A temporary decline in home advantage when moving to a new stadium.

Wilkinson, Todd ; Pollard, Richard

The home advantage in sport has been well documented in previous research, with a comprehensive review provided by Nevill and Holder (1999). One factor identified as contributing to the home advantage in sport is familiarity with the local playing facility (Pollard, 2002). This study compared home advantage during a team's last year in an old stadium to their first year in a new stadium. The records of thirty-seven North American professional baseball, basketball, and hockey teams that moved to new stadiums within the same metropolitan area between 1987 and 2001 were analyzed. The results showed a significant reduction in home advantage for teams performing in their first year in a new facility. Taking into account possible confounding factors such as crowd size and crowd density, Pollard quantified the effect of familiarity with the local playing facility as accounting for an estimated 24% of the home advantage.

Several previous studies have examined the effects on performance of moving into a new stadium. Quinn, Bursik, Borick and Raethz (2003) showed that Major League Baseball (MLB) teams significantly increased their overall winning percentage when moving to a new stadium, but teams from professional basketball (NBA), hockey (NHL) and football (NFL) did not evidence such an effect. Watson and Krantz (2003) documented a significantly increased home winning percentage for MLB teams moving to new stadiums, but not for teams in the NBA and NFL.

Loughead, Carton, Bray and Kim (2003) investigated the effect of moving to a new stadium by comparing home winning percentage in three time periods; immediately before the move, immediately after move, and following a period of relocation. Using a sample of 57 professional teams from hockey and basketball in North America and from soccer in England and Scotland, the authors found no significant differences in home winning percentage across the time periods. However, this analysis was based on a relatively small sample of games in each time period (8-10), and thus the lack of significant findings may have been affected by the different opponents for each team in the three time periods.

The present study seeks to extend previous findings on the home advantage. It should be noted that the method of analysis employed differs from the three aforementioned studies in its calculation of the home advantage. Rather than using purely a measure of home winning percentage, the analysis of home advantage used in this study is based on the percentage of a team's number of home wins to its total wins for a season. It should be emphasized that home winning percentage alone is not a valid measure of home advantage unless assessed simultaneously against away winning percentage. For example, a team that wins 60% of its home games would actually have a home disadvantage if it wins 70% of its away games. Also, the use of complete seasons of results, rather than just a small number of games, ensures that the comparisons made are based on balanced playing schedules, with the same teams being faced at home and away.

We hypothesize that since a team's familiarity with their playing facility will have increased by their second year in the new venue, the magnitude of their home advantage should also be greater in the second year than in the first. In other words, we anticipate that the decline in home advantage due to loss of familiarity will be temporary and last no longer than one season. Specifically, the null hypothesis was that there would be no significant difference in home advantage between teams' last year in the old stadium, first year in a new stadium, and second year in a new stadium. This was tested against the one-sided alternative that home advantage in the first year in the new stadium would be lower than in both the final year in the old stadium and the second year in the new stadium.

Method

Participants

The participant sample included 40 teams from Major League Baseball (MLB), the National Hockey League (NHL), and the National Basketball Association (NBA) that had moved to a new stadium during the period from October 1987 to September 2002. Teams from the National Football League (NFL) were not included due to that league's relatively smaller number of games played per year and therefore the inherent lack of reliability in calculating the home advantage for these teams. Further, only teams that moved to a new stadium within the same metropolitan area were included. This criterion was chosen to prevent the introduction of potential confounding variables associated with a team's move to a new city. The year 1987 was selected as the start date as no stadium changes were made in the three years previous to 1987, and at least one team has moved into a new stadium in every year since.

Several additional teams were eliminated from the analysis due to other methodological considerations. For instance, six teams moved mid-season, which prevented a comparison of balanced schedules (Seattle in MLB, Montreal, Ottawa and Toronto in the NHL, and Miami and Washington in the NBA). Additionally, three MLB teams (Cleveland, Texas, and Colorado) had seasons during their move affected by the baseball strike, and thus did not have a balanced schedule for comparison. Another team (Houston in MLB) was excluded due to their move from a domed stadium to an open-air stadium. Previous research has shown that the home advantage for domed and open-air baseball stadiums is not the same (Zeller & Jurkovac, 1988). Finally, an NBA team (Atlanta) was excluded from the analysis because their move spanned two seasons, during which they played home games in two different locations.

Materials

Major League Baseball, National Hockey League and National Basketball Association teams that had moved to a new stadium within the same city between 1987 and 2002 were identified through use of an internet website (ballparks.com). Team records from this same period of time were also identified through use of an internet website (shrpsports.com).

Procedure

The method for calculating home advantage was that of Pollard (1986). For each baseball and basketball team, home advantage was calculated as the percentage of home wins to total wins during a season. To account for ties and overtime losses in hockey, home advantage for these teams was calculated as the percentage of total points earned in home games divided by total points earned during the season. For each team, home advantage was calculated for three seasons: their last year in the old stadium, and for each of their first two years in the new stadium.

Statistical analysis

Because there was evidence of non-normality of the distribution of home advantage values, non-parametric methods were employed in the analysis. Friedman's test was used to investigate differences between the three time periods. A subsequent test of multiple comparisons (Hollander & Wolfe, 1973) was then applied in order to determine the extent to which home advantage in the first year in the new stadium was lower than in the final year in the old stadium, and in the second year in the new stadium.

Results

Table 1 presents mean home advantage for the three time periods, with the teams classified by sport. These results are shown in Figure 1 and clearly display the dip in home advantage that occurred in all the sports during the first year in the new stadium. Friedman's test, using all 40 teams, indicated a significant difference between the three time periods (p = .015). Table 2 gives p-values for the tests of multiple comparisons. A clear drop in home advantage in the first year in the new stadium is evident (p = .005), with the increase in the subsequent year also significant (p = .036). These results suggest that a significant drop in home advantage occurs during the initial season in a new stadium, but that by the following season home advantage has largely returned to previous levels.

[FIGURE 1 OMITTED]

Discussion

The results of the study provide support for the effect of familiarity with playing facility on the home advantage in sport. Home advantage was greater in teams' second season in their new stadium than their first season in the new stadium. In addition, there was little difference between the home advantage for teams in their second season in the new stadium versus their last season in the old stadium, suggesting that a two-year period is on average, sufficient for teams to gain familiarity with their new stadium and subsequently "rebound" to the level of home advantage they previously enjoyed in the old stadium. The results also replicate the findings of Pollard (2002) by evidencing a significant decrease in home advantage from a team's last year in their old stadium to their fast year in the new stadium.

In addition to our analysis of the home advantage, we examined whether three potential confounding factors may have had an impact on our findings: attendance, team quality, and changes over time in the overall league home advantage. First, the average attendance per home game in each of the three seasons studied was investigated. In each of the three leagues, average attendance per game was highest the first year in a new stadium. Since larger crowds would, if anything, be expected to increase home advantage, it can be concluded that change in attendance could not have been responsible for the decline in home advantage.

Several previous studies have explored the possibility that the quality of individual teams might be related to the magnitude of home advantage, but no clear cut conclusions have been reached. These include basketball (Harville & Smith, 1994; Madrigal & James, 1999), hockey (Bray, 1999) and these sports together with soccer (Loughead et al., 2003). We also examined the possibility that the changes in home advantage may have been affected by changes in team quality, which was quantified by team winning percentage. No significant relationship was found between change in team quality from season to season and change in home advantage. As a result, this factor can also be excluded as having an influence on the observed change in home advantage.

The third confounding factor analyzed was the overall change in home advantage over time. It has been documented that over the past two decades, the average home advantage has been declining in basketball and hockey, at the same time showing marked variations from year to year. Pollard and Pollard (2005) present a season by season analysis of home advantage for all the sports in this study. Because our analysis covers a three year period for each team, it is possible that the change in overall home advantage over time might have had an influence on the results. To account for this factor, home advantage was analyzed for each team in relation to the average home advantage in that sport for that year. This refinement made a slight but interesting change to the results (Table 3) with Friedman's test still showing a significant difference between the time periods (19 = .045). The mean drop in home advantage and subsequent recovery became very similar to each other in magnitude, both just under 2%. However, although the mean difference between the first and second season after relocation had increased, its significance was lowered (p =. 103). Nevertheless, the level of home advantage in the second year in the new stadium was now virtually the same as in the final year in the old.

The study is not without its limitations. For instance, teams from just three professional leagues in North America were included in the analysis. As such, it is possible that these results do not generalize to other sports (e.g., professional football) or to leagues in other countries. Many of the top soccer teams in Europe have, for instance, moved to new stadiums during the last decade. Therefore, it is suggested that further research be conducted to examine this rebound effect across other domains and contexts.

In addition, by examining team records on a complete season by season basis, we were unable to determine more precisely when increased familiarity with the facility resulted in a team's return to the previous home advantage enjoyed in the old stadium. It is possible, for instance, that familiarity had increased enough by mid-season for teams to have regained their full home advantage. Additional research using split- or partial-season records may serve to further enhance this understanding, although the lack of a balanced playing schedule during a specific part of a season will present problems.

With the proliferation of new stadiums having been built over the past decade, and with more new venues planned, the findings of the present study may be both valuable and timely for a number of professional organizations. These results suggest it may be in the interest of a team moving to a new stadium to increase its exposure to the new facility, possibly through additional scrimmages, pre-season work-outs or exhibition games, in an effort to more quickly familiarize them with the new facility. It may also serve to inform team organizations and interested parties as to the importance of familiarity with the playing facility as a factor in team performance.

References

Bray, S. R. (1999). The home advantage from an individual team perspective. Journal of Applied Sport Psychology, 11, 116-125.

Harville, D. A., & Smith, M. H. (1994). The home-court advantage: How large is it, and does it vary from team to team? American Statistician, 48, 22-28.

Hollander, M. & Wolfe, D.A. (1973). Nonparametric statistical methods. New York: Wiley.

Loughead, T. M., Carron, A. V., Bray, S. R., & Kim, A. J. (2003). Facility familiarity and the home advantage in professional sports. International Journal of Sport and Exercise Psychology, 1, 264-274.

Madrigal, R., & James, J. (1999). Team quality and the home advantage. Journal of Sport Behavior, 22, 381-398.

Nevill, A. M., & Holder, R. L. (1999). Home advantage in sport: An overview of studies on the advantage of playing at home. Sports Medicine, 28, 221-236.

Pollard, R. (1986). Home advantage in soccer: A retrospective analysis. Journal of Sports Sciences, 4, 237-248.

Pollard, R. (2002). Evidence of a reduced home advantage when a team moves to a new stadium. Journal of Sports Sciences, 20, 969-973.

Pollard, R., & Pollard, G. (2005). Long-term trends in home advantage in professional team sports in North America and England (1876-2003). Journal of Sports Sciences, 23, 337-350.

Quinn, K. G, Bursik, P. B., Botlek, C. P., & Raethz, L. (2003). Do new digs mean more wins? The relationship between a new venue and a professional sports team's competitive success. Journal of Sports Economics, 4, 167-182.

Watson, J. C., & Krantz, A. J. (2003). Home field advantage: New stadium construction and team performance in professional sports. Perceptual and Motor Skills, 97, 794-796.

Zeller, R. & Jurkovac, T. (1988). Doming the stadium: The case for baseball. Sport Place International, 3, 35-38.

Todd Wilkinson

University of Minnesota

Richard Pollard

California Polytechnic State University

Address Correspondance To: Todd Wilkinson, Dept of Psychology, Elliott Hall, Room N580, University of Minnesota, Minneapolis, MN 55455.

Table 1.
Mean home advantage before and after moving into a new stadium.

 First Second
 Last Season Season Season
 Number in Old in New in New
Sport League of Teams Stadium Stadium Stadium

Baseball MLB 7 54.75% 53.72% 55.11%
Basketball NBA 19 64.08% 61.68% 63.25%
Hockey NHL 14 56.03% 53.31% 55.24%
All sports 40 59.63% 57.36% 59.02%

Table 2.
Mean changes in home advantage from season to season.

Seasons Mean change p-value

First season in new stadium -- last -2.27% .005
season in old stadium

First season in new stadium -- second -1.66% .036
season in new stadium

Table 3.
Mean changes in home advantage from season to season, adjusted for
average league home advantage for each year in each sport.

Seasons Mean change p-value

First season in new stadium -- last -1.86% .013
season in old stadium

First season in new stadium -- second -1.98% .103
season in new stadium