The MIT Blackjack Team was a group of students and ex-students from the Massachusetts Institute of Technology, Harvard University, and other leading colleges who used card counting techniques and more sophisticated strategies to beat casinos at blackjack worldwide. The MIT Blackjack Team was made up of a collection of students from Boston area schools such as the Massachusetts Institute of Technology (MIT), Harvard Business School, and Harvard University. The goal of the team was simple and led to one of the gambling world’s biggest scandals of all time. Directed by Robert Luketic. With Jim Sturgess, Kate Bosworth, Kevin Spacey, Aaron Yoo. '21' is the fact-based story about six MIT students who were trained to become experts in card counting and subsequently took Vegas casinos for millions in winnings.
BOSTON (AP) — In the early years of the MIT Sloan Sports Analytics Conference, a couple hundred math whizzes gathered on the school’s campus hoping for a breakthrough that would make them the next Bill James.
As the event and the field grew, they aspired to professional front office jobs, like Theo Epstein or Daryl Morey.
Now the revered names at the cavernous convention center in Boston’s Back Bay are people like sports bettor-turned Dallas Mavericks number-cruncher Haralabos Voulgaris and other current or former professional gamblers who are using analytics to gain an edge as legal sports betting spreads across the United States.
“Math has always been in gambling to some degree,” said Jeff Ma, a professional bettor and a former member of the MIT blackjack team, who moderated this year’s gambling panel.
“You’re just seeing it go more mainstream,” since the U.S. Supreme Court lifted a ban on sports gambling, Ma said. “It’s more acceptable to talk about now.”
What started a dozen years ago with 175 people listening to a handful of panels in MIT classrooms has now grown into a two-day event with 3,500 attendees from 44 states and 33 countries, including more than 130 teams across the U.S. pro leagues and European soccer.
In addition to the panels featuring everyone from rapper Meek Mill on social justice to Olympic curling gold medalist Tyler George, there are workshops, research competitions, hackathons and live broadcasts and podcasts.
And, now that sports betting is possible — if not quite legal — nationwide, the self-proclaimed “dork-a-palooza” is increasingly turning its attention to gambling.
Perhaps not since the publication of Michael Lewis’ book “Moneyball,” which detailed Oakland Athletics general manager Billy Beane’s use of analytics to build a roster for his underfunded team, has an event so energized the statistical community.
“Stats and sports have always been associated for pretty much since sports have been around,” DraftKings founder Jason Robins said, crediting both James and Beane for popularizing analytics.
“They really popularized the notion that you could gain an advantage. In their case, it was on the field,” Robins said. “It was a natural transition to the viewer, who is not going to get the chance to be the general manager of a baseball team, to figure out outlets for that. And the most natural outlets are going to be fantasy and betting.”
This year’s conference included the first poker panel, hosted by fivethirtyeight.com founder Nate Silver and including 2017 World Series of Poker Main Event winner Scott Blumstein. There were at least two discussions of fantasy sports, one on predicting college basketball’s March Madness outcomes and one on how to keep gamblers from fixing games.
There was also a full panel on sports gambling, held in an enormous ballroom named for James.
Like the analytics pioneer, statistically inclined team executives like Beane, Epstein and Morey (a co-founder of the conference) faced resistance from skeptical old-timers.
But gamblers are open-minded to anything that will give them an edge.
“Everybody wants to use analytics” in betting, said Rufus Peabody, a professional bettor, podcast host and the founder of Massey-Peabody Analytics. “You have a lot of smart people with access to the data. You have to be constantly improving just to keep up.”
That’s where the Sloan conference comes in.
Over two days, college math majors rubbed elbows with team and tech executives looking for fresh ideas — and talented minds to implement them.
A research paper competition included topics like “How much do coaches matter?” and “The Eagle Takes Flight: The Economist’s Hierarchical Bayesian Approach to Golf.” At the root of all the math is the same question: How can you find an edge, whether it is in a free agent market or an NHL faceoff circle.
For the Oakland A’s of the late 1990s, as chronicled in “Moneyball,” it was on-base percentage. For sports bettors, it can be the over/under line in a single baseball game or even the final few innings.
Peabody said gamblers can have an edge in two ways: You can have better information, or be better at processing the information. Gamblers used to scour team blogs or local papers for injury information, but the internet has largely leveled the field.
So bettors are looking for a way to profit on the latter.
And there’s more to it than just crunching numbers.
“It’s one thing to be able to get the information or do the analytics, just crunch the numbers. But you have to actually interpret it. That’s as much of an art as anything,” said Robins, whose daily fantasy sports company moved into sports betting when the nationwide ban was lifted.
“So I think that gap is still there among people who just have a good instinct for the game and understand, based on some data or information, what to make of it.”
The MIT Blackjack Team was a group of students and ex-students from the Massachusetts Institute of Technology, Harvard University, and other leading colleges who used card counting techniques and more sophisticated strategies to beat casinos at blackjack worldwide. The team and its successors operated successfully from 1979 through the beginning of the 21st century. Many other blackjack teams have been formed around the world with the goal of beating the casinos.
