experts, the study consisted of gaining access to the finest private collections of marked cards known to exist (e.g., one collection consisted of 3000 decks, 500 unique marking systems), libraries (public and private), catalogs, newspapers, magazines, and most other written sources going back to the mid-1800s in the USA and even earlier in European sources. After compiling a massive list of card marking systems, we were able to categorize the systems from an algorithmic perspective so our technology partners in the computer vision world could develop the corresponding algorithms for detecting all systems within a category.
Q: How many different marking systems are there?
LDG: We don’t believe anyone can answer that question, but what was discovered after painstaking research is that we were able to determine the total number of possible card marking categories. From this point we looked at categories that exhibited overlapping traits, categories with inverse traits to other categories, and much more, all in an effort to ascertain the optimal number of categories. In short, we were able to define and quantify the problem in a way that has never been done before.
Q: What about future marking systems?
LDG: The Shark Trap system was developed to address future marking systems, too. If you come up with a way to mark the cards with apple juice, for example, although the specific system is not specifically listed or addressed in our study, the concept of adding a foreign substance to the card is. The Shark Trap system will detect all marking systems, past, present, and future.
If the industry doesn’t start thinking in terms of fighting technology with technology, they will remain vulnerable, and it will only get worse.
Q: What kind of technology is being employed? For example, how would the system detect a dauber and notify management?
LDG: We have several patents pending and more being drafted for filing, so I’ll have to keep my response brief. Over a three-year period, we experimented with all kinds of camera and sensor based technologies, ranging from line-scan cameras and independent lighting sources, where the camera looked up at a prismatic mirror to grab card images, to eventually settling on technologies that can hide in the palm of your hand.
The sensors are sophisticated enough to detect anomalies at the micron level (1/25,000″), so all forms of daub are relatively easy to detect. Assume that at some point a deck goes through our shuffler and two anomalies are detected, both on high cards. The next time the deck goes through the shuffler, two more high-card anomalies are detected. After the number of anomalies reaches a certain threshold, management is alerted that a statistical aberration has occurred and that the system is approaching a second alert, which would indicate a scam in progress. The reason for the first alert is to get the pit and surveillance involved in case they happen to be watching the same action and have suspicions. Once the second alert indicates a scam in progress, the system can be queried for the details of the marking system, which consists of multi-tier graphical exposés.
One way to better understand and appreciate the analysis is to look at it from a prosecutor’s standpoint. After five million scans of red-back Bee cards from ABC casino with the data all falling within a normal range, two anomalies are detected on two high cards at a particular time and date. Several anomalies are detected in subsequent rounds until most of the high cards are marked and, as you can see, a round-by-round accumulation of high-card anomalies given the sample size presents irrefutable evidence of foul play.
The system can also determine if the dauber is a pro or an amateur by measuring the size of the mark and quantity of the marking substance. It can inform management if the same daub has been played in the past, or is being played on other games, even on games in other casinos. Depending on the game, the system can even determine if more than one cheater is on the game, which would be the case if three anomalies were detected after a single round in a game where only two cards are dealt to each player. Although I’m leaving out most of the details, the system follows a very ordered and systematic approach to assessing all forms of inside and outside marked card scams. In the case of inside scams, the industry’s most dangerous marked card scam, they can be detected instantly as any deck marked previously already has several marks, perhaps dozens—there’s no progression or gradual increase in the number of marked cards.
Q: Are you familiar with SpectrumVision?
LDG: Yes, we have had an opportunity to test the unit, and although it’s not my intention to knock any product, it’s our contention that all independent testing units, devices, and systems are flawed for several reasons, but most importantly, they require human judgment to initiate a test. Someone must first make a decision to take the cards off the game for testing. Unfortunately, the consensus among experts is that most marked card scams are never suspected, let alone detected, so these solutions fail most of the time at the onset. I could provide a more detailed response, but, frankly, there’s no comparison. Independent systems simply cannot compete with Shark Trap’s real-time, tireless, invisible, reliable scientific analysis.
Q: I’m assuming that the system will detect IR marks?
LDG: Yes. By the way, the principle of infrared marks is largely misunderstood, mainly due to the advertisements of online suppliers of crooked gambling equipment. If a mark is truly in the infrared range, it’s not possible to see it with glasses or contacts—you can’t make the invisible visible. These marks require a hidden camera with special features, as cameras have a greater sensitivity to wider spectrum areas than can be seen by the human eye. If an infrared mark’s spectral curve bleeds into the visible spectrum, the mark can often be read by eye or enhanced with glasses/contacts, and when this is the case, the marks can be detected without special cameras or IR filters.
Q: Based on your research, do you have a sense of how big a threat marked card scams pose to the industry?
LDG: This is impossible to answer. What we do know, however, is that there is an undeniable increase in the online availability of the most sophisticated card marking systems the industry has ever encountered. Whether from China, Russia, Romania, or other countries, it’s not like the old days where you had to go underground and know who to contact to get the best stuff; today, just Google “marked cards” and the results will astound you. If the industry doesn’t start thinking in terms of fighting technology with technology, they will remain vulnerable, and it will only get worse. Also, according to our sources, there appears to be ample proof that marked card scams are on the rise. Several experts have been speaking about the advances in the field at different conferences, while other experts and consultants have been writing about it. For example, Willy Allison, owner of the World Game Protection Conference, in an article titled “Surveillance Report – On Your Mark”, which appeared in Global Gaming Business Magazine in January 2014, described the marked card threat as a “resurgence of marked cards” and the most recent advances as “scary new technology,” then went on to prophetically opine the following:
“We should be proactive in exploring new technology opportunities. The key to many emerging analytic technologies will be cameras and lenses. The ability for computers to analyze images from video to provide useful information is exciting. Smart camera technology could be a game-changer. Mark my words.”
Q: You stated that the system will also detect asymmetries; how does the system account for so many different back designs and ranges of variance associated with the manufacturing process and its inherent tolerances?
LDG: That’s another excellent question. After two years of experimenting with different algorithms, we finally developed an optimal approach that detects all forms of asymmetries on any back with 100% reliability. When you consider the fact that most experts estimate that 50%-70% of all casino playing cards exhibit some form of readable asymmetry, that Phil Ivey’s female partner recently claimed in a podcast that she could identify asymmetries on all cards, and that exploitable manufacturing variances can be as tiny as 1/200″ (sometimes even smaller), it appears to be another impossible problem to solve, but we did it. And we can prove it, which is our intention at this year’s G2E.
I should also mention that although most gamers understand that less-than-perfect cutting and punching of the cards can cause asymmetrical back designs, many don’t realize that the cards can be perfectly cut and punched and still exhibit asymmetries, which is why the term ‘edge sorting’ only describes a small part of the problem. The term fails to address other strategies and scams that exploit asymmetries but that are largely unknown to the industry. Frankly, our research uncovered some explosive revelations about the manufacturing of playing cards that would shock the industry, but here’s the good news: the Shark Trap system will detect, quantify, and manage any conceivable, exploitable form of asymmetry, even when the variances are so infinitesimal that only a few of the best players/cheaters in the world can read them, and even when the form of asymmetry is unknown.
