It's a sweltering August evening in Las Vegas, but it's cool inside the sports-wagering emporium at the M Resort Spa and Casino. On one of the dozens of TVs around the room, the Detroit Tigers are playing the White Sox in Chicago, and Jimmy E. has a feeling.

Jimmy, a fiftyish man in Bermuda shorts and Crocs, sips green tea as he surveys his options on three small monitors in front of him. The veteran sports bettor would rather not reveal his full name, but he's happy to share the details of his wagers. He has $8,600 on account that he can draw from. He also has a gut instinct that the next man up to bat, the Tigers' Jhonny Peralta, is going to strike out. Jimmy taps one of the touchscreens, placing a $250 bet on Peralta whiffing. The first pitch is a swing and a miss. Jimmy's a third of the way there.

The M Resort is a new casino 8 miles south of the glittering Vegas Strip. Its wood-accented sports book (gambling lingo for the casino within the casino where you can wager on sporting events) resembles a cross between an upscale sports bar and a midtier accounting firm. Some gamblers hunch over workstations identical to the one Jimmy sits at, feverishly entering fresh bets. Others wander around the floor clutching touchscreen tablet devices that they use to tap in spur-of-the-moment wagers. There are occasional cheers when a big play happens, but mostly there's a palpable tension, a sense that everyone around you is engaged in intense concentration. If the big-screen TVs were playing CNBC instead of ESPN, you might mistake the place for a trading floor.

Jimmy squints through his glasses at the screens in front of him as Peralta goes after the second pitch. Another miss. Newly updated odds appear. The count is 0 and 2, and at this point, Jimmy could hedge the bet he made 45 seconds ago by entering a counterwager—with attractively long odds—that Peralta will actually get on base. Jimmy opts to sit tight, which turns out to be a wise decision—the third pitch comes and Peralta goes down swinging. In mere seconds, $777 is credited to Jimmy's account. "My pending bets look favorable," he says. He has money down on the outcome of the game, but he has also bet on specific at-bats, the performances of various players, and which team will win a particular set of three innings.

When the White Sox's Mark Kotsay comes to the plate, Jimmy taps the touchscreen to bet $250 that Kotsay will be retired in any way other than a strikeout. If Jimmy is right, he will add $325 to his bankroll. But Kotsay swings, connects, and gets on base with a single—$250 gone forever. Jimmy looks a little drained by the accumulation of microdramas. "People say baseball is slow," he remarks as he flags down a cocktail waitress to freshen his green tea. "But it doesn't seem slow when you bet like this."

Jimmy is operating more like a daytrader than a traditional casino sports bettor, moving in and out of positions on the fly, looking for hedges, capitalizing on fleeting moments of inefficiency. And that is not an accident. Since 2009, the sports book at the M has been run byCantor Gaming, a division of the Wall Street financial services outfitCantor Fitzgerald. The resort's new bookmaker openly boasts that it's bringing the pace and style of Wall Street trading to betting on baseball, football, and basketball—and dabbling in everything from horse racing to golf.

This style of on-the-fly wagering while the game unfolds, known as in-running, didn't exist in Vegas casinos when Cantor Gaming arrived. Other big casinos in town give sports gamblers a narrow range of betting opportunities and a limited amount of time in which to bet. That appalls Lee Amaitis, formerly co-CEO of Cantor offshoot BGC Partners and now president and CEO of Cantor Gaming. "When Wall Street first opens, everybody starts trading; in this town, when a game gets going, everybodystops betting," he says. "That's the dumbest thing I ever heard. The game is the market. Why not let people bet the market?"

Amaitis is sitting at the rear of the M sports book, looking commanding in a sharp black suit and a white shirt that's open at the neck. His hair is slicked back in a style fashionable among financial execs around the time of the originalWall Street flick. "Most operators here do not have the technology or the capital investment to do what we are doing," he says in a gravelly Brooklyn accent. "They build great resorts, but they aren't figuring out how to calculate algorithms on Derek Jeter."

