In addition, crowd size, which we contend affects referee judgment, has more influence in the sports with the greatest home field advantage. Crowd size matters most in soccer (the sport with the highest home field advantage) and least in baseball (the sport with the lowest home team winning percentage) and is somewhere in between for the other sports. This is also consistent with referees mattering more in some sports (soccer) than others (baseball).
To answer the second question, referee bias also explains why the home field advantage is the same for a particular sport no matter where it is played. Whether baseball is being played in the United States or Japan, whether it’s basketball in the NBA, WNBA, or NCAA or soccer in France versus South Africa, the rules and, more important, the role of the referee are essentially the same, no matter where the game is played.
Finally, to answer the third question, referee bias also explains why the home team’s success rate hasn’t changed over a century. Although sports have altered their rules over time—raising and lowering the pitcher’s mound, introducing a shot clock and the three-point line—the official’s role in the game hasn’t changed much. Umpires still call balls and strikes, referees still call fouls and penalties, and for over a century these calls have been made by human beings—none of them immune from human psychology.
Although we will never be able to measure or test all the decisions an official makes, if we can see that some biased judgments are being made, it is likely there are other biases going the home team’s way that we don’t see. Think of the father who comes home early from work and catches his teenage daughter kissing her boyfriend. He’s upset about the kiss, but he’s more upset about what else she might be doing when he doesn’t happen to be looking.
Knowing what we now know, let’s revisit that Cubs-Brewers game, the ten-inning affair that ruined a summer day for Jack Moore of Trempealeau, Wisconsin. You wouldn’t deduce this by scanning a conventional box score or watching a SportsCenter highlight. But after revving up the Pitch f/x results, it becomes clear that when the umpire erroneously called a ball on a 3–2 pitch in the bottom of the tenth inning, enabling the winning run to score for the hosting Cubs, it marked the culmination of an afternoon filled with unfavorable decisions against the visiting Brewers.
According to Pitch f/x, Cubs hitters failed to swing at 25 pitches that were strikes. However, nearly a third of them were incorrectly called balls. As for the Brewers, they failed to swing at 32 pitches in the strike zone, only a quarter of which were called incorrectly as balls. Advantage, Cubs. In high-leverage situations, when batters had three balls, not a single strike was called on a Cubs hitter even when the ball was in the strike zone—including, of course, the final pitch of the game. But for Brewers hitters facing three-ball counts, every pitch in the strike zone was called a strike and half the pitches outside the strike zone were called strikes! Big advantage, Cubs.
Overall, the Brewers were deprived of three walks to which they were entitled and the Cubs were given two walks on strikes that were erroneously called balls, including the game-winner. That’s a difference of five base runners in a game that ended with a final score of 2–1 in extra innings.
One last point: Recall how the closer officials are to the crowds, the more likely they are to favor the home team. Wrigley Field has about the smallest amount of foul territory in the Major Leagues, so the umpire is uncommonly close to the restless natives. And remember how attendance influences the home field advantage. Wrigley Field seats 41,118 fans and is generally nearly full. In fact, despite a long history of losing seasons, the Cubs have won 54 percent of their home games—above the league average. (It’s just that they have been terrible on the road.) That particular afternoon drew a crowd of 41,204—more than 100 percent of capacity with standing room only.
When that mass of humanity on Chicago’s North Side yelled at the players, they weren’t affecting the outcome. When they yelled at the umpire, well … that’s another story entirely.
* Eliminating intentional walks.
* According to Major League Baseball, the 11 franchises whose ballparks were equipped at various times with QuesTec were the Arizona Diamondbacks, Boston Red Sox, Chicago White Sox, Cleveland Indians, Houston Astros, Los Angeles Angels of Anaheim, Milwaukee Brewers, New York Mets, New York Yankees, Oakland A’s, and Tampa Bay Rays.
* Pitch f/x is now in every ballpark, and thus, one could argue, umpires are now monitored everywhere. However, Pitch f/x—unlike QuesTec—is not being used to evaluate performance. There’s a big difference between casually monitoring umpires and bosses formally monitoring umpires.
THERE’S NO I IN TEAM
But there is an m and an e
After concluding that defense doesn’t necessarily win championships, we decided to examine another shopworn bit of sports wisdom. Before young athletes are capable of lacing their sneakers and putting on their cleats, they’re invariably taught, “There’s no I in team.” This spelling lesson is, of course, meant to reinforce the virtues of teamwork, stressing the importance of unity and the corrosive effects of attempts at personal glory. But does it accurately capture reality?
It’s in basketball that the no-I-in-team cliché is most often tossed around. If we were to compile a list of the top, say, five or six NBA players over the last 20 years, it probably would include Michael Jordan, Kobe Bryant, Tim Duncan, Shaquille O’Neal, Hakeem Olajuwon, and LeBron James. If there were no I in team, those stars wouldn’t much matter. A team that formed a symphonic whole, with five players suppressing and sublimating ego for a common goal, could surmount teams with the “I” players, stars willing and able to play selfishly when the situation calls for it. But that’s rarely the case. Since 1991, every year at least one of those players has appeared in the NBA finals. Go back another decade and add Larry Bird and Magic Johnson and now at least one of the top eight players has been featured in all but one NBA finals series for the last 30 years. In other words, a remarkably small number of select players have led their teams to the vast majority of NBA titles. Lacking one of the best players in history has all but precluded a team from winning.
We wondered how likely it is that an NBA team without a superstar wins a championship, makes it to the finals, or even makes it to the playoffs. We can define superstars in various ways, such as first-team all-stars, top five MVP vote-getters, or even those with the top five salaries. Pick your definition; it doesn’t much matter. The chart below shows what we found for the NBA.
PROBABILITY OF WINNING AN NBA CHAMPIONSHIP, MAKING FINALS, OR MAKING PLAYOFFS GIVEN NUMBER OF SUPERSTARS ON THE TEAM
A team with no starting all-star on the roster has virtually no chance—precisely, it’s 0.9 percent—of winning the NBA championship. More than 85 percent of NBA finals involve a superstar player and more than 90 percent of NBA titles belong to a team with a superstar. The graph also shows, not surprisingly, that as a team gains superstars, its chances of winning a title improve dramatically. One first-team all-star on the roster yields a 7.1 percent chance of winning a championship and a 16 percent chance of making it to the finals. A team fortunate enough to have two first-team all-star players stands a 25 percent chance of winning a championship and a 37 percent chance of making the finals. On the rare occasions when a team was somehow able to attract three first-team all-stars, it won a championship 39 percent of the time and made the finals 77 percent of the time.
The numbers are even more striking when we consider the top five MVP vote recipients. A team with one of those players stands a 15 percent chance of winning it all and a 31 percent chance of making the finals. Having two of those players yields a 48 percent chance of winning the championship and a 70 percent chance of making the finals.
Really, it’s no mystery why the Miami Heat fans celebrated deliriously when the team lured LeBron James in the summer of 2010. When James “took his talents to South Beach” and joined Dwyane Wade and Chris Bosh, it made the Heat a virtual lock to go deep into the playoffs. Having the
MVP (James) in addition to two other all-stars makes the Heat 98 percent likely to make the playoffs, 70 percent likely to make the finals, and 36 percent likely to win it all.
At some level this stands to reason, right? The superstars are usually going to be concentrated on the best basketball teams. The average winning percentage of teams with a first-team all-star on their roster is 56 percent. Two first-teamers and it’s 63 percent. Have a top-five MVP vote-getter and your team wins 64 percent of its games; that in itself calls the “no-I-in-team” shibboleth into question.
Here’s where it gets interesting. Even after controlling for the team’s winning percentage during the regular season, teams with superstars do measurably better in the playoffs. That is, a top-five MVP candidate improves his team’s chances of winning a championship by 12 percent and of getting to the finals by 23 percent even after accounting for the regular season success of the team. This implies that superstars are particularly valuable during the playoffs—ironically, the time when “team” is relentlessly stressed by coaches, media, and analysts.
What about the notion that a lineup of five solid players is better than a starting five of one superstar and four serviceable supporting role players? One way to test this idea is to look at the disparity among a team’s starters in terms of talent. Controlling for the same level of ability, do basketball teams with more evenly distributed talent fare better than teams with more dispersed talent? Measuring talent is difficult, but one reasonable metric is salary. Controlling for the average salary and winning percentage of teams, do teams with bigger differences in salaries among their starting players fare worse than teams whose salaries are spread more evenly among the players?
We find the opposite, in fact. Teams with more variable talent across their players are more likely to make the finals and more likely to win a championship than teams with more uniformly distributed talent. Again, this suggests that a superstar with a relatively weak supporting cast fares better than the team with five good players.
The same holds for the NHL and even soccer. Without a prolific goal scorer and/or goaltender/goalkeeper, survival in the postseason tends to be short-lived. There may be an I in Major League Baseball, too, but the effects are considerably weaker, in part because the team is larger, making it harder for one player to change the overall make-up of the team. Nonetheless, the bulk of World Series titles and appearances belong to the teams with a handful of elite superstars, both hitters and pitchers. But there are some examples of championship teams without a starting all-star player. (Name a “star” on the 2003 Florida Marlins. How about the 2002 Angels?) But this makes sense. Although we focus on individual achievements in baseball, it’s hard for any one player other than the starting pitcher, who pitches only one game out of five, to take control of a game. Even the best hitters come to bat only once every nine times.
If the evidence suggests that in basketball, hockey, and soccer a handful of individual players are extremely valuable for success, why do so many coaches and commentators place such heavy emphasis on the team? Perhaps it’s because admitting that in these sports the star matters as much as he does blunts the incentive for the rest of the team. Though teammates may be less valuable than the stars, they still have some value. They’re needed to grab rebounds, pass, block, chase loose balls, and defend. Sure, the superstar makes a big difference, but he can’t do it alone. In that sense, the team certainly does matter.
Stars tend to recognize this delicate balance. Remember how lustily Michael Jordan embraced the conventional wisdom that “defense wins championships,” a phrase that galvanized his teammates? Nonetheless even Jordan felt differently about the “no-I-in-team” truism. He recognized that not all players were created equal.
At his 2009 induction into the Basketball Hall of Fame, Jordan gave a speech that revealed much about his turbo-powered competitive drive. He told a story of once scoring 20 consecutive points late in a game to lead the Chicago Bulls to victory. Afterward, he was admonished by Tex Winter, the Bulls’ eminent longtime assistant coach, “Michael, there’s no I in team.” Jordan recalled his response: “I looked back at Tex, and said, ‘There’s an I in win. So which way do you want it?’ ”
OFF THE CHART
How Mike McCoy came to dominate the NFL draft
At the Dallas Cowboys’ team headquarters in spring 1991, a sense of optimism was leavened with a sense of unease. The most iconic franchise in the NFL had recently been sold to a swashbuckler from Arkansas, Jerry Jones. Just as Jones had made a fortune in the oil and gas business by taking bold risks, he’d leveraged his entire net worth to buy the ’Boys in 1989 for $140 million—$65 million for the team and $75 million for the stadium. Jones paid $90 million in cash and borrowed the rest against personal assets. His interest payments were $40,000 a day. His banker told him he was nuts. His father told him the same thing, but hey, this was the Dallas Cowboys.
The beginning of the Jones regime did not portend greatness. One of his first acts was to fire the longtime coach, the legendary and dignified Tom Landry, and replace him with his polar opposite, Jimmy Johnson, a brash (if notably well-coiffed) renegade who’d never coached in the NFL. He was fresh from a spectacularly successful, spectacularly controversial tenure at the University of Miami. Johnson was an old teammate and running buddy of Jones’s from their days at the University of Arkansas, and that counted for plenty. Jones conferred full football decision-making powers on Johnson, forcing out Tex Schramm, the only president the Cowboys had ever employed and, as his name suggests, a man with deep roots in the state. Worst of all, the Cowboys put a rotten product on the field, winning just 8 of 32 games in Jones’s first two seasons as owner.
