Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won
Page 14
Less cynically, like any underdog, small schools also thrill at the chance to “make a name for themselves” and elevate their profile on the off chance that the team can spring an upset. (Who had even heard of the small Hawaiian school Chaminade before its basketball team’s momentous upset of a Ralph Sampson–led top-ranked Virginia team in 1982?) Even in defeat, the small school usually appears on national television, a big draw to potential recruits. Plus, the players leave with a sense of how far they are—sometimes it’s not very—from the next level of competition. In short, everybody usually benefits from this arrangement.
It’s worth noting that scheduling a “patsy” opponent can backfire financially, sometimes spectacularly. In 2007, the University of Michigan football team (ranked fifth in the nation at the time) lost its home opener to tiny Division I-AA Appalachian State. Not only did such a disgraceful defeat ruin any chance of Michigan playing in a big (read: well-paying) bowl game regardless of what happened the rest of the season, but such an embarrassing loss—at home!—surely had an effect on alumni donations.
We found that home schedule padding accounts for roughly half of the home team advantage in college football. If we adjust for the quality of teams—or look at in-conference games, where the conference and not the big schools sets the schedule—home team winning percentage drops from 64 percent to 57 percent. Incredibly enough, that 57 percent is almost the exact same rate at which the home teams win in both the NFL and Arena football. For college hoops the numbers are similar. Of the impressive 69 percent home court advantage in NCAA basketball, a little more than half can be explained by early-season schedule padding. Accounting for these scheduling biases and strength of opponent, the home advantage in college basketball declines to 63 percent, the same as in the NBA.
But scheduling bias gets us only so far. It accounts for half of the home field advantage in college sports; it partially explains the home field advantage in the NBA and NHL. In Major League Baseball and the NFL—and, as it turns out, in soccer as well—it doesn’t explain it at all.
Conventional Wisdom #4: Teams win at home because they are built to take advantage of unique “home” characteristics.
Fulfilling the Portland coach’s prophecy, the Spurs pulled away in the second half and beat the Blazers handily, 99–84, in a contest that was virtually uncontested. From the start, the Blazers competed with no urgency or passion, missing nearly two-thirds of their shots and giving only periodic consideration to defense, as if resigned to defeat. The postgame locker room hardly called to mind the picture of despair. Just one more game in Minnesota and the road trip would be over. “A few more days, man,” said the Blazers’ center, Greg Oden, who hadn’t even played that night. “A few more days and we get to go home.”
As for the Spurs, they played generally unimpeachable basketball. They competed capably, defended capably, and shot the ball well. Unmistakably, the star of the game was Parker, who darted around the court, scoring 39 points in 35 minutes. In a battle of images, Spurs coach Gregg Popovich gushed that Parker was “a superstud again.” Nate McMillan, the Portland coach, likened Parker to “a roadrunner blowing by us.”
Sterlingly as Parker had played, he was, by his own admission, at his best when paired with Tim Duncan. In addition to coaching the team, Popovich was the team’s chief architect, and his decision to draft Parker as a complement to Duncan’s inside presence was a coup that had paid immense dividends—not least multiple NBA titles. The twenty-eighth selection in 2001, Parker represents one of the great steals in recent NBA drafts. When making personnel decisions, Popovich told us that he considers dozens of factors: How will the players fit into the tapestry of the team? How much will they cost? How will they feel knowing they’ll be operating in the considerable shadows of the Spurs’ three stars? How will they acclimate to a “small market” that lacks the beaches of Los Angeles and the nightlife of New York and Miami?
What he doesn’t worry about is how they will play specifically in the AT&T Center. From arena to arena, the baskets are 10 feet off the ground and 94 feet apart, 15 feet from the free throw line. Throughout the NBA, the playing surface is standardized. The games are always played indoors in climate-controlled venues. Even the placement of the decals on the court must conform to league regulations. It’s the same in the NHL. For all intents, a rink is a rink is a rink.
In the NFL, each field is 120 yards long, including the end zones, and roughly 53 yards wide; but the climate and playing conditions can vary immensely. A December game in San Diego, California, is played in a much different environment than a December game in Buffalo, New York. Is the home field advantage in football influenced by teams tailoring their rosters to the weather?
We didn’t find that to be the case.
Much as broadcasters talk about those poor teams from the tropical precincts—say, the Miami Dolphins and San Diego Chargers—faltering in the “frozen tundra” of Lambeau Field in Green Bay or thermally challenged lakeside stadiums in Buffalo and Cleveland, climate, we’ve found, is largely irrelevant. If NFL teams are built to take advantage of their home weather, we should observe cold weather teams winning disproportionately in cold weather games. We also should observe teams playing poorly when venturing to markedly different climates. Finally, we should observe that “domed teams,” because they play in climate-controlled chambers, are simply insensitive to home weather. Think of this last situation as a placebo test. If the control group experiences similar reactions even though they weren’t administered any treatment or medicine, you know it wasn’t the drug but something else that caused their reactions. Similarly, if domed teams seem to perform differently depending on the outside weather, it must be something else influencing the results. For example, just as the weather tends to get worse late in the NFL season in December, teams may play better at home late in the season, having nothing to do with weather at all.
After studying data from every NFL game from every season between 1985 and 2009—nearly 6,000 games—and matching those games to the outside temperature and wind, rain, and snow conditions, we found that cold weather teams* are no more likely to win at home when the weather is brutally cold, nor are warm weather teams more likely to win at home when the temperature is awfully hot. And the home winning percentages for dome teams immune from extreme weather conditions—our placebo test—do not vary with the weather any more than they do for cold and tropical weather teams. Even looking at the most extreme cases, when a warm weather team has to play in extremely cold weather or a cold weather team plays in humid and hot conditions, there is little to no unusual effect. Contrary to conventional wisdom, weather gives a team no additional home advantage. Either teams are not built to suit their home weather conditions or, if they are, it doesn’t seem to have much effect on the outcome of games.
What about baseball? After all, not only do the playing conditions vary but—one of the sport’s great appeals—each stadium is unique. Don’t the home players have an advantage, as they’re more familiar with their ballpark’s idiosyncrasies? Don’t teams stock their rosters with players who are better suited for their park’s features? And couldn’t this influence the home field advantage in baseball?
Yes and no.
There’s no question that the Boston Red Sox, for instance, have an advantage playing at Fenway Park. The Sox outfielders know the Green Monster, the notorious 37-foot-high left field wall, the way Thoreau knew Walden Pond. Unlike the opponent, they’re well acquainted with caroms and angles and the effect of the wind. (There’s even an ersatz Green Monster scheduled for construction at the Red Sox spring training facility so the organization’s minor leaguers can familiarize themselves with the wall’s distinctive features.) Similarly, Sox hitters know that although the Green Monster is high, it’s also deceptively shallow—barely 300 feet from home plate—and they adjust their swings accordingly. Surely this has an effect on Boston’s home winning percentage.
You might also surmise that baseball players are fam
iliar with the unique optics of their home park. The home batters see the ball better; the visiting pitchers exhibit less control. But we already know that once we’ve controlled for other factors—pitch count, game situation, and so forth—players don’t hit the ball appreciably better at home and pitchers don’t throw appreciably less accurately on the road. Thus, that can’t be the reason home teams win more games.
What about the notion that baseball teams win more games at home because they tailor their rosters to the idiosyncrasies of their ballparks? The teams that play in parks with, say, shallow right field porches recruit more left-handed hitters. The teams with uncharitable dimensions recruit superior pitchers and speedy outfielders. How much does this affect the home field advantage?