Cantor has generated a lot of buzz with the original technology behind its sports book operation, especially the eDeck tablets—touchscreen wireless devices that can be used to bet from anywhere in the casino. But the cornerstone of the operation is a piece of number-crunching software called Midas. It functions like the predictive computer programs that Amaitis dealt with on Wall Street: Midas acquires information, processes it, finds mathematical patterns and correlations, and uses all of that to divine the ever-shifting odds of sporting events. The system is robust enough to handle the play-by-play handicapping that keeps Jimmy E. glued to every pitch of the Tigers-White Sox game. During basketball season, things move so quickly that the bettors at the M have about eight seconds to consider a wager before the odds change.

Amaitis insists that Cantor Gaming's departure from the traditional style of sports books is the future, and some casinos are coming around to the idea. The Venetian and Palazzo, situated on the northern end of the Strip, launched Cantor's sports operations technology last fall. The Hard Rock Hotel and Casino just made a deal, and the Tropicana will be Cantorized by year's end. "This is going to become Wall Street in the desert," Amaitis says. "We are not building a sports-betting operation; we are building a trading operation."

To many people on Wall Street,gambling is a dirty word. Overlords at firms like Deutsche Bank and JP Morgan take great pains to differentiate the sober, serious profession of investing from the irresponsible, impulsive act of betting. Conversely, many traditional investment houses are eager to dismiss newfangled equity trading techniques as something closer to spins of the roulette wheel than to long-term investment strategies. "Wall Street hates being thought of as a gambling operation, but that's how it makes its money," says John Bogle, founder of the risk-averse index-fund titan Vanguard Group. "One of the big trading firms just announced that its average hold time on a stock was 11 seconds. That is not investing—it's gambling."

The stigma associated with the term became even more pronounced after the financial meltdown of late 2008. At Senate hearings this April in which Goldman Sachs execs were grilled about the collapse of their mortgage-backed securities—a collapse that threatened to bring down the entire US economy—the worst put-down that senator Claire McCaskill of Missouri could conceive of was, "You had less oversight than a pit boss in Las Vegas." Goldman CEO Lloyd Blankfein was dubbed the prince of casino capitalism by the media. In The New York Times, reporter Andrew Ross Sorkin wrote that Goldman's synthetic collateralized debt obligation was "no different than betting on the New York Yankees vs. the Oakland Athletics."

It was in the thick of this debacle, March 2009, that Cantor Gaming rolled out its M Resort sports book operation, an enterprise that flagrantly, deliberately blurs the line between investing and gambling. Why aren't they being pilloried?

For starters, Cantor Fitzgerald doesn't have to deal with stockholders, individual investors, or the press the way its higher-profile competitors on Wall Street do. Cantor is a private company that derives much of its income from being the middleman in trades between major banks. Also unlike others in the finance industry, it never begged the government for hundreds of millions of dollars to cover its bad investments. "Cantor is private, it didn't blow up, it didn't take TARP money," saysSteven Begleiter, former head of corporate strategy at Bear Stearns (who also won $1.6 million in the 2009 World Series of Poker). "I doubt Goldman Sachs would move into this business—in the current environment, it could generate unwanted newspaper headlines. But Cantor is small enough that nobody on Wall Street really worries about what they're up to."

Amaitis insists that Cantor's gambling operation is no more likely to cause a stink than the fact that Goldman Sachs bought four casinos. While it's true that the real estate division of Goldman did acquire the Stratosphere, two branches of Arizona Charlie's, and another Nevada outfit in April 2007, Goldman execs aren't telling The Wall Street Journal that they want to "turn gamblers into traders" the way Amaitis is. Pressed further on this point, Amaitis insists that he doesn't worry about the potential for stigma. "My job is not to judge the morality of why people gamble," he says. "My job is to invest in and create things that make money for my partners."

Amaitis anticipates that Cantor Gaming will control 15 percent of Nevada's sports betting this year. Judging by 2009's numbers, that would make for about $375 million in bets taken. Not money earned—just total bets taken. That figure is a pittance compared to the $62.9 billion total wagered in Vegas in the last fiscal year, and it's infinitesimal compared to the $140 trillion that passed through the hands of Cantor Fitzgerald.

The only way it makes sense for Cantor to sink what Amaitis pegs at "well over $100 million" into casino sports gambling is if he and his crew can expand the business by several orders of magnitude. And that does seem to be the plan. Amaitis views casino sports books as underdeveloped resources. Vegas casinos have traditionally regarded sports betting as an amenity for guests rather than as a serious opportunity for profit. In 2009, the average casino on the Strip made about $2.4 million a year from their sports operations compared to an average of $112 million per year from slots.

Casino executives neglect sports books because taking bets on athletic events seems like a risky proposition. They like guests to play craps, slots, and baccarat, games in which a favorable outcome for the house is all but guaranteed. From the point of view of casino owners, the result of a sporting event is incredibly uncertain, and they have no control over it. Experienced sports bettors, known as smart-money gamblers, can win far more steadily than someone playing roulette or pai gow poker. And a major upset can require a huge payout. For all these reasons, many casinos have decided that it's best to minimize their risks by posting odds that stay in line with those of the other casinos in town, keeping betting limits low, and discouraging wagers by expert gamblers. "The other bookmakers in this town are afraid to take bets, and they hate us," Amaitis says. "For me, not taking smart-money bets is like Cantor Fitzgerald not selling stock to Goldman Sachs. We do trades all day with Goldman, Deutsche, Citibank. You think those guys are stupid?"

Amaitis' scrappiness and his willingness to use technology to open untapped markets fit right in with the ethos of Cantor Fitzgerald. It is, after all, the firm that in 1999 became the first to do fully electronic US bond market trades with customers. The swagger even survived September 11, 2001, when its New York headquarters on floors 101 through 105 of One World Trade Center was destroyed. More than two-thirds of its New York staff perished, and the company's primary data center was decimated. But a skeleton crew of Cantor staffers was able to get back online and resume trading just 48 hours after the attack. Cantor pledged 25 percent of its profits over the next five years to the families of its 658 employees killed on 9/11. It gave the grieving families a total of $180 million, thanks in part to yet more lucrative technological advances like being the first firm to offer wireless trading on BlackBerries.

Play-by-Play Gambling

The NFL’s Vikings faced off against the Saints on September 9, 2010. If you were at a casino that used Cantor Gaming’s algorithm-powered betting system, you could’ve made thousands of bets as the game unfolded. Here’s a sample.

—Erik Malinowski, photo by Brian Finke

Time 14:56 left in the first quarter

Score 0-0

The Play New Orleans possesses the ball at its own 23-yard line.

The Action Betting $100 on a Saints touchdown will return $148 if it happens in this drive.

The Resolution Five plays later quarterback Drew Brees throws a 29-yard TD pass. You win.

Time 6:56 left in the first quarter

Score Saints up 7-0

The Play Minnesota is stuck with second down and 30 long yards to go.

The Action A pass seems likely. Not a bad time to bet big on a turnover. A $100 wager on an interception in this drive would yield a $783 payoff.

The Resolution The Vikings don't turn over the ball, but they do end the drive with a punt. Which means you punt $100.

Time 14:53 left in the first half

Score Saints up 7-0

The Play Minnesota has 89 yards to go if it wants to even the score.

The Action A $100 wager on a touchdown in this drive will bring in $362.

The Resolution Oof. After an epic drive, Vikings kicker Ryan Longwell completes a 41-yard field goal. That's $100 down the drain. Try not to dwell on the fact that a $100 bet on this drive ending in a field goal attempt would've made you $465.

Time 3:49 left in the first half

Score Saints up 7-3

The Play Minnesota has the ball on its own 18 and is again making risky pass attempts.

The Action A $100 wager on an interception in this drive will yield $418.

The Resolution Big risk, big reward! Three plays later, Brett Favre's pass is plucked from the air by the Saints' Jonathan Vilma.

Time 1:19 left in the first half

Score Saints up 7-3

The Play The Vikings have only 79 seconds to go 64 yards if they want to take the lead.

The Action A $100 bet on a touchdown in this drive would return $208.

The Resolution A 20-yard touchdown pass from Favre with 40 seconds to go. Translation: Ka-ching!

Cantor's latest innovation is the Midas algorithm, which is constantly being refined and fed reams of new statistical data. Even with Midas, however, sporting events are too volatile for Cantor to always end up on the right side of all the wagers it takes. But that's OK: The point isn't to nail the outcome of every contest; that's a sucker's game. There's only one sure thing in sports gambling: the standard commission, known as the vigorish, that casinos charge when they take bets. When your wins are effectively balancing out your losses, the vigorish starts to add up. In this light, Cantor's business model begins to look more like E-Trade than a conventional sports book. "The gaming business is nothing but a transaction business where you buy and sell sports instead of making a trade on the stock exchange," Amaitis says. "It's all about commission; and the more volume you do, the less risk you take."

By 2012, Amaitis predicts, his company's revenue will be $20 million. Cantor will make that money by taking far more bets than are made at present, enough that the vigorish will dwarf the income other casinos now make with their smaller (and safer) sports books. "My competitors want to hold 6 percent on a couple million dollars," he says. "I'm looking to hold 2 percent on a couplebillion dollars. Which would you rather have?"

Vegas has never seen this level of technological firepower, but Amaitis insists that it's just the beginning. He likens the Midas program's abilities to Wall Street's first foray into computerization nearly 40 years ago, before people realized how the increased speed of trading opened up many exciting—or frightening, depending on your point of view—new investment opportunities. This year, Cantor was accepting bets on the outcome of NFL games months before the season started. But Amaitis wishes he could take more granular bets before kickoff. After all, more opportunities to bet means more commissions for Cantor. He dreams of being able to offer programmed sports betting that will allow gamblers to put in bid orders, just as you would with an electronic stock purchase. Before the game begins, you'll be able to set your account so that a bet will automatically be placed if, for example, a team is ever a six-point underdog. You'd also be able to set it up to place a bet on the other side, say, if that number drops from six to four. Once your bids are in, you won't even have to watch the game. "I am salivating for that," Amaitis says.

The ideas that led to Midas were born almost 15 years ago, over a lavish dinner at Cliveden House, an elegant manor outside London. Amaitis was supping with Andrew Garrood, a mathematician turned derivatives trader he'd just met in a private suite at theAscot Racecourse. They were both there to watch a leg of the Royal Ascot, one of the most important horse-racing events in the UK. But they spent much of their time discussing the resemblance between brokers and bookies.

The two must have made for quite a contrast. Garrood is a genteel Englishman with an upper-crust accent. The street-smart Cantor exec, then head of the company's European brokerage operation, was deemed coarse and uncouth by London standards. (He was the target of a harassment suit by a former employee, and the British tabloids dubbed him the Brooklyn Bruiser.)But they were both excited by the idea of a Cantor-backed sports-betting enterprise.

The ubiquity of gambling in the UK was an eye-opener for Amaitis. There are legal betting shops on nearly every corner, and having a punt on a footy match—a wager on a soccer game—is as common as having a pint at the pub. "I'd visit clients, and they'd be on the phone with Ladbrokes making bets," Amaitis remembers, referring to the UK's popular chain of bookies. "I went back to my firm and told them that we should be in the bookmaking business."

Garrood knew enough about gambling to tell Amaitis that it had a great deal of overlap with the sort of work he did trading derivatives for Japan's Sanwa Bank. "There is absolutely no difference between making prices for currency or making lines for sporting events," Garrood says. "You don't know if the market will go up or down any more than you know who will win the next horse race."

The two chewed over the idea with their meal. "I was initially interested in that space as a client," Garrood says. "Lee was interested in it as a business opportunity." But by fall 1999, when Garrood read in theFinancial Times of London that Cantor was actually going to get into the bet-taking business, he immediately phoned Amaitis and said he wanted to help run the business.

By March 2000, he was on the Cantor payroll. Garrood was tasked with gathering the data and creating software that could calculate the in-running odds for an online sports-wagering operation. He developed the algorithms, and an IT team turned them into code. In 2004, the UK site Cantor Spreadfair was launched. It was a peer-to-peer sports gambling site, allowing users to set their own bets, which would be taken by other players (or by Cantor). The site allowed for spread betting, a volatile form of gambling in which the payoffs are based on how closely you predict the final point difference between two teams. They also set up a site called Cantor Index for "financial fixed odds"—allowing people to bet on the moment-to-moment changes in the stock market—as well as financial spread betting. (Such activities are illegal in the US but not in the UK.)

Although Cantor's UK gaming scheme was a moderate success, Amaitis viewed it as a proof of concept. He used it to pitch Vegas casinos on the prospect of allowing him to lease their sports books and introduce in-running to American gamblers. "Cantor Gaming approached everyone on the Strip and everyone turned them down; you've got a lot of control freaks in this town," says Joseph Magliarditi, now CEO of the Hard Rock Hotel but previously COO of the M Resort. "Lee approached me, and I got it immediately. Discussions went on for a year and a half, but I knew I wanted to bring Cantor to M."

Amaitis also presented his UK setup to theNevada Gaming Control Board as evidence that the wireless component of his master plan was sound and secure. "When we applied for our gaming license, Nevada tried cracking into our UK system a couple hundred times," Amaitis says. "They bounced in from IP addresses in the States and off of other IP addresses around the world. We always blocked them."

Cantor Gaming signed a deal with the M Resort in July 2008, while the $1 billion property was still under construction. As the ink dried on the deal, Garrood had about 10 months to create the algorithms for a computer system that could establish airtight betting lines and calculate odds on the fly. Inefficiencies, he knew, would be rooted out by bettors and cost Cantor dearly.

Working out of a skyscraper in the Canary Wharf business district of London, Garrood found that he was able to port some elements of Cantor's British betting site into the nascent Midas, but he would need all-new information on which to base his algorithms. (After all, Americans wouldn't want to bet on rugby or cricket.) "I had the perfect sauce recipe," he says. "But I needed different ingredients."

Amaitis made things easier for Garrood by purchasingLas Vegas Sports Consultants. This was the first time many in Vegas noticed the fledgling Cantor operation; LVSC had been the city's oddsmaker of choice since 1982, and now Cantor owned its vast library of research and stats.

The LVSC archive forms the bulk of the content that Midas analyzes. Garrood worked with a staff of two to build its algorithms, find patterns, and categorize the data. "Statistics carry you most of the way," he says. "The beautiful thing is that you can place values on predictingeverything. If the Patriots are playing the Colts, and the Patriots want to go for it on fourth and two from their own 28-yard line with two minutes left in the fourth quarter, people think it's ridiculous. It doesn't happen that often, but it does happen"—and it did happen when the teams faced off last November. "You can use those times—combined with thousands of correlating factors—to prove the statistical likelihood of the Patriots making a first down. And you can put a price on it."

In a town where most sports books still rely on back-of-the-envelope calculations and teams of guys in low tech war rooms, Garrood's mathematics give Cantor a level of comfort in its odds and point spreads that other Vegas bookmaking operations simply can't match. He also learned some things in the course of Midas' creation that go against the conventional wisdom among Vegas bookies and bettors. For example, when a star player is injured, a casino will sometimes move the line on a game by a point or two. But decades of stats tell Garrood that a star—who, let's face it, will be replaced by another high- caliber professional athlete—is usually worth a negligible amount in the spread. "Another old wives' tale is that teams perform better after they have a week off and the players are well rested," Garrood says. "That's absolute rubbish. So is the belief that a team like the Miami Dolphins can't win when they play in cold climates."

These are the sorts of misconceptions that can help Cantor make money. If you're being given 2-to-1 odds that the Dolphins will make a first down in Green Bay, and you have 30 seconds in which to decide whether that is a worthwhile bet, you will be basing your decision on what happened over the past couple of plays—a woefully unreliable sample—or on a widely held belief that a team from the sunny South will fare poorly in the frigid North. You'll be betting with your gut; Cantor bets with Midas.

