Tony McCoy found his bookie through a friend of a friend. It was October 2011, four weeks into football season. He’d spent all of last season testing a new betting system he’d created, and it had proven sound. Now he was ready to put money on it. He’d never had a bookie before, never spoken to one on the phone. He figured there would be rules, but he didn’t know what they were. Before he made the call, he wrote out all the bets he wanted for the week, four in total, and gave himself a pep talk. People do this all the time, he thought.

It didn’t help his nerves.

“It was crossing a line,” he tells me.

It’s a Sunday afternoon at the McCoy family home, where the living room is cluttered with Legos, coloring books, and other kiddie toys. Tony’s wife Jen and their two-year-old son are doing a puzzle on the floor. Tony and I sit at a round oak table a few feet away. While he talks to me, his gaze remains mostly on the 49ers/Giants game playing on a wall-mounted flat screen. Tony has money on San Francisco.

During a commercial for Allstate Insurance, he looks in my direction. “Technically,” he says, “it’s all illegal. Playing poker online, betting your friend 20 bucks at work — it’s all illegal. Betting with a bookie, though, was a line I didn’t want to cross. But I just finally said, ‘Aw, screw it.’ And I called him.”

Tony locked himself in the upstairs bathroom to make the call. He turned on the bathroom fan to drown out his son’s playful shrieks and the sound of his wife reprimanding the boy for jumping off the couch.

“The guy answers the phone right away,” Tony says. “He sounds exactly like you would expect a bookie to sound, like a cranky old Italian guy with one of those newsboy caps.”

The conversation went as follows:

Tony: “A friend of mine said I should call you.”

Bookie: “All right.”

Tony: “Yeah, my friend Steve.”

Bookie: “Who’s that?”

Tony: “Steve. He said he knows you pretty good.”

Bookie: “What’s his last name?”

Tony: “Brown. Steve Brown.”

Bookie: “White guy? Kinda tall?”

Tony: “Right, right. That’s him.”

Bookie: “All right, I know him. But who are you?”

Tony: “I’m a friend of his.”

Bookie: “What kind of friend?”

Tony: “Uh, pretty good friends, I guess.”

Bookie: “Have you met him?”

Tony: “Who? Steve? Well, yeah, I’ve met him.”

Bookie: “You play poker with him?”

Tony: “No, I never played poker with him, but I know he plays.”

Bookie: “All right, well, what do you want?”

Tony: “Um, Steve tells me you might be able to help me out for this Sunday.”

Bookie: “What about Sunday?”

Tony: “Well, I’m looking to, uh, bet on a game on Sunday.”

[Long pause.]

Bookie: “And you’re Steve’s friend, huh?”

Tony: “Yeah.”

Bookie: “You been knowing him a long time?”

Tony: “Yeah, I guess so.”

Bookie: “Well, he’s okay, so I guess you’re all right. Tell him what you want, then give him the money, and he’ll call me.”

Tony did as the bookie instructed, driving to Steve’s house in North Park to drop off the envelope with $220 cash (including $20 in “juice” — a commission paid to bookies for losing bets). His picks were written on the outside.

“I had $50 each on Washington plus-one over St. Louis, Kansas City plus-three over Minnesota, and Tennessee plus-one over Cleveland. Plus a $50 parlay on all three games. The parlay pays six to one, but you have to win all three games to win the bet. All together, I had $200 on the line. Get this — Steve actually laughed at my picks. I can’t say I blamed him, because, honestly, I didn’t like the picks all that much myself. The system liked them, though, and that’s what counts. Steve doesn’t pick games any better than I do.”

Sunday came. Tony won all three games.

“Steve doesn’t laugh at me anymore,” he says.

The bookie settles up with clients on Tuesdays. So on Tuesday, October 4, Tony followed the bookie’s directions to an Arby’s parking lot in Oceanside. He figured the guy wouldn’t be too happy about having to hand over so much cash at their first meeting. Jen, on the other hand, was more worried about the sketchy nature of the meet-up.

“An Arby’s parking lot?” she says, looking up at us from her spot on the floor. “I was afraid he’d get jumped or something. I don’t know.”

Tony laughs. “It wasn’t anything like that.”

The bookie turned out to be a shabby, pale man sporting a comb-over and a pair of ill-fitting slacks. The two shook hands. The bookie pulled a roll of bills from his pocket, peeling off four one-hundred dollar bills, two twenties, and a ten. If the bookie was bothered by having to pay out that amount, it didn’t show. Tony wondered if the bookie was a middleman, if maybe someone scarier would come along in the event you didn’t pay up. This guy didn’t appear to be the kind who would, or could, break your legs.

“Not that I’ll ever know,” Tony says. “I never bet what I don’t have.”

Over the next six weeks, Tony won $90, won $150, lost $30, won $90, won $25, and lost $15. At the start of this week’s games, he’s up $690, not including the $220 he started with. So, the money he has on the line this weekend is all profit.

Even so, he says, “Every time I win, I’m surprised. But I really do think it’s a good system.”

∗ ∗ ∗

The whole thing started with the promise of a trip to Italy for Tony and Jen’s tenth anniversary. Something they’ve been dreaming about since they were first married. Seven years later, they found themselves with only three years to save for a $10,000 trip.

“Personally, I like to gamble,” Tony says. “It’s fun. But it also helps to have a goal. When I was playing online poker a few years ago, I told my wife that anything I win, we’re going to Italy with it.”

It didn’t work out that way. But not because he didn’t win any money.

“That year, I won $2600 playing poker online. But when I took the money out, we made the mistake of putting it in the checking account,” he says. “Slowly, but surely, we used it on bills and going out to dinner and stuff. It’s long gone.”

Now he’s trying again, this time gambling on football. He keeps his winnings in a baggie at the back of the freezer.

This week, he has 13 bets going, five with the bookie at $50 each (including a teaser that pays 2.8:1), five with friends and work buddies (for a total of $120), and four small bets online (where he’s trying out a second system for college football), totaling $30.

“It’s good to have your bets spread out. You don’t want all your money on one or two games,” he says.

All told, if he wins everything, he’ll be up $490 for the week. Losing it all will mean $425 gone.

If Jen weren’t here now, listening and occasionally piping in with words of praise for her husband’s methodology and precision, I wouldn’t believe it when Tony tells me she supports the $400 gambling risks he takes each week. Their 120-square-foot den certainly doesn’t read “we’ve got lots of money to blow.” But do they?

When I ask, both Tony and Jen laugh.

“I’m a regular working guy with bills and a mortgage,” Tony says. “We’re not quite paycheck-to-paycheck, but not a whole lot better than that, either. I certainly can’t afford to be betting the grocery money or anything like that. And I don’t. Ever.”

“He wouldn’t,” Jen pipes in. “He’s not that guy.”

She goes on to explain that last year, when he agreed to take her to the horse races at Del Mar, he wouldn’t let her bet more than a dollar or two on each race. And on a recent trip to Vegas (their first), he preferred sightseeing to casinos; she had to talk him into putting $10 into the slot machines.

I tell them I’m confused. Here’s a guy who has $400 riding on the outcome of 13 football games today. And yet he’s not a gambler?

Tony says he differentiates between betting on things like poker and sports, where a person can manipulate the odds to be in their favor, and games of chance like roulette, craps, and slot machines, where whether you win or lose is “one hundred percent luck.” Winning and losing in sports, he estimates, is 40 percent luck on any given day. The rest of it is that “the system works.” But even a winning system is not enough. It’s also important to “remain unemotional” when making picks, “never bet more than you can afford,” keep an accurate track of wins and losses, and “never, ever ‘chase.’”

“You can have six good weeks in a row, winning $50 every week,” he says. “That might not sound impressive, but that’s $300 to the good. Then, let’s say you lose $100 the next week. A good gambler stops right there, and he’s still up $200. But if you’re a chaser, you bet $100 on the Sunday-night game, trying to get back to even for the week. You lose that, so then you bet $200 on the Monday-night game, still trying to get even. You lose that, too. And now, instead of being up $200 for seven weeks, you’re down $100. That’s chasing.”

What keeps Tony from being “that guy?”

When he loses, his natural inclination is to lower the amount of his bets and to bet less often. “I’m also really big into record-keeping,” he says. “I’d say, as a gambler, my best quality is record-keeping. That and bankroll management.”

Bankroll management, he explains, is where a bettor starts with a given amount of money and allows himself only a small percentage of that total for each bet. The starting size of the bankroll doesn’t matter, as long as it’s enough to cover four or five losing streaks in a row. A good bankroll manager never bets more than he can afford.

It occurs to me that Tony’s $400 in bets for this week is almost half his total bankroll. When I mention this, he laughs. “Ideally, I meant.”

To demonstrate his proficiency in record-keeping, he flips open a laptop, retrieves a file, and shows me a spreadsheet titled “Poker Wins/Losses.” It’s dated July 2010. The columns are topped with a mess of acronyms and symbols. I point to the column marked “D/N” and ask what it means. “Double or Nothing,” he says, which was his favorite type of tournament. Ten guys start, half double their money, half get nothing.

“See here?” He points to the column marked “RNG TOTAL,” where he listed the cumulative amount he was up (or down) for the month. “I can see at a glance when I was running good.” The numbers in bold, he explains, are new monthly highs. “Look at all that bold. That was a good month.”

At the bottom of the page, below the columns and rows, he has notes typed for every day he played. I ask if he really kept this up every single day.

“No,” he says. “The screen cuts it off. Those notes are for every match I played.”

One reads, “7/21 — lost one on some experimental shit. A new regular-type guy re-raised my button raise. I thought he was making a move and shoved.”

The rest of the words run off screen. Tony laughs. “That’s funny,” he says. “I shoved.”

Before I can ask what that means he says, “I’m a really systematic person. I’m attracted to patterns and statistics and all those types of things. But the other thing that keeps me out of danger is that it hurts me to give away money.”

He pauses to watch Eli Manning complete a long pass.

“Damn it,” he says to the television. Then to me, “Yeah, I hate losing more than I enjoy winning. If I win, I’m, like, ‘All right, great.’ But if I lose, I’m up all night thinking about it. It kills me.”

Jen leans in from the kitchen where she’s now standing over a pot of pasta.

“Oh, please!” she says. “You’re not that low-key when you win. When you take that baggie out of the freezer, you sniff your money.” She turns to me. “He does. Then he tries to make me sniff it, and he says, ‘Nothing smells better than won money.’”

Tony smiles.

“That’s true,” he says. “I do like counting up my cash.”

He leaves the room briefly, returning with a freezer bag. He opens it and pulls out a roll of bills wrapped in a rubber band.

“Feel how cold it is,” he says.

I do.

“You’d be surprised how fast it warms up.”

∗ ∗ ∗

That first time in the parking lot at Arby’s, Tony sat in his car and watched others drive into the lot. He stayed on the phone with Jen (at her insistence), described each vehicle to her, and together they tried to guess which would belong to the bookie. Anytime a man parked and got out of a car, Tony watched for him to scan the lot, to look around as if he were waiting for someone. But everyone who parked went straight into the restaurant. Ten or so minutes past the meeting time, Tony hung up with Jen and called the bookie to let him know he was waiting.

Two seconds later, a “schleppy” guy stepped out from inside Arby’s and looked around. Tony still laughs at the ordinariness of this whole business. The shiny Lincoln Town Car he’d been expecting turned out to be a ten-year-old Nissan Maxima. And the intimidating muscle man that collects the money in the movies is, in real life, no different than any other guy trying to make a living.

Jen, too, laughs about how nervous she was. “We should have known. [Our friend] Steve isn’t the type to be involved with anything crazy.”

The bookie became even more ordinary in Tony’s eyes when, during a subsequent meeting in the parking lot at Arby’s (where all their exchanges take place), the guy mentioned that his kids were inside. Eventually, Tony realized that Arby’s isn’t merely a safe, convenient meeting spot. It’s where the bookie’s kids like to eat.

“This guy might as well be a shoe-repair man, or a guy who owns a corner store,” Tony says. “He just seems like someone who happened into this line of work.”

∗ ∗ ∗

Despite the long pass, the Giants’ drive stalls. San Francisco takes over. Tony shakes his head.

“If it was just me, I never would have bet on San Francisco to win this one,” he says. “I’m telling you, it’s the system.”

This is at least the third time he’s said this. I tell him I want to know exactly how it works.

Before he explains this system of his, he warns me. “It’s not that it’s advanced,” he says. “It’s just that it’s math on top of math on top of math. And if you’re not intrigued by numbers, I promise it’ll be boring as hell.”

Behind his back, Jen nods her head exaggeratedly.

I promise to do my best to stay awake. Tony promises to keep it as simple as possible.

First, he goes back three years, to when he saw a story about a professional sports gambler on TV. Tony had played poker seriously in the past and had studied up on it as well. He’d learned that while most people think of poker solely as a game of chance, the best players are logical thinkers who rely more on statistics, odds, and math than they do luck or feelings.

“Have you ever watched professional poker players at a tournament?” he asks. “They can tell you the precise odds of a specific card coming out next. It’s math. There’s luck involved, but math plays a big part.”

At first, Tony didn’t understand how the same idea could apply to betting on sports. Back then, the only sports betting he’d done was the occasional $20 wager with friends. But after the television show featuring the professional sports bettor, he knew there had to be a method to it. So he went to the library and picked up a book titled Beat The Sports Books: An Insider’s Guide to Betting The NFL. He read bits of it, flipping through to the interesting parts. After getting what he thought was the general idea, though still skeptical, he decided to give it a try.

“I hate to lose a nickel,” he says. “I use coupons.”

For the whole 2009–2010 football season, Tony watched ESPN and NFL Countdown shows. He read about each team online, who’d been injured, what the weather might be like on game day, and so on. He made his picks on paper as if he were gambling — but without putting money on them. He showed his wife his “bets” before the games, so he would have a witness to his wins and losses.

In the end, he lost more games than he won.

“I was bad,” he says. “I won, like, 40 percent of my bets. I remember thinking I actually could have made money by making my best possible pick and then betting the opposite. That’s how bad my picks were.”

In the off-season that followed, Tony went back to the library and checked out the sports-betting book again. This time he read it thoroughly. At night, while Jen watched television and his baby slept, he took notes. On Saturday afternoons, he donned headphones and read at his desk. Every now and again, he went into the upstairs bathroom and read there. Eventually, he understood the author’s message: you can’t beat Vegas.

Vegas has the best computers and the best handicappers. On top of that, they open their lines first to the big-money bettors, then move the line according to how those guys bet. By the time your average Joe steps up to bet, Tony says, “the spread is perfect.” When Vegas makes a line (or a “spread,” as Tony calls it), their line is based not just on how much they think a team should be favored, they also factor in the public’s perception of how good the team is.

“That’s the part where a bettor like myself can make some money,” he says. “You try to find games where the Vegas line is off due to the fact that they’re compensating for public thinking.”

He gives me two examples of teams that have made him money so far this year: the Philadelphia Eagles and the San Diego Chargers. Philadelphia, he says, got a handful of highly regarded free-agent signings in the off-season. “And they were already pretty good before that.” They began the year calling themselves The Dream Team. Some experts even predicted the Eagles were going to the Super Bowl.

“No matter how often Philadelphia loses, people still seem to believe they’re going to be great,” Tony explains. “But they’re not.”

For San Diego, the better-than-they-actually-are reputation has lasted a few seasons.

“Every year, some major sportscaster picks them to go to the Super Bowl,” Tony says. “They do have a lot of talent. They’re always supposed to be so good. And for some reason, it seems like San Diego always figures out a way to lose or to just barely win.”

He pulls up a website called Goldsheet.com on his laptop.

“See? San Diego is 2–6 against the spread. If you bet against them every week, you’d have won six times and lost twice.” [Since then, San Diego has lost three more games against the spread.]

But Tony’s system isn’t just about betting against public opinion. It’s more complicated than that.

The book taught him that he’d have to create his own lines to compare to those Vegas came up with. To do so, he’d have to watch every single game, then rank the teams in order, based on what he saw. From there, he’d have to look at which teams were matched up for the following week and decide — based on numbers, not opinions or biases — who was likely to win each game and by how much.

