A Fantasy Sports Wizard's Winning Formula
Tuesday, June 10, 2014
Posted by: Meg Van Petten
Some 30 million Americans play, but
is there a secret formula to winning fantasy sports games?
EARLY ON A Friday evening in South Bend, Ind., Cory Albertson is
among the small subset of students on the University of Notre Dame campus not
prepping for a party, a date or both. At 29 years old, he is past his partying
phase—he even gave up beer as part of a gluten-free diet he recently adopted. A
graduate student in the business school, he is fixated on other, more lucrative
extracurricular activities. From a small table near the window in his
fourth-floor apartment, the Catholic school's iconic Golden Dome visible in the
distance, Albertson sits feeding numbers into tens of thousands of rows and
columns that populate a spreadsheet on his laptop. A Bob Marley remix plays
softly in the background.
With his close-cropped red hair, broad shoulders and easy grin,
Albertson seems plucked from the Fighting Irish recruiting catalog. He is
enjoying his return to campus life but increasingly finds it difficult to make
time for his classes. He considers it hard to believe, though more plausible by
the day, that the side business he started last year with $200 could actually
make him rich. Even more incredible: His business is playing fantasy sports.Using tactics more familiar to a hedge-fund manager than to your
average sports enthusiast, Albertson is earning thousands of dollars almost
every day. One NFL Sunday, he took home more than $100,000. "It is like
securities trading," Albertson explains, "and athletes are the
On this night, Albertson is making projections for how each
player will perform in the evening's professional basketball games—12 games,
involving a total of almost 300 players. It is painstaking, almost menial work.
As fuel for his calculations, Albertson imports statistics from the Web into a
complex algorithm he built over a period of months, and then tweaks the numbers
to reflect injury updates, matchups and other factors. He adjusts for
consistency—giving different weight to a player who scores 11 points almost
every game than to one who is prone to going off for 20 points one night and
recording a measly two points the next. He factors in potential fatigue if, for
example, a player had multiple games in recent days, and accounts for trends in
Albertson also needs to ensure his information is up to the
minute. As he calibrates, he clicks back and forth between the spreadsheet and
his Twitter feed, where he pays special attention to the dispatches of the
newspaper beat reporters for the teams playing that night. "If someone
slips and falls in the locker room before the game, it's going to be the beat
reporter who knows it first," Albertson says.
It is 5:30 p.m., about an hour and a half before the first game
tips off. As Albertson speaks, Gordon Hayward, a Utah Jazz shooting guard, is
downgraded to doubtful for that night's game because of a bad hip. Albertson
adds the new information to the spreadsheet and updates the entry for Alec
Burks, Hayward's backup. If Hayward sits out, Burks is likely to get more
playing time. Eventually, Albertson punches the button to let the algorithm do
its work. The pinwheel turns and the laptop whirs as the machine cycles through
millions of calculations. Albertson waits.
More than 33 million Americans play fantasy sports, which has
mushroomed into a $3.3 billion industry and become the purest expression of our
sports-obsessed culture. (A primer for those whose interests run more toward
"Real Housewives" and "Glee": In fantasy sports, as the
name suggests, fans get to live out their dreams of owning and managing their
own teams. They pick real professional athletes to form a roster, and then
compete against other "owners" based on the statistical performance
of the chosen players. In this world, Denver Broncos quarterback Peyton Manning
is a god.)
For most players—"addicts" may be a better
word—fantasy sports is a hobby, albeit one that consumes their weekends and
keeps them checking their smartphones in the middle of the night to get the
final stats from the late games. It is the bane of many a spouse's existence.
Albertson is in a completely different league. In between
attending business-school classes and cramming for exams, Albertson plays a new
high-volume variety of fantasy sports that allows him to compete in hundreds of
leagues each day. He may be the most consistently profitable fantasy sports
enthusiast in the country. And the reason is that to him, fantasy sports isn't
really about sports. It's about data. He doesn't trust his eyes to tell him who
the best players are; he only trusts the numbers, which tell him who is
overvalued and who is undervalued. Albertson's strategy is almost literally a
formula, and it bears little resemblance to casual fandom. "It's all about
having logical inputs that lead to logical outputs," he says.
The result of all this evening's research and effort is that
Albertson's algorithm determined the optimal lineup for a fantasy basketball
team. More precisely, on Albertson's command, it produced hundreds of lineups
that collectively represent a diverse portfolio that will, he hopes, generate
thousands of dollars in profits by the end of the night's games, in about six
He doesn't make money every night, and in fact ends up in the
red on a regular basis. But he wins more than he loses, the benchmark all good
traders strive to meet. Last year, Albertson took home more than $200,000. This
year, he projects his business will make almost a million dollars.
THE MONEY IS burning a hole in Albertson's pocket. On a snowy Chicago
afternoon, he folds his 6-foot-5 frame into the driver's seat of a white Tesla
Model S for a test drive. The salesman launches into his spiel for the $110,000
electric sedan as Albertson zips through the sludge of a side street and toward
the wide open lanes of Interstate 90. "This is awesome, man. It's like
driving a spaceship," Albertson mutters.
There is little about him, including the rented Hyundai
039010.KQ +1.96% he parked outside the
dealership a few minutes earlier, to suggest he can afford to drop six figures
on a car. You can hardly blame the salesman for asking: "What kind of
business are you in?" Albertson guns it off the freeway entrance ramp,
pushing the speedometer past 85 and the electric car toward the Windy City
skyline in the distance. "I'm self-employed," he says. "I have a
fantasy sports business." The salesman, not surprisingly, says most of his
buddies are into fantasy sports. But he appears troubled by one thing.
"Isn't that sorta like gambling?" he asks.
Albertson hates that question—"I take it as an
insult," he tells me later—but he hears it all the time, from classmates,
prospective girlfriends and his parents. Albertson grew up in Warsaw, Ind., as
part of a conservative, blue-collar family in which his father worked for an
orthopedic-device manufacturer and his mother stayed home. He attended Ball
State University, where he studied criminal justice, for a time anyway. When he
started college, in 2004, the online-poker boom was in full roar and he spent
as much time on the virtual felt as he did studying. But like a small number of
online-poker aficionados, he was drawn more to the game's analytical components
than to the gambling rush it provided. He became fascinated by data and its
applications, keeping detailed records on his opponents and studying the
mathematically correct choice in thousands of possible situations that can
arise in a poker game.
Albertson doesn't like to gamble and says he hopes to never set
foot in another casino. He is a health nut who practices yoga and is traveling
to Haiti this summer to help train small-business owners. It pains him that
anyone, much less his parents, might see him as some sort of degenerate.
"They are pretty religious. To say we come from different worlds now would
be an understatement. I think they worry a lot," he says.
Last year, Albertson saw a chance to return to mainstream
culture. He applied and was admitted to Notre Dame's business school, where as
part of the orientation he was required to take a career aptitude test. The
software was astute: It said he should become a securities trader. As it
happens, Albertson had recently been approached by an old poker buddy, a
Chicago native named Taylor Caby, who was starting a fantasy sports website.
Caby needed players to populate the games on his site and offered Albertson a
rebate on some of his entry fees if he would try his hand at a few contests.
Albertson concluded that fantasy sports was like poker—a contest
dominated in the long run by skill, not luck. He teamed up with a friend from
the poker circuit, Ray Coburn of Jackson, N.J., to build the algorithm. Each
man put in about $200, they recall. But they made a key observation early: With
its use of vast reams of historical data, fantasy sports is similar to
stock-market investing. And much like the top investors, they decided the way
to get the most consistent results was to take as much chance out of the
equation as possible. "It felt like a problem we could solve," Coburn
Instead of coming up with one ideal lineup, they built a system
that designed a range of lineups that would create profits under a variety of
scenarios. Sure, Peyton Manning may be the best play, but what if he gets hurt,
or has a bad game? Better not to have Manning in every lineup. In investing
terms, they opted for a diversified portfolio rather than a concentrated bet.
concluded that fantasy sports was like poker—a contest dominated in the long
run by skill, not luck.
Until very recently, it wasn't possible to make a living playing
fantasy sports. A league can be free, with only bragging rights on the line,
but typically each owner puts up a modest entry fee, with the winner collecting
the bulk of the prize pool at the end of the season. Given that the seasons for
professional sports leagues last anywhere from four to seven months, the payout
on an hourly basis for the average fantasy league doesn't even come close to
the minimum wage. Albertson plays a new, faster version called daily fantasy,
which compresses the typical season-long fantasy league into a single day. So,
for example, instead of drafting a baseball team for the season and tracking
the players over the course of six months and 162 games, a daily fantasy
contest is decided based on who can pick the best players for a single night's
worth of games. The people who drafted a traditional fantasy baseball team this
March won't know the outcome of the league until the end of September. In daily
fantasy, people can draft a team on any morning of the week and know the
outcome by the end of the night—and then start over again the next day.
This version of the game popped up about five years ago, and a
handful of startups have since drawn more than $100 million in venture capital.
Comcast Ventures, the venture-capital arm of the cable conglomerate, is one of
the backers of
FanDuel, the biggest daily fantasy site at the moment. Barry
Diller's IAC/InterActiveCorp IACI -0.55% put money into a site
called DraftStreet. Daily fantasy "only really hit the mainstream in about
the last 18 months," says Paul Charchian, chairman of the Fantasy Sports
Trade Association. "Charch," as he's known, says there has been more
venture capital and other investor money allocated to daily fantasy websites in
the past few years than in the history of fantasy sports. The appeal for
players is twofold, he says: For one thing, many people simply like the shorter
time frame. But it also provides a way for serious players to make big money.
In traditional fantasy, people didn't try to make a living on it, because
"it took too long getting a return on your money," Charchian says.
Daily fantasy's emergence also coincides with the rise of sports
analytics, a once-obscure, nerd-populated corner of the sports world that has
revolutionized the industry in recent years. The trend got perhaps its most
mainstream exposure with the Michael Lewis book (and subsequent movie starring
Brad Pitt) "Moneyball," which showed how the Oakland Athletics
baseball team used statistics to rigorously measure players, upending much of
the conventional wisdom about the game. (The Boston Red Sox also embraced analytics, and
the team won the World Series in 2013, its third championship
in a decade.) But analytics have now been fully embraced by fans as well. In 2006,
the Massachusetts Institute of Technology started an annual event, called the
Sloan Sports Analytics Conference, which drew less than 200 people in its first
year. ESPN columnist Bill Simmons dubbed it Dorkapalooza. This year, it drew
nearly 3,000 attendees and sold out more than two months in advance.
Not surprisingly, gamblers are also embracing analytics. The
winner of the Las Vegas Hilton's SuperContest, the most prestigious
sports-betting event of the year, in which each participant picks five NFL games
each week, was David Frohardt-Lane, who has a graduate degree in statistics
from the University of Chicago and for eight years ran the equities desk at
Getco, a multibillion-dollar investment firm that has since merged with Knight Capital
KCG -0.33% to form KCG Holdings.
His prize for winning the SuperContest: more than half a million dollars.
The man many consider the most successful basketball bettor in
the world, a Canadian named Haralabos Voulgaris, compiles vast quantities of
data and has a team of assistants to analyze NBA games. Voulgaris reportedly
makes millions of dollars each year but declined to speak to WSJ.Money about
Sports gambling is illegal in most of the country, and the top
officials of the major sports leagues routinely denounce any suggestion of
expanding sports betting. Fantasy sports, however, is legal in all but five
states, and the leagues love it—they say it drives television viewership and
overall engagement among fans, without introducing the unsavory aspects of
full-fledged gambling. This is one reason why many analysts and investors think
the daily fantasy industry is poised to explode, and why the sites that offer
the games are aggressively marketing on sports radio and other mainstream
outlets. Andrew Wiggins, a friend of Albertson's and the founder of
a daily fantasy site, says the future of the industry is based on attracting
casual fans. "If this is going to get huge, we need the guys who are going
to buy in for $20. They do it for fun," Wiggins says. But if that happens,
number-crunching sharks like Albertson will be lying in wait. If casual players
embrace daily fantasy in bigger numbers, Albertson says, "then we'll
really be a printing press."
A FEW HOURS after the Tesla test drive, sitting about 20 rows from
the court at Chicago's United Center and munching on a gluten-free hot dog,
Albertson asks a question that is on no one else's mind in the arena: "How
come D.J. Augustin is not in the game?"
The Chicago Bulls are playing the Philadelphia 76ers in a
meaningless midseason game, and as the first quarter draws to a close, Augustin
is sitting on the bench. Augustin until recently was almost washed up in
professional basketball. At age 26, he had been cut by three teams before
latching on with the Bulls when the team's starting point guard, Derrick Rose,
went down with a knee injury. The Bulls also had recently traded their best
remaining player, Luol Deng, in a move that most fans interpreted as raising
the white flag on the season.
Albertson—or, rather, his algorithm—sees Augustin's stock
beginning to rise. On this night, it identifies Augustin as one of its top picks,
even though the player was chosen by only 2.4 percent of Albertson's
competitors in one of his biggest contests. And the algorithm is right.
Augustin comes into the game late in the first quarter and goes
on to lead the Bulls to a 103-78 victory. More important to Albertson, Augustin
fills up the stat sheet, finishing with a total of 19 points, eight assists,
three steals and four rebounds. Albertson and Coburn's business ends the night
with a nice profit—almost $9,000. The algorithm isn't right all the time, and
even loses money on many nights. But Albertson says he is thinking of pursuing
investors, so he can invest outside money as well as his own—sort of like a
fantasy sports hedge fund. "It's a big enough sample size, there is
clearly something to it," says Albertson. "It's not going to be
perfect, but it can be pretty damn good."