Blackjack can be legally beaten by a skilled player. Beyond the basic strategy of when to hit and when to stand, individual players can use card counting, shuffle tracking, or hole carding to improve their odds. Since the early 1960s, a large number of card counting schemes have been published, and casinos have adjusted the rules of play in an attempt to counter the most popular methods. The idea behind all card counting is that, because a low card is usually bad and a high card usually good, and as cards already seen since the last shuffle cannot be at the top of the deck and thus drawn, the counter can determine the high and low cards that have already been played. They thus know the probability of getting a high card (10,J,Q,K,A) as compared to a low card (2,3,4,5,6).
In 1979, six MIT students and residents of the Burton-Conner House at MIT taught themselves card-counting. They traveled to Atlantic City during the spring break to win their fortune. The group went their separate ways when most of them graduated in May of that year. Most never gambled again, but some of them maintained an avid interest in card counting and remained in Cambridge, Massachusetts. Two of them, J.P. Massar and Jonathan, offered a course on blackjack for MIT's January, 1980 Independent Activities Period (IAP), during which classes may be offered on almost any subject.
In late November 1979, Dave, a professional blackjack player contacted one of the card-counting students, J.P. Massar, after seeing a notice for the blackjack course. He proposed forming a new group to go to Atlantic City to take advantage of the New Jersey Casino Control Commission's recent ruling that made it illegal for the Atlantic City casinos to ban card counters. Casinos instead have to take other countermeasures like shuffling the cards earlier than normal, using more decks of cards, or offering games with worse rules to destroy the advantage gained by counting—even though these all negatively impact the non-counter as well.[1]
The group of four players, a professional gambler, and an investor who put up most of their capital ($5,000), went to Atlantic City in late December. They recruited more MIT students as players at the January blackjack class. They played intermittently through May 1980 and increased their capital four-fold, but were nonetheless more like a loose group sharing capital than a team with consistent strategies and quality control.
In May 1980, J. P. Massar, known as 'Mr. M' in the History Channel documentary, overheard a conversation about professional blackjack at a Chinese restaurant in Cambridge. He introduced himself to the speaker, Bill Kaplan, a 1980 Harvard MBA graduate who had run a successful blackjack team in Las Vegas three years earlier. Kaplan had earned his BA at Harvard in 1977 and delayed his admission to Harvard Business School for a year, when he moved to Las Vegas and formed a team of blackjack players using his own research and statistical analysis of the game. Using funds he received on graduation as Harvard's outstanding scholar-athlete, Kaplan generated more than a 35 fold rate of return in fewer than nine months of play.[2]
Kaplan continued to run his Las Vegas blackjack team as a sideline while attending Harvard Business School but, by the time of his graduation in May 1980, the players were so 'burnt out' in Nevada they were forced to hit the international circuit. Not feeling he could continue to manage the team successfully while they traveled throughout Europe and elsewhere, encountering different rules, playing conditions, and casino practices, Kaplan parted ways with his teammates, who then splintered into multiple small playing teams in pursuit of more favorable conditions throughout the world.
After meeting Kaplan and hearing about his blackjack successes, Massar asked Kaplan if he was interested in going with a few of Massar's blackjack-playing friends to Atlantic City to observe their play. Given the fortuitous timing (Kaplan's parting with his Las Vegas team), he agreed to go in the hopes of putting together a new local team that he could train and manage.
Kaplan observed Massar and his teammates playing for a weekend in Atlantic City. He noted that each of the players used a different, and overcomplicated, card counting strategy. This resulted in error rates that undermined the benefits of the more complicated strategies. Upon returning to Cambridge, Kaplan detailed the problems he observed to Massar.
Kaplan said he would back a team but it had to be run as a business with formal management procedures, a required counting and betting system, strict training and player approval processes, and careful tracking of all casino play. A couple of the players were initially averse to the idea. They had no interest in having to learn a new playing system, being put through 'trial by fire' checkout procedures before being approved to play, being supervised in the casinos, or having to fill out detailed player sheets (such as casino, cash in and cash out totals, time period, betting strategy and limits, and the rest) for every playing session. However, their keen interest in the game coupled with Kaplan's successful track record won out.
The newly capitalised 'bank' of the MIT Blackjack Team started on 1 August 1980. The investment stake was $89,000, with both outside investors and players putting up the capital. Ten players, including Kaplan, Massar, Jonathan, Goose, and 'Big Dave' (aka 'coach', to distinguish from the Dave in the first round) played on this bank. Ten weeks later they more than doubled the original stake. Profits per hour played at the tables were $162.50, statistically equivalent to the projected rate of $170/hour detailed in the investor offering prospectus. Per the terms of the investment offering, players and investors split the profits with players paid in proportion to their playing hours and computer simulated win rates. Over the ten-week period of this first bank, players, mostly undergraduates, earned an average of over $80/hour while investors achieved an annualized return in excess of 250%.
The team often recruited students through flyers and the players' friends from college campuses across the country. The team tested potential members to find out if they were suitable candidates and, if they were, the team thoroughly trained the new members for free. Fully trained players had to pass an intense 'trial by fire,' consisting of playing through 8 six-deck shoes with almost perfect play, and then undergo further training, supervision, and similar check-outs in actual casino play until they could become full stakes players.
