“It’s my life.”
“That’s me!” says the famous Iraqi sculptor.
Alaa Al-Saffar is in the artist’s studio connected to his small home on a sunny California day when he points to a faded photograph on his table. It is a portrait from his old life working for a murderous dictator. The man in the photo wearing blue jeans and with hair below his shoulders is the only person in a room of Iraqi officials not dressed in a suit or a military uniform. He looks hilariously lost. It is as if he’d been on his way to a gallery opening and stumbled into a cabinet meeting. The effect of Al-Saffar’s presence is so disorienting that if you were to glance at this photo you would notice him before you realized the man standing beside him is Saddam Hussein.
It was neither by accident nor by choice that Alaa Al-Saffar found himself working for one of the most brutal leaders of the twentieth century. A few times every year, Al-Saffar received an envelope stuffed with cash. He didn’t need to read the note inside to know that it was an invitation from the president of Iraq. “Saddam was an easy person to get on with,” Al-Saffar says. “At least with us artists.” But his personal and professional opinions about this man nicknamed the Butcher of Baghdad didn’t really matter. He’d been summoned enough to understand that honoring the whims of Hussein was a minor inconvenience that Iraqi artists had to deal with every so often. It was like a teeth cleaning.
He tolerated Hussein as a patron only because it allowed him to pursue the life of an artist. The son of a painter, Al-Saffar studied art in Baghdad and left his native country to pursue a master’s degree in Switzerland, which afforded him the opportunity to visit Paris and Rome to study Rodin and Picasso, Chagall and Dalí, Michelangelo and Da Vinci. He returned to Baghdad in the 1980s and immediately distinguished himself by winning a series of prizes for his sculptures, paintings, and sketches. He would never have the stability of a steady job or a reliable paycheck, and there might be some months when he worked from the garage of his Baghdad home because he couldn’t afford his studio. But he was fine with the choices he made. He accepted that his commitment to his art required sacrifices. “He worked nonstop,” recalls Zinah Al-Saffar, the oldest of his children. “His dream was to have a big statue in the United States or have his paintings on display in one of the biggest galleries in the world.”
But that’s when he became one of the dictator’s favorite artists. The attention of Iraq’s president was a powerful amplifier that could make or break careers, and when Hussein invited artists to private competitions, they didn’t think twice about taking his money. “It’s not that you are forced,” says Natiq Al-Alousi, another Iraqi artist. “It’s up to you whether you want to participate or not. And everybody wants to participate. That’s income. That’s work. Everybody is looking for work.” By the time his presence was requested in late 2002, the year before the United States invaded his homeland, Al-Saffar was used to taking a commission from someone who used chemical weapons against his own people while treating artists like heroes. He was especially proud of the last sculpture that he proposed to this man who was about to burrow into a spider hole. It was meant to be a gargantuan bronze piece to reflect his national pride. The sculpture would be his most ambitious project yet. On top of the whole thing, like the bride and groom figurines on a wedding cake, was an enormous palm tree. “Iraq is famous for the palm,” Al-Saffar says. “Like California.”
Hussein thought it was almost perfect. But the dictator had some notes. He suggested one thing that he felt was missing from the sculpture: a massive statue of Saddam Hussein atop the palm. Al-Saffar knew what to say: “I say okay,” he said. He agreed that his sculpture could use a massive statue of Saddam Hussein standing on a palm tree.
But the war began. Saddam Hussein was captured and killed. Alaa Al-Saffar’s sculpture would never come to exist beyond the scope of his imagination.
Instead he spent the days after Hussein’s downfall chiseling stone in his garden1 and making art for which there was no market. Al-Saffar felt like he was on the verge of a breakthrough when he won a national commission for a memorial sculpture, but the funding for that project went dry before he could start his work, and then his life fell apart because of it. When his name was publicly mentioned in the news for winning the commission, he became a target of Al-Qaeda and the insurgent militant groups as they consolidated power in the aftermath of Hussein’s reign. (As it turned out, Al-Qaeda wasn’t full of art buffs.) It was suddenly dangerous to be Alaa Al-Saffar.
His daughter was getting ready to bring her children to school one day when she says she noticed an envelope slipped under the garage door of the family’s home. This one wasn’t stuffed with cash. Inside the envelope was a death threat for her father. “It was an order for him to stop being an artist,” Zinah Al-Saffar says. “Not a request.”
There would be no more painting and sculpting in his garden. If he kept working, he would be killed. “They called me an infidel2 and told me to stop,” he said. “So I stopped.”
Zinah soon moved to Southern California with her family because her husband had worked for the U.S. Army during the war, and she had to watch from afar as her father struggled to adjust to this life without art. She could see how tenuous his situation was. Every time she called home, he tried to reassure her that he was safe. “But I knew,” she says. “I was there when he got the letter. I’m the one who found it.”
She urged him to leave. “You don’t belong there!” she told him. He wouldn’t listen. When she finally persuaded him to visit, he landed in Los Angeles with a temporary visa. His trip was only supposed to be a vacation until they found themselves staring down the worst part of vacation: the end. Zinah couldn’t bear the idea of her endangered father returning home. “Dad, I don’t think you should go back,” she said. “Stay here.”
He knew that she was right. In that moment he accepted that his old life was over. The famous Iraqi sculptor was in his sixties when he decided it was time for a new one.
He filed the paperwork to inform the proper authorities that he was in grave peril back home, that he feared persecution if he returned to his native land, and that he needed to be protected by the United States. And thus began the asylum process for the artist from Baghdad. Alaa Al-Saffar’s new life was about to be seized by the hot hand.
Justin Grimm wasn’t sure why he was being called into his boss’s office. It was another oppressive day in Frisco, Texas, the latest stop on the carousel of minor-league baseball teams Grimm had been riding since he’d been drafted by the Texas Rangers. He’d already played for the Hickory Crawdads and Myrtle Beach Pelicans. Now he was a Frisco RoughRider. He expected to be in Double-A ball for a while after the Rangers had invited pretty much everyone with a realistic shot at playing for them to spring training that season and had not invited him. But he’d been pitching well lately, and the people who drafted him were paying attention, which was why his manager needed to see him on this particular Thursday afternoon.
The Texas Rangers needed a starting pitcher on Saturday night against the Houston Astros. They decided against every reasonable expectation, including Justin Grimm’s, that the man for the job was Justin Grimm.
In front of a sellout crowd of nearly fifty thousand people, including a former president of the United States, he soon found himself strolling to the mound with wobbling knees as the television announcer said to the audience at home, “Who knows what’s in the stomach of Justin Grimm?” As it turned out, not much. Grimm was so nervous that he hadn’t eaten all day. He toed the rubber, took a deep breath, tricked himself into thinking his big-league debut was just like any other game, stared at his catcher for the sign, and fired a ninety-one-mile-per-hour fastball. His ability to throw a baseball with amazing velocity and pinpoint accuracy was one of the reasons the Rangers thought he was ready to leap from Double-A on two days’ notice, but he would later admit that he had no earthly idea where the ball was going when it left his right hand. It zipped down the middle for a called strike. The first pitch of Justin Grimm’s major-league career was perfect.
Grimm let the leadoff hitter get on base with a single, but he got the next hitter out, and he was beginning to settle down and forget that he was living his childhood dreams when Jed Lowrie walked to the plate.
For a professional athlete, Jed Lowrie was awfully peculiar. He might have been the best player on the worst team in baseball that season. The most striking physical attribute of this undersized six-foot, 180-pound shortstop was his piercing blue eyes. He represented the few dozen professional baseball players with a college degree. And he wouldn’t have been in the batter’s box that day against Justin Grimm if he didn’t have one.
Lowrie was a political science major at Stanford University the first time an improbable major-league scout named Sig Mejdal came to see him. Mejdal studied aeronautical engineering in college and operations research and cognitive psychology in graduate school, and he paid his tuition through a side gig as a blackjack dealer at the local casino.3 He was working for Lockheed Martin and NASA as a literal rocket scientist when he decided that he wanted to work for a baseball team instead. His first job in baseball was in fantasy baseball: Mejdal was the quantitative analyst for a lunatic owner in a wildly competitive fantasy league. That invaluable experience led to a position in the fledgling statistical department of the St. Louis Cardinals—a real job with a real team. Mejdal built a model that used a prospect’s college statistics to project his future in the big leagues, and his computer told him the best college baseball player in the country was playing at Stanford.
But when he went to see that player in person, the most remarkable thing about Lowrie was how unremarkable he was. Jed Lowrie resembled a Major League Baseball player as much as Sig Mejdal carried himself like a grizzled scout. Mejdal resisted the urge to break his laptop for long enough to remember that he didn’t care how Lowrie looked, only how he played, and he played the way his algorithm suggested. Mejdal begged his bosses to draft Lowrie. When they passed on him with their first pick, Lowrie was selected by the Boston Red Sox instead. It would soon become clear that the Cardinals should have listened to Mejdal. Lowrie made it to the big leagues, just as the algorithm said he would, and he was coming into his own when Mejdal’s boss was named the general manager of the Astros on December 8, 2011. Less than a week later, he traded for Lowrie. Less than a month later, he poached Mejdal. Finally they were on the same MLB team.
By the afternoon of June 16, 2012, when he was staring at Justin Grimm on the mound, Jed Lowrie was once again proving Sig Mejdal right. He was on pace to hit more homers in his first season with the Astros than in his previous four seasons combined. He’d already hit eight home runs that month after never having smashed more than nine in a full season. But that was mostly because it had never really been his goal to hit home runs. When he was in high school, Lowrie’s field didn’t have a fence. There was no incentive for him to hit the ball high and far in the air. The only way that he could guarantee himself a home run was by hitting it so hard that it kept rolling long enough for him to round the bases. That became his goal: hit the ball as hard as he could.
But the sport of baseball was on the brink of its own version of the three-point revolution that put a heavy premium on strikeouts for pitchers and home runs for batters. This one at-bat in this one meaningless game between the Rangers and Astros was an unexpected peek at the future. Grimm wanted a strikeout. Lowrie wanted a home run.
Grimm’s parents were escorted to the best seats in the house, right behind home plate, in time to watch their son throw the first pitch to Lowrie and hear the umpire Bill Miller call the fastball a strike. There was nothing memorable about this pitch. It was one of the hundreds of thousands of pitches that make up a baseball season. Lowrie rearranged the dirt in the batter’s box. Grimm looked at his catcher for the sign and conveniently ignored the fact that he was about to throw his next pitch as hard as he could almost directly at his family.
It was around this time when the showdown between Grimm and Lowrie subjected itself to the fickle biases of human judgment.
As soon as he took the mound, Grimm had the advantage over Lowrie. The pitcher always has the advantage over the batter in a game where even the best hitters fail more than they succeed. But his advantage in this particular situation was even bigger. Grimm had never pitched in the big leagues, and there was no reliable scouting report for Lowrie to study. He wouldn’t know how the pitches would look until they were hurtling at him. “You don’t know what they do,” he says. Grimm’s inexperience meant that Lowrie was basically wearing blinders when he stepped to the plate.
The odds of this at-bat shifted further in Grimm’s favor as he prepared to deliver the next pitch. The catcher Mike Napoli moved several inches away from Lowrie. By crouching on the outside corner, he was attempting to fool the umpire. Miller might be tricked into calling a ball a strike if the pitcher didn’t make the catcher move. Grimm and Napoli were conspiring to make the strike zone slightly bigger. It worked.
Grimm put the ball where his catcher wanted: low and outside. Lowrie recognized that it was low and outside and figured that it would be called a ball. Miller called a strike. Lowrie couldn’t believe it. He turned his whole body and registered his discontent with a vicious stare. The silent protest was Lowrie’s equivalent of screaming in the umpire’s face given how rarely he argued the merits of a called strike. “Only when it’s a ball,” he mutters to me years later.
By this point of his umpiring career, Bill Miller was impervious to Jed Lowrie’s stink eye. Miller was basically a lifer behind the plate. He first started thinking about umpiring when he was still in middle school. He made some extra cash in high school by calling Little League games, quit his college baseball team to be a high school umpire, and went to umpiring school to work in the minor leagues as soon as he could.4 He’d been in the big leagues fifteen years, and he loved his job. He loved making a judgment after every pitch, and he especially loved coming to Texas,5 where the clubhouse attendants Hoggy and Cornbread made sure there were always heaps of jumbo shrimp, brisket, and apple crisp waiting for him as he took off his uniform. He was even used to the one part of umpiring that could make him not love his job: the complaints. To complain about balls and strikes was as much a part of the game as the buckets of sunflower seeds in every dugout. There were times when it seemed like the purpose of Miller’s job was to call pitches and then get berated about how bad he was at his job. A good umpire had to acknowledge his mistakes, however, and Miller was objectively a good umpire.
But there was one more thing about Miller that made his mere presence behind the plate better for Grimm than Lowrie. He was a friend of pitchers.6 Miller called about four more strikes per game than the average umpire. In fact his strike zone was among the biggest in all of baseball. “High pitches, low pitches, inside pitches, outside pitches, pitches to left-handed hitters, pitches to right-handed hitters—Miller is almost always calling more strikes,” read one description of his style on the baseball blog Hardball Times.7
It wasn’t possible to say with definitive proof whether any umpire was right or wrong when Miller made it to the big leagues. But then came a system called PITCHf/x that planted high-resolution cameras in every ballpark and tracked in precise detail everything you could have ever wanted to know about any given pitch: its speed, its trajectory, and, most important, its location. It was technology that could make people like Bill Miller obsolete. Human error had always been a part of baseball if for no other reason than it had to be. But now it didn’t. There was nothing other than the inertia of a hidebound game that was stopping Miller from being replaced with a machine. In the meantime, Miller’s bosses developed a method known as Zone Evaluation, which used PITCHf/x data as the unimpeachable standard to determine how often umpires like Miller were right and wrong. It was an ominous reminder that the robots would never be wrong. But until they came along, balls and strikes would still be called by humans. The upshot of PITCHf/x was that it was suddenly possible for anyone, not just Miller or his bosses but everyone with access to the internet, to see for themselves if a pitch was a ball or a strike as soon as it crossed the plate. In this particular instance, the umpire believed the blur of a fastball had grazed the outside corner of the plate, but really it had missed the plate by an infinitesimal amount that made all the difference.
It was called a strike. It should have been a ball.8 Jed Lowrie was right. Bill Miller was wrong. Justin Grimm was lucky.
Lowrie now had two strikes against him as Grimm looked at his catcher for the sign and saw one finger: the pinkie. A pinkie in the major leagues is the same as the pinkie in Double-A ball and every level of the game all the way down to Little League. In the universal language of baseball, the pinkie is an inside fastball. Grimm nodded in agreement with his catcher. They both felt the inside fastball was the pitch that would strike out Lowrie. The cameras in the stadium that night would record this fastball at ninety-three miles per hour as it whirred over the inside edge. The precise baseball terminology for this kind of pitch is “filthy.”
If the umpire were a robot making decisions using robot logic according to a box over the plate that only a robot could see with robot eyes, he would have initiated a strike-three call after such a filthy pitch.
Bill Miller was not a robot. He was very much a human being. And the umpire with the biggest strike zone in the game called this borderline pitch a ball.
The one thing that’s hard to appreciate about MLB umpires when drunk fans are screaming obscenities at them is that they are phenomenally good at their jobs. They manage to nail about 87 percent of their calls,9 and they almost never miss the obvious balls and strikes. Their success rate on those pitches is 99 percent. It’s the toss-up calls within inches of the strike zone that give them fits. Even the most reliable umpires are right about those pitches only 60 percent of the time.
But that wasn’t the only reason Miller called the pitch from Grimm to Lowrie a ball. A few years later, this pitch was one of millions over five MLB seasons analyzed by a team of economists intrigued by the behavior of professional decision-makers in high-stakes situations. What they wanted to know was how umpires acted immediately after calling two consecutive strikes. And what they really wanted to know was whether it had any effect on the next pitch.
In a perfect world, it wouldn’t. Major League Baseball is not that perfect world. If the last two pitches were called strikes by the umpire, he was 2.1 percentage points less likely to call a strike. Miller’s strike zone had shrunk for this pitch. The fact that Miller had already called two strikes and Grimm had thrown another near-strike actually conspired against him.
Grimm betrayed no emotions about being robbed of the first strikeout of his big-league career. He pursed his lips and stared at the catcher for his next sign. There it was again: the pinkie. He didn’t need to nod this time. As soon as he saw that pinkie, he went back to his set and fired another inside fastball. The runner on first base took off for second base. The catcher threw a perfect strike of his own to catch him stealing for the second out. Grimm exhaled again. He was now one pitch away from the end of the inning.
Grimm went into his windup and uncorked a ninety-four-mile-per-hour fastball right over the heart of the plate. It turned out to be a huge mistake. Lowrie pounced on the second chance afforded to him by Miller not calling a third strike. The instincts that he’d been training since his days in high school on a field without a fence took over. He swung. The ball cracked off his bat and climbed for the abyss between right and center field. Grimm’s mother covered her mouth in horror. But the ball was still going. Grimm’s sister had shielded her eyes at the sight of contact. And still the ball kept going. She finally allowed herself to peek at the worst possible moment: when it sailed over the fence.
Jed Lowrie hadn’t just hit the ball hard. He’d hit a home run.
The one thing that Justin Grimm, Jed Lowrie, and Alaa Al-Saffar have in common is that each of their predicaments can be explained by taking a trip to a casino.
A casino is a good place to study decision-making because casinos are where people go to make bad decisions. This laboratory of willful stupidity has the controls that an experimental psychologist desires, but it also lures subjects who are real people betting their own real money, which is why every spin of the roulette wheel has all the makings of a rollicking study of human behavior waiting to be written. Rachel Croson and Jim Sundali decided to write it.
Croson was a behavioral economist who focused on financial decision-making. The kinds of mistakes that people make in their dealings with money were the cognitive errors that power the entire casino industry. Sundali knew that from firsthand experience. Before he was a scholar of managerial sciences, he was a stockbroker. When he went back to school for his Ph.D. he studied under an Israeli psychologist named Amnon Rapoport, who had been college roommates and inseparable friends with Amos Tversky. They had met waiting in line to register as undergraduate psychology students.
Sundali admired Rapoport the way that any student would idolize a professor who ushered him into a Wonka factory of ideas. They also had a closer relationship than most students and professors. When Sundali went on a skydiving trip with his college buddies, for example, Rapoport announced that he wanted in. “Which in and of itself was sort of remarkable, because he had a major heart condition and had been off for the semester getting his heart fixed,” Sundali says. “I didn’t know if I should be taking him skydiving.” He took him anyway. The scholar piled into the car with two former college athletes who had barely graduated but still found gainful employment managing other people’s money for august financial institutions. Sundali told his friends about a paper he’d read in Rapoport’s class that claimed the hot hand was a myth, and they were barely on the highway before they commenced their grilling of the egghead. Rapoport took such a beating that he finally blurted out, “Is there any piece of evidence I could give you to show this isn’t real?” Sundali’s friends concluded there was no amount of evidence that would change their beliefs about the hot hand. “I’m glad you have faith, and I’m never going to discount someone’s faith,” Rapoport said. “If you have faith, evidence doesn’t matter.” The car pulled into the hangar and they were handed liability waiver forms. Rapoport gulped. This man who had suffered a series of heart attacks couldn’t answer the health questions truthfully. He pulled Sundali aside.
“What should I do?” he asked.
“Well, if you want to go skydiving,” Sundali said, “check yes and sign your name.”
Rapoport ignored the evidence and went with faith.
The other reason that Sundali thought studying gambling would be worth his time was that his place of employment was the University of Nevada, Reno. There were as many casinos on the fringes of campus as there were coffee shops. When they set about publishing their series of papers about decision-making in gambling, Croson and Sundali looked at wagers made at the rapid roulette wheel in a Harrah’s casino.10 “We got this huge stack of old IBM paper with the betting patterns of every player,” Sundali says. “We know what they bet and when they bet it.” Sundali later got a call from one of his students who worked for a manufacturer of slot machines and offered a mother lode of seventeen million slot pulls. “It’s not the best game,” Sundali says, “but it was the best data set.” But they hit the real jackpot when one of Sundali’s students in an executive MBA course told him that he worked for one of the local casinos. Sundali mentioned that he was about to send a Ph.D. student to the casino floor to record bets on a roulette table. But his student promised to do him one better. He could obtain the security videotapes from his casino’s eye in the sky. “We had this Ph.D. student all excited,” Croson says. “And now they’re in a basement watching videotapes frame by frame and putting them into a spreadsheet.” But now Sundali and Croson were also sitting on a gold mine: eighteen hours of an overhead shot of the same roulette table. They had 139 gamblers, 904 spins, and 24,131 bets at their disposal.
They started by looking at the fifty-fifty bets in roulette. Those were the wagers like red versus black—basically heads versus tails. There were 531 bets after gamblers had witnessed at least one spin of the wheel. They were split roughly down the middle between bets on the same outcome (red after red) and bets on the opposite outcome (black after red): 52 percent versus 48 percent.
But when Sundali and Croson analyzed the bets after streaks of two, the ratio suddenly flipped: 49 percent versus 51 percent. It tilted toward bets of the opposite outcome. That is, after the wheel landed on red, 48 percent of the next bets were on black. But after the wheel landed on red twice, 51 percent of the next bets were on black. That number climbed to 52 percent after three reds, 58 percent after four reds, 65 percent after five reds, and a whopping 85 percent after six or more reds. “They place significantly more bets against the streak than with the streak,” Sundali and Croson wrote.11
This was the corollary of the hot hand: the gambler’s fallacy.
The gambler’s fallacy has been infecting the brains of unsuspecting bettors for as long as they have been charitably donating their money to casinos. The first person to coin this phenomenon was a French mathematician, statistician, physicist, and god knows what else named Pierre-Simon Laplace. He was Kahneman and Tversky hundreds of years before Kahneman and Tversky. If he were alive today, Laplace would have his own podcast. He would be famous after giving a TED Talk. If he were really unlucky, he might even be invited to Davos. But he came to these conclusions a long time ago. In one chapter of his seminal 1814 book, Philosophical Essay on Probabilities, he used the example of the French lottery to describe the gambler’s fallacy, noting how foolish it was that people favored the numbers that hadn’t been picked in a while. “The past ought to have no influence upon the future,” he wrote.12 This was especially true in casino games that stole people’s money—games like roulette. For the same reasons that people were tricked by the lottery, people also overreacted to streaks in the casino.
The gambler’s fallacy is different from the hot-hand fallacy. Stephen Curry makes three shots in a row, and everyone in the arena expects him to make a fourth. That’s the hot-hand fallacy. But the roulette wheel lands on red three times in a row, and everyone in the casino puts their money on black. That’s the gambler’s fallacy.
The issue is how we perceive outcomes that we feel we control (basketball) and how we perceive outcomes that we know are beyond our control (roulette). When there’s a streak that defies the odds, we understand that it will even out, and we place our wagers accordingly. We internalize the regression to the mean. But that’s not what happens when we are the ones challenging probability. When it’s Stephen Curry chasing that fourth consecutive shot, we believe he’s temporarily beaten the statistical inevitability. He’s on fire!
It’s even possible for us to believe in both the hot-hand fallacy and the gambler’s fallacy at the same time. Peter Ayton and Ilan Fischer came up with a clever way to illustrate this contradiction.13 They took some of their psychology students into a lecture hall and passed out a three-page quiz for extra credit. On each page was a jumble of @s and #s. It looked as if someone had butt-dialed a tweet.
As they tried to make sense of these seemingly incomprehensible sequences, the students were told that the symbols were the disguised results of six experiments their professors had conducted: basketball shots, coin flips, soccer goals, dice throws, tennis serves, and roulette spins. What the students weren’t told was that those experiments never really happened. The series had been spit out by a computer that randomly generated sequences with eleven at signs and ten hashtags. Ayton and Fischer split their fake experiments into two categories. They pitted human performance (basketball shots, soccer goals, tennis serves) versus pure chance (coin flips, dice throws, and, yes, roulette spins). The students looked at twenty-eight sequences with different alternation rates between the at signs and hashtags. The visual differences were striking. This is how a series with a low alternation rate looked compared with a series with a high alternation rate:
Ayton and Fischer said that each string of @s and #s could be makes or misses (as in basketball shots) or reds or blacks (as in roulette). And that’s pretty much all the information their students were given. It was up to them to figure out which sequences represented what tasks, and the professors offered extra credit to any of their students who outperformed the class average.
By then Ayton had been thinking about the hot hand and gambler’s fallacy for many years. One of the first things he’d written that anyone outside academia had bothered reading was a short essay for a popular science magazine in which he tried to replicate Gilovich, Tversky, and Vallone’s basketball paper in soccer. After studying the goals of the English Premier League’s top scorers, he reached the same conclusion. “Any belief in the ‘hot foot’ is also a fallacy,” he wrote.14 He soon experienced the same ruckus. His essay caused such an uproar in his native England that he was invited on the radio to debate Ron Atkinson, a longtime soccer manager better known as Big Ron, about whether a hooligan would be right to believe in streaks. Big Ron had some nits to pick. “You’ve never been in the dressing room!” Big Ron yelled. “I’ve been in the dressing room. I know what it’s like.” Ayton was delighted about Big Ron’s furor. He was getting precisely the same incredulous reception from soccer that Tversky had received from basketball.
But this particular experiment was different from most of Ayton’s work because he couldn’t predict beforehand if it would yield anything interesting about the hot hand or gambler’s fallacy. “Often when you run an experiment, you’re pretty certain what’s going to happen, or at least you think you are, which makes you wonder why you’re going to run the experiment,” he says. “But with this one, I really didn’t have a clue.”
What happened next wouldn’t have surprised Laplace. The students guessed the streaky sequences were basketball shots and the runs that seemed random were roulette spins. In their minds, @ @ @ @ @ was the product of human performance, but @ # # # @ was pure chance. When humans had control, they believed in the hot hand. When they were bystanders, they believed in the gambler’s fallacy.
The roulette tables in Las Vegas are designed with those bystanders in mind. Right next to the wheels in today’s casinos are electronic scoreboards that update after every spin with real-time statistics: the red versus black split, the distribution of numbers, and, most important, the last twenty numbers. It’s an impressive display that is entirely meaningless. The ostentatious screens are as helpful to gamblers as a livestream of the Bellagio fountains. “There’s no reason to put up a board with the prior numbers if it’s a completely random game,” Sundali says. “But they know the bettors care about the prior numbers.”
They care because they believe there are patterns to be cracked. The casinos are targeting clients who believe in the gambler’s fallacy. They’re providing just enough information for those people suffering from this particular bias of the mind to make regrettable decisions. Those electronic scoreboards might as well be neon billboards that scream: please give us all of your money.
That brings us to the next thing that Croson and Sundali studied: Did the gamblers who embodied the gambler’s fallacy also believe in the hot hand? They did. When those people playing games of chance thought they had the hot hand, they didn’t leave the table until they cooled off. They ordered the same drinks and performed the same lucky rituals as long as they kept winning. They did everything in their power to keep playing for the reasons that Stephen Curry tries to keep shooting: because they’re hot.
Croson and Sundali looked at the gamblers at this particular roulette table in this particular casino and found that 80 percent of their subjects walked away from the table after losing, but only 20 percent voluntarily stopped playing after a winning spin. You might think it makes sense to leave after winning and quit while ahead. This is roulette! You’re going to lose eventually. It would be smart to take your profit and splurge on a steak dinner. Maybe that’s what you believe. But the casinos know it’s not how you behave.
This wasn’t the only evidence of the hot hand they uncovered. Croson and Sundali also learned that if gamblers won fifty-fifty bets, they became more aggressive. The roulette players who won their fifty-fifty bets spread their chips over fourteen numbers on the next spin.
The unlucky ones who lost made only nine bets the next time around. The last thing that Croson and Sundali found when they replayed the videos of three nights at the roulette table was their coolest discovery yet. It was one thing to determine that the average roulette player believed in the hot-hand fallacy and the gambler’s fallacy. But what about the individual roulette player? It turns out the people who bet according to the gambler’s fallacy were the same ones who bet according to the hot-hand fallacy. That is, if you bet on red after black, you also probably bet more after winning a bet. “People seem to believe that people can get ‘hot,’” Ayton and Fischer had concluded, “but that inanimate devices cannot.”15 Croson and Sundali also emphasized that phrase: “seem to believe.” “We don’t actually know their beliefs,” Sundali says. “All we’re measuring is their behavior.” That was the point of bringing their experiment to the casino. They could evaluate what their subjects did instead of what they said they would do.
But what if we extend the principles of the gambler’s fallacy beyond the neon lights of the casino? What if they apply to baseball umpires, too? And what if people with jobs where the stakes are greater than calling balls and strikes also behave according to the gambler’s fallacy?
What if that bias is what determines whether a famous Iraqi sculptor gets asylum in the United States?
The paper was called “Refugee Roulette.” It was the most comprehensive study of U.S. immigration ever published, based on more than four hundred thousand asylum cases, including those overseen by the people entrusted with making the single most important decision of another person’s life: judges.16 It was an explosive piece of scholarship, and the law professors who wrote the paper made sure it would be read as widely as possible by giving it that name. The conclusion of their analysis of asylum applicants like Alaa Al-Saffar was a powerful repudiation of the way that justice is supposed to work. “Whether the asylum applicant is able to live safely in the United States or is deported to a country in which he claims to fear persecution,” the authors wrote, “is very seriously influenced by a spin of the wheel of chance.”17
They were right. And yet they had no idea how right they really were.
The upshot of their paper was that immigrants subject themselves to that wheel of fortune as soon as they apply for asylum. Their odds of staying in the United States shift based on circumstances beyond their control. There is no such thing as a level playing field. In fact the same application could easily produce different outcomes. Chinese asylum seekers win 7 percent of their cases in Atlanta and 76 percent of their cases in Orlando. One immigration judge grants asylum in 6 percent of cases. Another immigration judge in the same courthouse grants asylum in 91 percent of cases. In cities like Miami, New York, and Los Angeles, 32 percent of the judges deviate wildly from the average rates. The outliers aren’t the exceptions. They’re the expectation.
“There is remarkable variation in decision-making from one official to the next, from one office to the next, from one region to the next, from one Court of Appeals to the next, and from one year to the next,” the researchers wrote.18
The most startling takeaway of this disturbing study was that nothing in a case was more important than the judge—not the asylum applicants, not the country they were coming from, not the skills they would bring to the United States, and not even why they were fleeing in the first place. There is a great deal of randomness that goes into assigning judges. A deserving immigrant who could do as much for America as America does for him might very well get denied asylum simply because he got stuck with a strict judge in Atlanta instead of a lenient one in Orlando.
But the process is especially cruel to asylum hopefuls because it’s not only about the who and where. It’s also about the when.
Their chances of staying in the United States depend on whether the judge who was randomly assigned their case recently granted asylum in a completely unrelated case. That’s how arbitrary it can be. The asylum court is basically a casino.
Remember those economists who wrote that paper about baseball umpires? Baseball umpires weren’t the only people they studied for signs of the gambler’s fallacy. They also studied the decision-making habits of a collection of experts whose decisions actually mattered. They studied asylum judges.
Bruce Einhorn was one of their subjects. Before he was a U.S. Immigration Court judge, Einhorn worked for the Department of Justice, where he quite literally helped write the nation’s asylum law. But first he was an undergraduate student at Columbia University, curious about the sort of psychology that can shape a judge’s supposedly impartial judgments. “I find that stuff fascinating,” he says. “I don’t think all judges do.” Einhorn trained laboratory rats in the Skinner box and studied their behavior. He ended up learning something about himself along the way. “I learned that I don’t want to be a rat in a box,” he says.
That much he proved in his two decades on the bench. Einhorn was a generous judge. He had a higher asylum grant rate than most of his peers. He did not say yes to everyone who came before him, but he said yes enough that it became clear that his default response was not to say no. “It’s always easier to deny relief than to grant it,” he says. There were a surprising number of incentives to saying no. Anyone who said yes as often as Einhorn risked putting himself in the crosshairs of bureaucracy. “You’re facing the possibility that you’re known as a wimp,” he says, “instead of being known as a person of integrity who perceives himself to be what he is: a judge.”
That even the people tasked with the authority of federal judges could look at the same cases and come to a wide variety of conclusions made an economist named Kelly Shue wonder if they were also prone to cognitive biases. She mentioned to her colleagues Toby Moskowitz and Daniel Chen that she had some data pertaining to asylum judges. She wanted to see if these real people in the real world with real stakes were subject to the gambler’s fallacy. The caseloads are completely random, and judges are encouraged to complete as many as possible in as little time as possible, which makes the asylum court a breeding ground for bias.
What happens in one asylum case theoretically should have no effect on what happens in the next case. The problem with theory is that it doesn’t account for the whims of people wearing robes to work. The judges are not robots. They are human beings like Bill Miller, the Major League Baseball umpire.
Judges occupy a complicated place in public society. We give them the authority to play god. They are our chosen enforcers of the natural order. But when they feel it’s their job to balance things out, they become the opposite of the basketball fans cheering for a streak to continue. It is the judge’s obligation to ignore the streak and make an impartial call. They often fail to uphold that standard. They notice the streak. They believe there shouldn’t be a streak. So they end the streak.
Shue, Chen, and Moskowitz analyzed more than 150,000 decisions from 357 asylum judges. They calculated the average grant rate to be 29 percent. But when they looked at the sequence of cases, they could identify when the average shifted. And they found that judges were less likely to grant asylum immediately after they granted asylum in their last case. That is crushing. It means that an immigrant in need of asylum—someone who has already survived terrible hardship, lived through unimaginable misery, and outlasted awful odds—is automatically penalized for something that had nothing to do with the application. The depressing thing is that the statistics get more depressing. If a judge has granted asylum in two straight cases, he is 5.5 percentage points less likely to give asylum in the next case than if he’d denied two straight cases, regardless of the merits of the applicant. The judges were as vulnerable to the gambler’s fallacy as the baseball umpires and the drunk roulette players. And someone like Alaa Al-Saffar could be screwed by this spin of the wheel.
“The judges understand that cases are brought to them randomly, but as soon as you say random, we have a really strange sense of what random means,” Moskowitz says. “Most of us don’t understand it very well. They think that random means I have six cases today, and half should be positive and half should be negative. They don’t think you can get three in a row of the same type followed by three in a row of a different type. They think it should be alternating. And it’s just not.”
If anyone had bothered asking him to fix the system, Moskowitz would have proposed one simple tweak: change the way that asylum judges get assigned cases. He thinks every asylum applicant should be heard twice. “The judge makes the decision, and there’s another judge who reviews the decision,” Moskowitz says. “But they get to see the cases in a different order.” That would eliminate the issue of two judges looking at the same application and seeing different cases simply because one of them has just granted asylum and the other has just denied asylum. It isn’t the most practical solution, considering the judges already have far too many cases on their dockets without doubling their workload, but it is the right one.
Moskowitz likes this idea so much that he’s implemented the behavioral adjustment in his own life. When his teaching assistants graded exams, he made sure that each one was graded twice, at which point he averaged the scores together. But after brainstorming ways to beat the gambler’s fallacy, Moskowitz realized the flaw in his own system. “There could be some bias there,” he says. “You see a couple great exams in a row, and it might influence what you do on the next exam.” He took his own advice. His teaching assistants are now given the same exams to grade but in a different order. “That,” he says, “is a pretty fair and unbiased assessment.”
Alaa Al-Saffar wouldn’t have such a luxury. There would be several variables in the complicated equation of his asylum: which judge he got, where he got him, and when he got him. That was the human effect of the gambler’s fallacy. The judges weren’t betting on black or red, and they weren’t calling balls or strikes. They were deciding whether he lived or died.
Alaa Al-Saffar moved into a community for senior citizens in Southern California, a few turns off a road called Avocado Avenue, not long after he applied for asylum in the U.S. He converted his garage into a cramped artist’s studio and got to work.
His relocation had awakened the creative energy inside of him. “I was exploding,” he says. The cheapest means of expressing himself was painting. He produced a series called Lovely Dancers, a collection of buxom women in various states of undress. “In my country, if I do this, they kill me,” he says. But once he was allowed to do anything he wanted, he wanted to do something that he’d never been allowed to do.
Al-Saffar was settling into his unlikely, vaguely American life when I visited him on a blazing summer afternoon. I had read a story in the San Diego Union-Tribune about him, and now I was sitting across from him. He greeted me with a bottle of water and two Starbucks Frappuccinos, vanilla and mocha, insisting that I choose one. He wore the artist’s uniform of loose black T-shirt, black pants, and socks with sandals. He smelled of stale cigarettes. He sat behind a table with an ashtray, scattered pencils, and a laptop playing highlights from that day’s soccer matches. He barely had to stand to get to his easel. There were artifacts from his old life scattered around his workspace, and he dusted off a few shoeboxes with his press clippings, his stamp that won a national contest, and photos of his sculptures in all their glory. He even removed a thick binder from the end of his bookshelf and beamed with pride as he showed me his secret vice: a hidden stash of ouzo and arak.
He spent most of his time in this studio connected to his home. He’d been offered a proper workspace at a nearby college, but he said the gas would be too expensive. When his friends promised to drive him, he begrudgingly declined. “It’s too much,” he said. “That’s a place for a younger man.” Al-Saffar liked working from home. That was how he preferred to work in Baghdad, too. The proximity was crucial to him considering he was still always thinking about work. To him there was no such thing as work-life balance. “It’s my life,” he said. “Without work, there is no life.”
But there were two issues with his work and his new life. The first problem was the rocks. When he visited his local Blick Art Materials shop, he was surprised to learn that even the smallest hunks of sculpting material stretched his meager budget. The second problem was one that he couldn’t solve by painting instead of sculpting.
Al-Saffar had been in Southern California for several years already, and he still couldn’t say with any confidence if he would be allowed to stay. He’d left his old life and sought a new life at the exact moment in American history when the people who deserved compassion were treated with disdain. He couldn’t escape his old life because his new life was now at the mercy of the U.S. asylum system.