Since it would be impossible to consider every ballpark and how different types of players might be better suited to each, we looked at the most obvious case in baseball and the one likely to have the biggest impact: “hitter-friendly” ballparks versus “pitcher-friendly” ballparks. The Sabermetrics community helped us identify which ballparks historically were hitter-friendly, using total number of runs, hits, extra-base hits, and home runs produced in each ballpark by all teams each season. We then asked: Do teams from hitters’ parks outhit their visitors by more than teams from pitchers’ parks do? If teams from hitters’ parks are being stacked with sluggers, we should see them outhit their visiting opponents by a wider margin than that of teams from pitchers’ parks. Yet we don’t. Teams from hitters’ ballparks outhit their visitors by the same amount as home teams in pitchers’ parks do—same differences in batting average, home runs, doubles, triples, slugging percentage, and runs created. We even found this to be the case when a hitting team plays host to a team from a pitchers’ ballpark, where you’d expect the widest difference.
We also looked at how teams from hitters’ ballparks play away from home. If their lineup is stocked with hitters, they should hit better than other teams no matter where they play. (Plus, by looking at other ballparks we also remove any other home advantages, such as crowd, familiarity with field of play, and travel.) But we found that teams that play at home in hitters’ ballparks hit no better on the road than teams that play host in pitchers’ ballparks do when they’re on the road, even going so far as to control for the same stadium. That is, the Colorado Rockies (who play in a hitters’ park) hit as well as the New York Mets (who play in a pitchers’ park) when they each play in Busch Stadium in St. Louis.
All this evidence indicates that either teams aren’t stacking their rosters to suit their home stadiums or, if they are, it’s not making much of a difference. Bear in mind that the home field advantage in sports is lowest in baseball. Even if “roster tailoring” is a factor in some cases, it doesn’t get us very far in explaining the home field advantage phenomenon overall.
We should add that deception and “dark arts” don’t seem to be much of an explanation for the home advantage, either. In past eras, it was different. In 1900, for instance, a shortstop on the visiting Cincinnati Reds noticed that the Phillies’ third-base coach stood in a puddle each inning. When the shortstop investigated, he found that under the puddle was a wooden box. It turned out that a Phillies backup catcher sat in the outfield bleachers armed with high-powered lenses and stole signs from the visiting team. Then, using a buzzer that was connected to the wooden box with wires that ran under the field, he used Morse code to convey the pitch to the third-base coach. The coach then relayed the information to the batter. Little wonder the Phillies won two-thirds of their games at home and fewer than half on the road.
Through the years, other home teams have used elaborate plots to steal signs from the visitors. For years, home groundskeepers in baseball would water the field into a bog when speedy visiting teams were in town. The Boston Celtics were notorious for jacking up the heat in the visitors’ locker rooms so that halftime resembled a session in a sauna. The University of Iowa football team once ordered the visiting locker rooms painted pink, hoping it would make the opponent feel passive or emasculated. It’s unclear if any of this worked—and it probably didn’t—but because of the standardized league rules, the stiff deterring punishment for cheating, and surveillance technology, it would be hard to pull off this kind of skulduggery today.
So let’s take stock of all we know: When athletes are at home, they don’t seem to hit or pitch better in baseball, shoot free throws better in basketball, slap goals better in hockey shootouts, or pass better in football. The home crowd doesn’t appear to be helping the home team or harming the visitors. We checked “the vicissitudes of travel” off the list. And although scheduling bias against the road team explains some of the home field advantage, particularly in college sports, it’s irrelevant in many sports. The notion that teams are assembled to take advantage of unique home characteristics isn’t borne out, either.
Yet if home teams are winning more games so consistently, players on those teams surely must be doing something better than their opponents. What else is giving the home team its sizable edge?
Thanks to the quirks of the NBA schedule, the Blazers and the Spurs played again four nights later. Duncan, San Antonio’s exceptional big man, had regained his health and was back in the lineup. If the Spurs had beaten the Blazers by 15 points when he was on the bench, surely they would crush them with Duncan in the game. Right?