The group combined individual play with a team approach of counters and big players to maximize opportunities and disguise the betting patterns that card counting produces. In a 2002 interview in Blackjack Forum magazine,[3] John Chang, an MIT undergrad who joined the team in late 1980 (and became MIT team co-manager in the mid-1980s and 1990s), reported that, in addition to classic card counting and blackjack team techniques, at various times the group used advanced shuffle and ace tracking techniques. While the MIT team's card counting techniques can give players an overall edge of about 2 percent, some of the MIT team's methods have been established as gaining players an overall edge of about 4 percent.[citation needed] In his interview, Chang reported that the MIT team had difficulty attaining such edges in actual play, and their overall results had been best with straight card counting.
The MIT Team's approach was originally developed by Al Francesco, elected by professional gamblers as one of the original seven inductees into the Blackjack Hall of Fame. Blackjack team play was first written about by Ken Uston, an early member of Al Francesco's teams. Uston's book on blackjack team play, Million Dollar Blackjack, was published shortly before the founding of the first MIT team. Kaplan enhanced Francesco's team methods and used them for the MIT team. The team concept enabled players and investors to leverage both their time and money, reducing their 'risk of ruin' while also making it more difficult for casinos to detect card counting at their tables.
The MIT Blackjack Team continued to play throughout the 1980s, growing to as many as 35 players in 1984 with a capitalization of as much as $350,000. Having played and run successful teams since 1977, Kaplan reached a point in late 1984 where he could not show his face in any casino without being followed by the casino personnel in search of his team members. As a consequence he decided to fall back on his growing real estate investment and development company, his 'day job' since 1980, and stopped managing the team. He continued for another year or so as an occasional player and investor in the team, now being run by Massar, Chang and Bill Rubin, a player who joined the team in 1984.
The MIT Blackjack Team ran at least 22 partnerships in the time period from late 1979 through 1989. At least 70 people played on the team in some capacity (either as counters, Big Players, or in various supporting roles) over that time span. Every partnership was profitable during this time period, after paying all expenses as well as the players' and managers' share of the winnings, with returns to investors ranging from 4%/year to over 300%/year.
In 1992, Bill Kaplan, J.P. Massar, and John Chang decided to capitalize on the opening of Foxwoods Casino in nearby Connecticut, where they planned to train new players. Acting as the General Partner, they formed a Massachusetts Limited Partnership in June 1992 called Strategic Investments to bankroll the new team. Structured similar to the numerous real estate development limited partnerships that Kaplan had formed, the limited partnership raised a million dollars, significantly more money than any of their previous teams, with a method based on Edward Thorp's high low system. It involved three players: a big player, a controller, and a spotter. The spotter checked when the deck went positive with card counting, the controller would bet small constantly, wasting money, and verifying the spotter's count. Once the controller found a positive, he would signal to the big player. He would make a massive bet, and win big. Confident with this new funding, the three general partners ramped up their recruitment and training efforts to capitalize on the opportunity.
Over the next two years, the MIT Team grew to nearly 80 players, including groups and players in Cambridge, New York, New Jersey, Pennsylvania, California, Illinois, and Washington. Sarah McCord, who joined the team in 1983 as an MIT student and later moved to California, was added as a partner soon after SI was formed and became responsible for training and recruitment of West Coast players.
At various times, there were nearly 30 players playing simultaneously at different casinos around the world, including Native American casinos throughout the country, Las Vegas, Atlantic City, Canada, and island locations. Never before had casinos throughout the world seen such an organized and scientific onslaught directed at the game. While the profits rolled in, so did the 'heat' from the casinos, and many MIT Team members were identified and barred. These members were replaced by fresh players from MIT, Harvard, and other colleges and companies, and play continued. Eventually, investigators hired by casinos realized that many of those they had banned had addresses in or near Cambridge, and the connection to MIT and a formalized team became clear. The detectives obtained copies of recent MIT yearbooks and added photographs from it to their image database.
With its leading players banned from most casinos and other more lucrative investment opportunities opening up at the end of the recession, Strategic Investments paid out its substantial earnings to players and investors and dissolved its partnership on December 31, 1993.
After the dissolution of Strategic Investments, a few of the players took their winnings and split off into two independent groups. The Amphibians were primarily led by Semyon Dukach, with Dukach as the big player, Katie Lilienkamp (a controller), and Andy Bloch (a spotter). The other team was the Reptiles, led by Mike Aponte, Manlio Lopez and Wes Atamian. These teams had various legal structures, and at times million dollar banks and 50+ players. By 2000 the 15+ year reign of the MIT Blackjack Teams came to an end as players drifted into other pursuits.
In 1999, a member of the Amphibians won at Max Rubin's 3rd Annual Blackjack Ball competition. The event was featured in an October 1999 Cigar Aficionado article, which said the winner earned the unofficial title 'Most Feared Man in the Casino Business'.[4]
Several members of the two teams have used their expertise to start public speaking careers as well as businesses teaching others how to count cards. For example: