Location Variations in Professional Football
Acker, Jon C., Journal of Sport Behavior
"H.L. Mencken's average Boobus Americanus is a wild-eyed, cheering sports junkie. He is the 10th man at Dodger Stadium (Baseball), the 12th man at the Coliseum (Football), the 6th at the Forum (Basketball & Hockey). He is the home-field advantage sports reporters write about (Shirley, 1986, Part 3, p. 1)."
While home-field advantage is often personified, like in the above statement, it is a much more complicated phenomenon. The terms "home-field advantage," or "home advantage" are used to describe the persistence of home teams winning a majority of games. It has been shown that home teams win between 53% and 70% of their games depending on the sport and level of competition (Courneya & Carron, 1992). Home-field advantage (HFA) is a subject which has been extensively studied since the 1970s, but has resulted in very few concrete conclusions, since "little systematic research has been carried out to determine its nature or causes (McGuire et al., 1992, p. 149)." This is "due to the difficulties in controlling home advantage variables," which make conclusions inferential (Courneya & Carron, 1990, p. 313). Courneya & Carron (1992) list the wide and varied explanations for HFA which include:
biological-based theories of territoriality and circadian rhythm changes, to social psychological-based 'drive' theories (e.g., social facilitation) and social cognitive theories (e.g., self-preservation, perceived social support), to sociological-based theories of community celebration (e.g., ritual integration). (p. 14)
Many studies have looked at numerous aspects of HFA, including spectators' impact on officiating (Lehman & Reifman, 1987; Askins, 1978), effects of travel and the length of a home stand (Courneya & Carron, 1991), its presence in women's sports (Gayton, et al., 1987), audience size influences (Schwartz & Barsky, 1977), and team quality (Schwartz & Barsky, 1977; Snyder & Purdy, 1985), just to name a few. The purpose of this research, however, is to quantify HFA spatially for individual National Football League (NFL) teams and their metropolitan areas, something that has not been attempted to the author's knowledge in any sport on any competition level.
Professional football was chosen for this study due to its ever increasing popularity and preeminence in athletics and entertainment. Football is overwhelmingly the public's sport of choice. A Gallup poll (Gallup & Newport, 1992) discovered that 38% of Americans chose football as their favorite sport to watch, followed by baseball at 16% and basketball at 12%. Also, professional football is, perhaps, only surpassed in sporting fervor by its collegiate cousin. However, as the NFL continues to expand, adding Jacksonville and Carolina in 1995, all the while the Canadian Football League (CFL) further infiltrates the U.S., one cannot fathom the magnitude the sport would optain in the future should these two leagues coalesce as the American Football League (AFL) did with the NFL in 1966 creating the aptly named Super Bowl.
As Shirley (1986, Part 3, p. 1) pointed out, in football, HFA is often called "the 12th man," suggesting it is akin to the advantage of having an additional player. Others equate it to the home team's familiarity with site tangibles like the stadium, the local climate, field conditions and other peculiarities. After the 1994 National Football Conference championship game George Seifert, coach of the 49ers, stated:
"We've played on bad fields and artificial turf and in stadiums where the noise is mindboggling. That's the home-field advantage and you learn it after you've been in the league awhile (Amore, 1995, p. C4)."
Certain teams possess reputations as having profound home advantages. Denver is one such location. Rabun (1986) claimed that:
"Denver is blessed with a tremendous home-field advantage. The stadium is filled with orange-garbed fanatics who can drown out the opposition's signals and then, on cue, take on a cathederal-like silence so their heros can hear the snap count."
But is there really an advantage, one that can be calculated for each team and its location, rather than simply looking at wins and loses? Although it may be difficult to qualify and quantify the causal variables that lead to HFA, it is relatively easy to quantify their effects on each sport and on individual teams. This study seeks to answer the following research questions: (1) How does HFA differ spatially for each NFL team and its metropolitan area, (2) How does HFA change with varying team strength and (3) How do domed stadiums affect HFA?
Data and Data Attributes
This study utilizes game scores from the seven NFL seasons from 1988 - 1994. During those years the league was comprised of twenty-eight teams, each playing a sixteen game season schedule (8 home and 8 away games). In all there were 1,568 home games over the study period. Only regular season games were included in the determination of a HFA. In each game the difference in scores between the home team and the visiting team was calculated. The resultant integers were categorized into attributes of individual teams and the number of season wins. To be consistent, and provide each team with an equal number of consecutive years at the same location, 1988 was selected as the study's inception date because of the St. Louis Cardinals' move to Arizona after the 1987 season.
Table 1 Yearly Statistics for Entire League Spread Central Tendencies Home Record Year Mean Median Mode W - L - T % 1988 2.835 3 3 131 - 92 - 1 .587 1989 3.339 3 3 128 - 95 - 1 .574 1990 3.576 3 -3 131 - 93 - 0 .585 1991 3.348 3 3 132 - 92 - 0 .589 1992 2.978 3 3 136 - 88 - 0 .607 1993 2.781 3 -3 123 - 101 - 0 .549 1994 1.496 3 -3 128 - 96 - 0 .571 TOTAL 2.908 3 -3 909 - 657 - 2 .580
To determine an individual team's HFA, one needs to consider three factors: the mean points-margin for each team, a measure of team quality and the overall league mean points-margin. To determine the team quality factor the mean points-margin from teams with the same number of regular season victories (i.e. all the teams that had 1, 2, 3 ... season wins) are used as a measure of effect. Fourteen mean differences are thus obtained during these seven seasons as the number of season wins ranged from 1-14. Afterwards, a regression slope is calculated among these fourteen aggregated season victories.
The home field advantage is then found by determining the mean points-margin for each team (i.e. the average of all 56 home games, 7 seasons * 8 games), finding the average number of season victories for the team (regular season wins 17 seasons), and plotting the values referenced against the regression slope. A team's HFA is obtained [TABULAR DATA FOR TABLE 2 OMITTED] by adding or subtracting its distance from the slope to the league mean points-margin (i.e. the average of all 1,568 home games). This league or generic average serves as the base point for all individual teams. If the team mean points-margin falls above the regression slope it outperforms the expectations based on team quality, and its distance above the slope is added to the league mean. If a team falls below the slope it underperforms its team quality expectations and its distance from the slope is subtracted from the league mean. Thus the team farthest above the slope has the highest HFA, while the one furthermost below has the lowest.
One drawback in these calculations is that it must be assumed that all schedule strengths are equal over the study period. This may be an accurate assumption in ascertaining the individual team's mean points-margin, since they all have an equal number of games and a large N (N=56). However, the number of games played varies enormously in the determination of the team quality factor (e.g. 1 and 2 season wins occurred three times, 3*8=N=24; versus 9 season wins which occurred 27 times, 27*8=N=216). (see Table 3)
HFA differences between domed and open stadiums were calculated by multiplying the team HFA by the number of years played in domed or open stadiums, adding the results of each category and dividing by the total number of years played in each stadium type for the entire league. A difference of means t-test was performed to resolve whether the differences were significant. Home field advantage is defined as the number of points a team scores in excess of the visiting team after factoring out the biases of team quality. The terms "points-margin" and "spread" are used interchangeably in this paper. These terms refer to the difference between the home team's points or score, and the visiting team's points (e.g. Home team 24 Visiting team 35 = -11 spread or points-margin).
During the six years of the study period the home team outscored the opponent by an average of 291 points (league mean points-margin) and a median of three points (see Table 1). Thus, three points can be identified as the generic advantage posed by a home field in professional football. The spread for individual games, however, varied enormously (+54 to -51) [ILLUSTRATION FOR FIGURE 1 OMITTED]. Edwards (1979) in examining the results of 349 NFL games from 1974-76 also discovered that "home teams in the professional ranks had about a 3-point advantage" (p. 411). Given the scoring system in football, based on three-point field goals, six-point touchdowns, one-point conversions, and more recently the two-point conversion, certain point-spreads become more prevalent, such as 3,4,6,7,10 ... The results of overtime are also apparent, as the extra period minimizes deadlocked contests. During the seven seasons 78 games went into overtime with only two games remaining at a stalemate. Ironically, in those 78 games the home teams lost slightly more than they won (35-41-2). It is clearly evident in Figure 1 that the most common spread in football is three points, plus or minus, with the home team losing slightly more by that margin than winning. The effects of HFA become more evident when viewing the higher positive spreads and their increased frequencies for the home team.
Table 3 Team Quality Statistics 1988 - 1994 Season Mean Mean Mean Wins PF PA Spread W - L - T % 1 11.38 25.63 -14.25 0 - 24 - 0 .000 2 11.71 18.04 -6.33 4 - 20 - 0 .167 3 17.13 24.68 -7.56 19 - 53 - 0 .264 4 15.48 19.75 -4.27 30 - 66 - 0 .313 5 19.06 22.13 -3.07 52 - 84 - 0 .382 6 19.69 20.54 -0.86 74 - 86 - 0 .463 7 19.55 19.78 -0.23 77 - 75 - 0 .507 8 21.72 17.13 4.59 92 - 59 - 1 .609 9 22.16 16.73 5.43 143 - 72 - 1 .664 10 24.42 17.97 6.44 135 - 57 - 0 .703 11 23.36 14.47 8.89 125 - 35 - 0 .781 12 23.28 13.18 10.10 96 - 16 - 0 .857 13 28.23 16.15 12.08 36 - 4 - 0 .900 14 27.44 13.28 14.16 26 - 6 - 0 .813 Note. Mean PF = Mean Points For Teams with # Season Win(s), Mean PA = Mean Points Against, Mean Spread = average points-margin for 7 years, W, L, T = wins, loses and ties for home teams with # season win(s). Table 4 Playoff Statistics 1988 - 1994 Mean Mean Mean Mean H.W. V.W. PF PA Spread W - L % 14 11.00 27.13 12.63 14.50 6 - 2 .750 13 11.13 36.13 14.25 21.88 8 - 0 1.000 12 10.37 23.00 16.89 6.11 14 - 5 .737 11 9.78 19.83 18.06 1.78 12 - 6 .667 10 9.67 25.33 18.44 6.89 5 - 4 .556 9 9.00 28.50 20.50 8.00 3 - 1 .750 Note. H.W. = Home team season wins, Mean V.W. = visiting team season wins Mean PF = Mean Points For the home team, Mean PA = Mean Points Against the home team, Mean Spread = average Points-Margin W & L = wins and loses for home teams
The most striking finding was a nearly perfect positive linear relationship (r= .986) among the season victories' (team quality) mean points-margins [ILLUSTRATION FOR FIGURE 2 OMITTED]. The team quality slope indicates that the mean advantage of a team increases approximately two-points per season win, and becomes positive at seven victories. Consequently, a home team who wins four games in a season should be outscored by the visiting team by an average of five points, while a home team who wins nine games in a season should outscore their opponents by about five points.
A seemingly cyclical pattern is evident for the yearly league HFA (see Table 1). In 1988 the league HFA rose from 2.835 points to 3.576 points in 1990. Since 1990 it fell consistently, but gradually through 1993, and then precipitously in 1994 to its low at 1.496 points. Even though the league's yearly HFA has fallen profoundly since 1990 there has been no corresponding decrease in home team victories (131 vs. 128). In fact, home teams won just as many games in 1994 as they did in 1989 when the league HFA was more than twice as large.
Individually, HFA is much more prominent for certain teams/locations than for others (see Table 2). Houston manifests the largest HFA of any team, while Cincinnati, Denver, Kansas City and Washington also exhibit profound advantages. To the contrary, Indianapolis has essentially no home advantage, while the Los Angeles Raiders, Miami, Tampa Bay and Dallas display minimal advantages.
One must keep in mind though that the team HFA results factor out bias for team quality (i.e. strength). Teams such as San Francisco (Home Record 43-13, HFA 2.79) and the New York Giants (38-18, 1.98), who were among the best winning home teams during the study period, have such high expectations based on their mean season-victories that their HFA comes in lower in relation to the team quality slope expectations. While New England (23-33, 3.39), with one of the worst home records, maintains a higher than expected HFA. The effects of team quality have been examined in other home advantage studies as well. Schwartz & Barsky (1977), in studying baseball and hockey games, created a dichotomy of team strengths ("superior/first" and "inferior/second"), and then looked at the outcomes between the various matchups (i.e. 1-1, 1-2, 2-1, 2-2) (p. 646).