It was a system that required Al-Saffar to scale a small mountain of paperwork before he could even formally apply for asylum. The application contained such tedious instructions that anyone who finishes reading them should be granted citizenship on the spot. There was a purpose to all this paperwork. It was meant to suss out who was serious about needing to be saved. Al-Saffar had to demonstrate a credible fear of death back in Iraq and prove that he was under persecution for his race, religion, nationality, politics, or, in his case, membership in a particular social group. He recalled the threatening letter under the garage door that his daughter found and explained he was an artist who feared harm if he were to return to Iraq.
The next step in the asylum process was an interview with a Department of Homeland Security asylum officer whose job it was to put a face to all that paperwork. Al-Saffar was granted his first interview a few months after he applied. It seemed fast, all things bureaucratic considered, and his family was elated. This was surely a good sign. Once he was in a room with an officer, Al-Saffar attempted to explain all the reasons he was seeking asylum. “I want to do something here—for America,” the Iraqi sculptor said. It was a heartwarming gesture that would have little effect on his application. Instead he would have to convince the government that his motives for staying in the United States went beyond his burning desire to make art. He would be allowed to stay only by arguing persuasively that there wasn’t another choice. Al-Saffar had to prove that he was petrified. The asylum officer charged with establishing the credibility of his fears needed to know more about Al-Saffar’s past, which meant he needed to know whether he knew Saddam Hussein.
I am poor man, Al-Saffar recalled saying. I am artist.
Did he work for Saddam Hussein?
Yes.
What kind of work?
I am sculptor, he pleaded.
Did he feel in danger when he was in Iraq?
In my country, I need five eyes: one, two, three, four, five, he said, pointing to the left side, right side, and back side of his head.
And how many eyes did he need in America?
Just one, he said.
The officer who interviewed Al-Saffar could have handled his application in several ways. He could have granted him asylum on the spot. He could have denied him asylum. Or he could kick his case down the road and refer him to a federal immigration judge to decide whether he deserved asylum.
It had been years since that interview and months since his last contact with the asylum office when I found myself drinking Frappuccinos on that hot summer day with Al-Saffar. The U.S. asylum system was in crisis. The wave of migrants from Central American countries flooding the border had overwhelmed the officers and immigration judges responsible for the backlog of applications. The immigration system would have collapsed under the weight of a 100 percent or 200 percent or 300 percent increase, but what actually happened in the years before Al-Saffar applied for asylum was a catastrophic deluge. The real number of asylum cases had increased over the course of five years by 1,750 percent.19 The immensity of that figure is impossible to comprehend at the scale of the U.S. immigration system. So let’s simplify it. Imagine you’re in the habit of keeping a daily to-do list. On a busy day, you have ten tasks. You can derive a sense of satisfaction when you cross off every item. But an increase like the one that disturbed the asylum system would amount to ten chores becoming 185. That’s no longer a daily to-do list. It’s a monthly calendar packed into twenty-four hours.
There were 320,663 people like Alaa Al-Saffar with pending cases,20 enough to fill four NFL stadiums, and the immigration courts were overworked to the point where they weren’t actually working. The backlog kept growing. Meanwhile the government had adopted a last-in, first-out approach in which the first applicants to be heard would be the ones who arrived last. The agency was plowing through asylum filings at such a glacial pace that it seemed like it would never get to Al-Saffar’s. With so many migrants at the border, it was easy to ignore the asylum candidates who already had temporary visas, the hopeful Americans like Al-Saffar. The month that I went to see him ended with the affirmative asylum division of U.S. Citizenship and Immigration Services clearing about seven thousand cases and leaving hundreds of thousands on the docket. There were some who were granted or denied asylum on the spot, but the great majority of the applicants were interviewed and then referred to federal judges. They joined Al-Saffar in asylum purgatory. “Each way, you are dead,” he says. “Sleep or awake.”
During the years that his application was on hold, his family regularly called the local asylum office for an update. They were tortured with the same response each time. “Pending, pending, pending,” Zinah says. The whole ordeal had been so anguishing that she was no longer sure she’d made the right decision encouraging her father to leave his old life in Iraq and seek a new life in the United States. That was how dreadful the asylum process could be: Alaa Al-Saffar’s own daughter thought he might be happier in a place where there was a good chance that he would be killed for being an artist. “I wanted him to stay here for his safety,” she says. “But now, believe it or not, after all this waiting, the case is still pending. I regret it. Did I do the right thing by keeping him here? I don’t know. He for sure misses home.”
Al-Saffar liked to walk around his new home to clear his mind of this existential torment. He waved to his neighbors with American flags in their rock gardens. He was a popular guy around the community. His friends rushed over to Zinah to say hello whenever she visited. “I want people to know how talented he is,” she says. “How sad it is that talented people don’t get recognized. I want him to get recognized before his time is over.”
But there was one more thing that Zinah wanted for her father.
“I want him to have a choice in life and a choice in art,” she says. “Not to always be dictated by somebody.”
Alaa Al-Saffar passed the time as he waited for the results of his asylum application the only way he knew how. He continued to make art. His goal was to build a monumental public sculpture in the middle of his American town. He wanted to give back to the place he was still hoping would take him in. He even made an immaculate little model of his magnum opus. It was a map of the United States combined with the American flag, held together by a word in English and repeated in the many languages of the famous Iraqi sculptor’s native land: “freedom.” He couldn’t be sure that he would get to experience it for himself.
Alaa Al-Saffar’s uncertainty. Jed Lowrie’s home run. Nick Hagen’s bumper crop and David Booth’s rules of basketball. Rob Reiner’s movies, Rebecca Clarke’s sonatas, and William Shakespeare’s plays. Mark Turmell’s addictive game and, of course, Stephen Curry’s shooting extravaganza.
By now we know what the hot hand looks like. But there is one thing we still don’t know.
Should we believe it when we see it?