As is so often the case for sad-sack teams in the socialist world of team sports, hope for the Cowboys came in the form of draft picks. The best picks go to the worst teams, though this can be fool’s gold. In 1991 Dallas held a Texas-sized helping of selections: ten picks in the first four rounds, five in the first round alone, including the very first pick. When Johnson crowed, “We’re dictatin’ this whole draft,” he wasn’t met with much resistance. But if this draft had the potential to be a turning point for the franchise, it also had the potential to be Jones’s personal Waterloo: You picked early and often and you still couldn’t turn the franchise around?
Reflecting the new owner’s passion for speculation, Dallas traded more than any other team in the league. In the 26 months since Jones had bought the team, the Cowboys had made 29 deals, including the infamous “Herschel Walker trade” with the Minnesota Vikings, an 18-player swap that was less a transaction than an act of larceny. In exchange for the aging Walker and four modest picks, the Cowboys received a bounty of five players plus eight future picks. Two decades later, this still stands as a benchmark for lopsided trades. It ended up paying substantial dividends for the Cowboys in the years to come, but Johnson recalls that at the time it made other teams wary of doing business with the Cowboys for fear of getting similarly scalped.
However, with their stockpile of selections, it was logical to assume that on draft day other teams would offer to trade picks. That created a problem: With the clock ticking, how was Johnson to know whether he’d be better served to entertain the offer of swapping, say, one of the team’s third-round picks for Green Bay’s fifth- and seventh-round picks? “I can’t assess value that fast,” Johnson complained at a predraft meeting among Dallas executives. “No one can!” There were nods all around.
Mike McCoy, a Cowboys executive and minority owner, piped up: “Let me see what I can come up with to fix that.”
In 1981, McCoy, then an Arkansas petroleum engineer, had partnered with Jones to form the Arkoma Production Company. Jones made the deals, and McCoy was the driller. As Inc. magazine once put it, they perfected a low-risk strategy of drilling holes that other companies suspected were fertile but wouldn’t commit to exploring. The success rate on “wildcats”—speculative wells that come cheap but seldom yield a payoff—is around
5 percent. According to Inc., Jones and McCoy struck 2,000 wells and made money on more than 500 of them. In 1986, a state-run gas company in Arkansas, Arkla, bought Arkoma for $175 million. (Aside: Sheffield Nelson, the former head of Arkla, ran for governor in 1990 and lost to the incumbent—an ambitious Democrat named Bill Clinton—in part because of questions surrounding the generous payout he authorized to Jones and McCoy.) When Jones purchased the Cowboys, McCoy joined him. Jones, after all, often referred to McCoy as “one of the brightest minds I’ve ever been around.”
This business of trying to quantify draft picks? As McCoy saw it, he’d spent his entire life solving numerical puzzles and trying to tilt the odds in his favor. This was simply another application. He recalls thinking to himself: “How hard could this really be?”
Not hard at all, it turned out. McCoy asked Dallas’s player personnel department for a list of all the NFL trades that had been conducted on draft day going back four years. He assigned an arbitrary point value to the first pick in each draft round and then used all the prior draft trades to refine the relative values of every pick. As McCoy recalls, “The point was to create a graphical depiction of how the NFL valued draft picks, based upon their own actions—not how they should have been valued.”
McCoy didn’t make any subjective judgments; he simply took the existing information and plotted the data points. After two days of “fiddling” (his word) and plotting picks on a graph, he presented a chart that assigned a numerical value to every draft position. “It was basically a price list,” McCoy says. “It was what Walmart would do, only with football players, not jeans or toothpaste.” The first pick of the first round was worth X. The last pick of the last round—known as Mr. Irrelevant—was worth only Y. A sample of the point totals appears in the table below.
Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won Page 18