In comparing the two conferences, the American Conference (HFA 3.37) performs much better at home than the National Conference (HFA 2.75). In fact, the four top HFA teams are AFC teams. Also interesting are resemblances between the two New York teams and the two Los Angeles teams. The difference between the Jets and Giants is only .33 points, whereas the difference between the Raiders and Rams is .89 points. This may indicate that HFA is predominantly influenced by the local community and not greatly altered by erratic attributes of an individual team. This would dispel the notion that the Jets' often hapless seasons are the result of a "homeless feeling" from playing in Giants Stadium (Vecsey, 1993, p. B9).
Calculating the effects of HFA in the playoffs is much more complicated due to small N's, schedule strength disparities, and a phenomena known as "home field disadvantage." Table 4 displays how the 66 playoff games (not including Super Bowl games, which theoretically have no home team) match up when comparing the team quality of the home team versus the team quality of the visiting team. Notice that there is no clear cut relationship for team quality solely. This may be due to the small N's, which range between four and nineteen, or more probably indicates that the HFA of the individual teams must also be factored in. Remember that every season victory adds approximately two points to the HFA, but individual variations among these playoff teams range from .96 points HFA for the L.A. Raiders to 8.58 points for Houston. Thus, the individual variations may offset the advantages based only on team quality. Home field disadvantage is an added complexity. While contrary to all conventional wisdom the phenomena does seem to manifest itself more often in playoff and championship events, and has been studied extensively (Benjafield et al., 1989; Gayton et al., 1987; Heaton & Sigall, 1989). This episode is more commonly referred to as a "choke." In professional football the recurring Super Bowl loses of Buffalo and Denver bring this term to mind, even though the site is supposed to be neutral. Also, within the study period the home team was 9-5 (.642) in the conference championships, a much lower winning percentage than the remaining playoff games (39-13,. 750).
There is some evidence that "domed" stadiums amplify the effects of HFA (Horn, 1988), and this study's results may help confirm this. Domes were utilized for six teams throughout the study period (Detroit, Houston, Indianapolis, Minnesota, New Orleans, Seattle), with Atlanta initiating play in their domed stadium in 1993. In comparing teams playing in domed stadiums versus open stadiums, the domed teams averaged a HFA of 3.22 points, while the open stadiums averaged 3.01 points. These differences were not significant, but are still notable. It is generally thought that the heightened noise level is responsible for this phenomena.
This study has shown that home field advantage is a tangible force in professional football. All teams, except Indianapolis, possess a home field advantage of varying degrees of importance. Indianapolis's home field was neutral.
Taking for granted the generally beneficial affects of a home field does not mean the outcome is assured. "Psychologically, what is supposed to be the home-court advantage sometimes seems to create a home-court disadvantage by providing a team with a false sense of security (Anderson, 1993, Section 8, p. 1)."
Contradictory evidence was found regarding the principal determinants of HFA. This study found some evidence that the overall effects of team quality weigh more heavily in constructing the advantage. A team that wins two additional games per season (i.e. 4 points) will surpass the average league-advantage (i.e. 2.91 points) posed by a home field. However, the similarities of HFA in duplicate markets (e.g. Los Angeles and New York) suggest spatial conditions may predominate in constructing the advantage. Therefore, this study cannot say for certain which force is most integral in determining home advantage. Schwartz & Barsky (1977) were also unable to resolve this question stating, "game location and team quality are equally important determinants of performance" (p. 649). Nevertheless, team quality is a crucial factor in the advantage, which prompted Anderson (1993) to comment, "... the home court is like home cooking: It depends on who's cooking" (Section 8, p. 1).
These results are the reality of the past seven seasons. It should not be taken for granted that these figures will remain stationary over time. More than likely, home field advantage will change as the dynamics of the cities and the teams change affecting home attendance, crowd participation, and team quality. Nonetheless, the results of HFA cannot be predicted in that these variables are to some extent linked. As Edwards (1979) put it, "crowds are more dense when the home team is having a winning season, so it is difficult to separate the effects of fan support from team talent" (p. 433). One thing is for certain though, as George Seifert coach of the 1994 Superbowl champion San Francisco 49ers put it, "If you can get the home-field advantage, by God you want it ("Complaints Pour," 1995)."
Anderson, D. (1993, April 25). Home Court Can't Always be Trusted. New York Times, Section 8, p.1.
Amore, D. (1995, January 16). Davis Turns Tables on Cowboys. The Hartfort Courant, p. c4.
Askins, R. L. (1978) The Official Reacting to Pressure. Referee, 3, 17-20.
Benjafield, J., Liddell, W. W., & Benjafield, I. (1989) Is there a Home Field Disadvantage in Professional Sports Championships? Journal of Social Behavior and Personality, 17, 45-50.
Complaints Pour in Regarding Turf at Candlestick. (1995, January 16). Sacremento Bee, p. D5.
Courneya, K. S., & Carron, A. V. (1990) Batting First Versus Last: Implications for the Home Advantage. Journal of Sport and Exercise Psychology, 12, 312-316.
Courneya, K. S., & Carron, A. V. (1991) Effects of Travel and Length of Home Stand/Road Trip on the Home Advantage. Journal of Sport and Exercise Psychology, 13, 42-49.
Courneya, K. S., & Carron, A. V. (1992) The Home Advantage in Sport Competitions: A Literature Review. Journal of Sport and Exercise Psychology, 14, 13-27.
Edwards, J. (1979) The Home-Field Advantage. In J.H. Goldstein (Ed.) Sports, Games, and Play: Social and Psychological Viewpoints (pp. 409-438). Hillsdale, N J: Erlbaum.
Gallup, G. Jr., & Newport, F. (1992) Football Remains America's Number One Spectator Sport. The Gallup Poll Monthly, 325, 36-40.
Gayton, W. F., Mutrie, S. A. & Hearns, J. F. (1987) Home Advantage: Does It Exist in Women's Sports. Perceptual and Motor Skills, 65, 653-654.
Gayton, W. F., Matthews, G. R., & Nickless, C. J. (1987) The Home Disadvantage in Sport Championships: Does it Exist in Hockey? Journal of Sport Psychology, 9, 183-185.
Heaton, A. W., & Sigall, H. (1989) The Championship Choke Revisited: The Role of Fear of Acquiring a Negative Identity. Journal of Applied Social Psychology, 19, 1019-1033.
Horn, J. C. (1988) Dome-inating the Game. Psychology Today, 22: No. 10, 10.
Lehman, D. R. & Reifman, A. (1987) Spectator Influence on Basketball Officiating. Journal of Social Psychology, 127, 673-675.
McGuire, E. J., Courneya, K. S., Widmeyer, W. N, & Carron, A. V. (1992) Aggression as a Potential Mediator of the Home Advantage in Professional Ice Hockey. Journal of Sport and Exercise Psychology, 14, 148-158.
Rabun, M. (1986, October 6) Pro Files. UPI Sports Writer.
Schwartz, B., & Barsky, S. F. (1977) The Home Advantage. Social Forces, 55, 641-661.
Shirley, B. (1986, June 17) The Sports Fan: For many, the playing of games becomes something else entirely. Los Angeles Times, Part 3, p. 1.
Snyder, E. E., & Purdy, D. A. (1985) The Home Advantage in Collegiate Basketball. Sociology of Sport Journal, 2, 352-356.
Vecsey, G. (1993, October 27). Jets Need Home-Field Advantage. New York Times, p. B9.
For further information, please contact:
Jon C. Acker Department of Educational Research University of Alabama Tuscaloosa, AL 35486…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Location Variations in Professional Football. Contributors: Acker, Jon C. - Author. Journal title: Journal of Sport Behavior. Volume: 20. Issue: 3 Publication date: September 1997. Page number: 247+. © 1999 University of South Alabama. COPYRIGHT 1997 Gale Group.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.