But this time the two teams played in Portland. This time the Blazers would have the exuberant PA announcer, the dance teams, the 20,000 partisan fans. This time the Blazers shot more free throws. This time the Spurs committed more fouls and turnovers. This time the Blazers won 102–84, a whopping 33-point swing from their game only 96 hours earlier.
Maybe these athletes and coaches are right, after all, to adopt a defeatist attitude when heading off on the road.
But why?
* They include every league in Uruguay, Australia, Paraguay, Scotland, Japan, South Africa, England, Argentina, the Netherlands, Spain, Greece, Germany, Chile, Mexico, Italy, Honduras, Russia, Costa Rica, Brazil, France, El Salvador, Peru, Venezuela, and the United States.
* For the playoffs an issue that has to be taken into account is that teams are typically seeded so that better teams get to play at home and worse teams are on the road more often. For this reason alone we would expect the home team to win more often in the playoffs. That is, home teams would win more of their fair share of games no matter where they played. Therefore, to compute the home field advantage in the playoffs accurately, we have to adjust for the quality of teams. Specifically, if Team A hosts Team B and Team A is a much better team, we first calculate how often you’d expect Team A to win if it played on a neutral field and compare that to how often Team A actually beats Team B when playing at home. The results? If you adjust for team quality, the home field advantage is almost exactly the same during the playoffs as it is during the regular season: For MLB it is 54 percent, for the NBA it is 61 percent, for the NFL it is 57 percent, and for the NHL it is 57 percent. These numbers are, once again, remarkably consistent.
* Although shootouts occur only after the game has been tied and hence the two teams are evenly matched, implying that a 50–50 split of shootouts should be expected, the same could be said of overtime periods. In overtime, the teams also enter tied, yet the home ice advantage is still present in overtime.
* If you were curious, pitches in the dirt occur 1.5 percent of the time at home and on the road.
* Cold weather teams are Buffalo, Pittsburgh, New York Giants, New York Jets, Philadelphia, Washington, Baltimore, Green Bay, Chicago, Denver, New England, Cleveland, Cincinnati, and Kansas City. Dome teams are obvious, and the rest are considered warm weather teams. Note, too, that there are several teams from a cold weather climate that play in domes: Minnesota, Indianapolis, St. Louis, and Detroit. We also adjust for the prior winning percentage of each team in each game to control for team quality.
SO, WHAT IS DRIVING THE HOME FIELD ADVANTAGE?
Hint: Vocal fans
matter, but not in the way you might think
It had the makings of a nearly perfect day. Jack Moore had just finished his sophomore year at the University of Wisconsin and was home for a few summer weeks, living with his folks. Marooned in the Mississippi River town of Trempealeau, Wisconsin, Jack was blissfully free of pressure, with generous rations of free time. He had a job coaching baseball, but the games didn’t start until the evening. On this Friday of the 2009 Fourth of July weekend, Jack’s beloved Milwaukee Brewers were playing an afternoon road game against their rivals the Chicago Cubs.
Air-conditioning blasting, Jack flicked on the cable to the regional sports network and sat down on the couch to watch. The Brewers were coming off a magical 2008 season in which they won 90 games and reached the playoffs. In the off-season, Milwaukee’s ace, C. C. Sabathia, was poached by the Yankees. It was the numbingly familiar fate of a small-market team: The Brewers had been unable to match New York’s $161 million contract offer. Jack was okay with that. A math major, he knew the economic realities and understood why Milwaukee could not afford to retain a star at those prices. Besides, the Brewers’ 2009 incarnation was easy to root for, a fun team with a winning record, filled with young and energetic players.
The game was a rare Wrigley Field pitching duel pitting the Cubs’ ace, Carlos Zambrano, then a Cy Young Award candidate, against Milwaukee’s veteran Jeff Suppan. The game was tied 1–1 after nine innings, which was all good with Jack, a former high school baseball player who was thoroughly capable of appreciating a low-scoring affair. “It was one of those games,” he recalls, “that remind you why you like baseball so much.”