Tony didn’t have the time, the inclination, or the recording equipment to watch every NFL game every week. But he knew of one site that did its own computer rankings, and he decided to look for others. Turns out there are 50 or 60 different computer rankings. He started reviewing the rankings and deciding which ones he liked best.

“I found one where their method is to actually rate a team based on how successful each play is,” he says. “So I found all these different computer systems, and instead of doing the work of ranking teams myself, I just take their rankings and average them.”

He slides a piece of paper across the table, a tight grid of numbers and letters, handwritten in pencil, slightly smudged and creased but still readable. The tops of five columns are labeled with the initials of the website that creates the ranking. Below that, each column ranks all 32 teams in the NFL by a given set of numbers. Column one reads: 1) NYJ + 35 +11.1; 2) GB + 28.2 +8.9; 3) HOU + 26.2 +8.3; and so on. No single column is the same as the others.

At the bottom of the page is another, smaller grid. This one lists the matchups for the week and the number of points by which each of the above sites expects the winning team to win. For each game, Tony has averaged these numbers to create his own line, which he’ll then compare to Vegas’ line.

“When I make my number, if the Vegas number is at least three points different, then I’ll bet,” he says.

Like most successful bettors, Tony bets in “units.” The more confident he feels, the more units he bets. For him, one unit is $25. When the difference between his number and Vegas is three points, he’ll bet two units ($50). If the difference is five or more, he’ll bet three ($75). And if it’s seven or more, he’ll bet either three or four units, depending on how optimistic he feels.

“So far this year,” he says, “I’ve never lost a three- or four-unit bet.”

He stops talking and watches the flat-screen.

“Sorry,” he says. I’m hoping the Giants don’t score right now. And they…just scored a touchdown. So now they’re going to be losing by one after they kick this extra point. But I’m actually behind by five because I need the Giants to lose by four.”

When the ball clears the goalposts, Tony turns his attention to a computer monitor on a desk to the right of the television. The sound is off, but the screen shows the Detroit/Chicago game. He explains that he’s doing something called “middling” with this game.

“I have Chicago plus points and Detroit plus points, in two different bets,” he explains. “A friend of mine said, ‘Hey, I think Chicago will win by five.’ ‘You wanna bet?’ I said, and he said, ‘Yeah.’ So we bet $40.”

Detroit plus-three is the Vegas spread this week, Tony explains. But his trusty system told him Detroit should be even money in this game, meaning neither team should be getting points. So he bet $50 on Detroit plus-three with his bookie.

“So, if somehow Detroit would win by four or less, or Chicago would win by two or one, then I win both!” he says. “That’s every gambler’s fantasy. You’d call all your friends and brag, ‘I won both sides of the game.’ It’s ridiculous.”

Right now, the likelihood of winning both is diminishing. Halfway through the third quarter, Chicago is blowing out Detroit, 34–6. Tony says it’s his worst-case scenario, but still he’ll only be out ten bucks on this game.

“The thing you gotta realize is, even with a great system, variance is a killer. Let’s say Team A is 75 percent better than Team B. For the NFL, that’s a pretty big difference. Still, that means three out of every four times they play, the better team will lose. So now, not only do you have to avoid that one time the better team will lose, you also have to hope they beat the spread the other three times. The truth is, you can be right about who’s the better team and still lose your bet.”

∗ ∗ ∗

This week, Tony has nothing on the Sunday-night game and nothing on Monday, either, so by 5:00 p.m., he’s done betting for the week. The results? He lost $4 online from his college bets. And he’s down $185 with the bookie. But he’s up $120 with friends and coworkers, which means he’s only down $65 for the week.

“My third losing week of the season,” he says. “I’m a little disappointed, but it could have been much worse. I’m still way up, overall.”

(A few days from now, in the parking lot at Arby’s, when it’s Tony’s turn to pay, the bookie will say, “So you’re human after all.”)

Right now, Tony has $690 more than he started with seven weeks ago. That’s an average of almost $100 per week. Even if he continues this trend, with just six weeks left in the regular season and another 17 next year, he’ll only be able to put $3000 toward the trip to Italy. But Tony has another plan.

“If I can end this season with at least $1000,” he says, “Next year, I’m going to double my bets.”