Spend enough time at the M Resort sports book and you will hear gamblers talking about Midas in hushed, reverent tones. Invariably, they gesture toward a locked wooden door at the front of the room, where they imagine the supercomputer resides. Listen to them for a few minutes and you get the sense that Midas is a sort of bookie HAL 9000, chomping on cigars, dicing numbers, and dictating point spreads from a jewel-encrusted silicon throne.

In reality, Midas is 120,000 lines of code, a piece of predictive software that largely resides in a nondescript office building about 14 miles north of the M. It was here that the final step of Midas' development was overseen by Cantor's CTO, Sunny Tara, who has since left to form his own software company. Prior to joining Cantor Gaming, the diminutive engineer had been the senior director of enterprise architecture and services at Harrah's Entertainment, the largest gaming company in the world. At Cantor, he and his team of about 15 took the statistical matrices Garrood created and made them into a fully functional piece of enterprise software.

Dressed in jeans and a light blue button-down shirt, dark hair combed neatly, Tara speaks with the patient earnestness of a comp-sci professor addressing undergrads. "The computer gets the info, the algorithm churns millions of combinations of calculations, it feeds out a price, and that gets displayed on the customer's terminal in just a few milliseconds," Tara says. "It is very similar to a Wall Street model."

He makes it sound so simple. But it was a logistical nightmare. "The challenge was to do multiple games, simultaneously, with different rules and criteria and various inputs," Tara says. "Yet it's the same Midas doing all of them at the same time. The code had to accommodate apples and oranges. But we were able to leverage what Andrew already had in England and what Cantor already had for its financial markets."

The heart of the operation exists inside a 2,000-square-foot back room that I was not allowed to enter, where there are racks of servers and a handful of Panasonic flatscreens on the wall being monitored by a control-room staff. It's connected to the M Resort and other Cantor facilities by a dedicated 50-megabit Ethernet circuit.

To illustrate how it works, take an August game that pitted the Atlanta Braves against the San Francisco Giants. With the score tied 2-2 in the bottom of the sixth, the Braves' Eric Hinske comes up to bat, and the money line on his team is minus 117, which means that for every $117 you bet on the Braves, you can win only $100. Though the score is tied at this moment, Midas believes that the Braves are a favorite to win.

Then the pitch comes in—an 82-mph changeup— and Hinske blasts it out of the park for a home run. Midas finds out about this a split second after the spectators at Turner Field. Cantor commissions several data processing services to send live game info directly to its control room; whichever feed arrives first is the one that Midas takes. It absorbs the new developments and correlates them with similar events that have happened in the past. The money line dips to minus 327. Winning $100 on the Braves will now mean putting $327 at risk. A play later, when the Braves' Rick Ankiel grounds out, the process repeats itself, and winning $100 on the Braves now requires a bet of $283.

It all happens in seconds, and Midas may well be crunching the numbers for other games, in other sports, at the same time. After the day's games are over, Cantor programmers scour the stats for mistakes or new correlations that could make Midas stronger for the next day's matchups. "Every game is a new market," Tara says.

Amaitis acknowledges that Midas still misses some factors. "There is a thing called momentum, and a computer doesn't understand how to calculate that," he says. Not yet, anyway. "Over a series of years and results and matchups, Midas will come to understand momentum," Amaitis says. "Three years from now, it will be 1,000 times smarter than it currently is."

Mike Colbert, Cantor Gaming's 30-year-old sports book director, has been around gambling all of his adult life. He's taken bets at Caesars Palace on the Strip and at a dilapidated joint called the Plaza. "Prior to coming here, I would get sick to my stomach when bets came in," he says. "You're booking on your heels the entire time. That was before I had the computer. I know how Midas was created, and I trust the numbers it generates."

Still, just because Cantor Gaming is reasonably confident in the forecasts it gets from Midas doesn't mean it's not exposed to potential losses. If betting is too lopsided—say, if everyone in the M bets that a field goal attempt will be successful and no one bets against it—then Cantor doesn't take in enough losing bets to pay the winners. Like all bookies, Colbert adjusts odds and point spreads in an effort to balance the wagers he takes, making sure there's plenty of money riding on both potential outcomes of a game. But sometimes, when the circumstances are right and an edge seems to be there, Cantor Gaming isn't above taking a speculative position the same way that firms on Wall Street do. Such was the case with the 2010 Super Bowl. Before the playoffs were even over, Midas determined that the Indianapolis Colts were at minus 2 to win it all, meaning they were a two-point favorite over the opposition. But Colbert guessed that the other sports books in town would make the Colts a four-point favorite or better. If he posted the line that Midas told him to post, it was certain to bring in a lot of bets on the favored Colts. But Cantor would have no advantage. So when Colbert opened his line, he had Indianapolis at minus 3, knowing that this point spread would still draw wagers from savvy bettors while giving him one point of leeway over the outcome that Midas actually predicted. (A single point might not seem like much, but from a statistician's point of view, it's a fairly healthy cushion.)

Serious professional sports bettors went nuts for the M Resort's line on the Super Bowl. Bill Krackomberger, a big-boned guy in a peak-billed Kangol who sits front and center in the M sports book on almost every game day, wagered $25,000 on the Colts as soon as the number was posted. When the point spreads from rival casinos appeared, they were even more extreme than Colbert anticipated—they had the Colts at minus 5.5 and minus 6. Colbert moved the Cantor line to 4 and eventually to 4.5. Again, not because he thought it was the right number but because he could still attract all the Colts action he wanted while giving himself a statistical cushion. Later Krackomberger cursed himself for not betting more at minus 3.

Adam Meyer, a professional handicapper from Florida, flew in and placed a series of wagers that totaled $1.2 million; his betting scheme included $700,000 on the Colts to have the lead at the end of the first half; the sum, he says, is far beyond what most other books in Vegas would allow him to wager. Meyer grabbed one of the M Resort's wireless tablets and, over the course of the game, placed other bets in the comfort of the casino's cocktail lounge.

The M garnered 10 percent of Nevada's $82.7 million in Super Bowl wagering, and much of that money was riding on the Colts winning by at least 5 points. Midas had actually predicted that the Colts wouldn't do that well, and they ended up doing far worse—they lost by 14. Which means that Cantor cleaned up. However, that big win has done nothing to discourage sharp bettors from trying to beat Cantor. Meyer is already in the process of making a series of $200,000 bets on various NFL games this season. And Krackomberger says a friend of his wrote a computer program that was designed to outthink Midas on in-running wagers at baseball. "That kid is the sharpest of the sharp," Krackomberger muses. "And after a few months, he waved the white flag. He just couldn't win."

Cantor Gaming's short-term strategy is clear: Move into more Vegas casinos and attract more gamblers to its sports books. But ultimately, the company is thinking beyond casinos, beyond Vegas, and beyond sports betting. Someday soon, Amaitis hopes, Cantor will take bets on just about anything that a price can be attached to, and it will take them from just about anywhere in Nevada.

For starters, Amaitis wants to take advantage of Nevada mobile-gaming laws to allow anyone in the state with a smartphone or Internet connection to make in-running bets with Cantor. The next step after that is "event wagering," giving odds on anything with an uncertain outcome. "We would offer bets on elections," he says. "We would do the Academy Awards andAmerican Idol. We can build algorithms to make lines oneverything."

Cantor Fitzgerald actually tried to dip a toe into these waters a few years ago with a separate division called Cantor Exchange. Its first market was going to be Hollywood futures, allowing people to trade on movie box-office results. The exchange's launch was torpedoed earlier this year by objections from the Motion Picture Association of America, but Amaitis says it might return in some other form.

Cantor Gaming is also laying the groundwork to allow gamblers at its sports books to wager on movements within the stock market, just like the Cantor Index in the UK. Amaitis acknowledges that the current laws and the recent economic meltdown make this an unsavory proposition at present, but he wants to be ready for a day when bettors can gamble on the performance of Microsoft and AMD just as they currently wager on the Yankees or the Athletics. An online futures exchange where you can place wagers on the performance of stocks without actually owning them—the ultimate synthetic derivative. What could possibly go wrong?

Michael Kaplan (mkap@nyc.rr.com)wrote about Next Media Animation in issue 18.09.