Full opinion text
MEMORANDUM OF DECISION STEFAN R. UNDERHILL, District Judge. In March 2009, the defendant, Quinnipiac University (“Quinnipiac” or the “University”), announced plans to cut three of its sports teams: the women’s volleyball team, the men’s golf team, and the men’s outdoor track team. Contemporaneously, the University pledged to create a new varsity sport, competitive cheerleading, for the 2009-10 season. Those decisions form the basis of this lawsuit. Plaintiffs Stephanie Biediger, Kayla Lawler, Erin Overdevest, Kristen Corinaldesi, and Logan Riker are five current Quinnipiac women’s varsity volleyball players, and plaintiff Robin Lamott Sparks is their coach. Together, they allege that Quinnipiac’s decision to eliminate its volleyball team violates Title IX of the Education Amendments of 1972 (20 U.S.C. § 162 et seq.) and the regulations adopted pursuant thereto (34 C.F.R. Part 106) (“Title IX”). On May 22, 2009, I granted the plaintiffs’ motion for a preliminary injunction, holding that the manner in which Quinnipiac managed its varsity rosters — essentially, by setting artificial ceilings for men’s varsity teams and floors for women’s varsity teams — deprived female athletes of equal athletic participation opportunities. See Biediger v. Quinnipiac Univ., 616 F.Supp.2d 277 (D.Conn.2009). On May 20, 2010, I certified the following class of plaintiffs seeking injunctive relief, pursuant to Rule 23(b)(2) of the Federal Rules of Civil Procedure: All present, prospective, and future female students at Quinnipiac University who are harmed by and want to end Quinnipiac University’s sex discrimination in: (1) the allocation of athletic participation opportunities; (2) the allocation of athletic financial assistance; and (3) the allocation of benefits provided to varsity athletes. Biediger v. Quinnipiac Univ., No. 09cv621 (SRU), 2010 WL 2017773 (D.Conn. May 20, 2010). Although the plaintiffs allege several theories for relief under Title IX, the parties agreed to sever and try independently the plaintiffs’ first claim: that Quinnipiac discriminates on the basis of sex in its allocation of athletic participation opportunities. The parties tried that claim in a bench trial held from June 21 to June 25, 2010. My findings of fact and conclusions of law are set forth herein. I conclude, as a matter of law, that Quinnipiac discriminated on the basis of sex during the 2009-10 academic year by failing to provide equal athletic participation opportunities for women. Specifically, I hold that the University’s competitive cheerleading team does not qualify as a varsity sport for the purposes of Title IX and, therefore, its members may not be counted as athletic participants under the statute. Competitive cheer may, some time in the future, qualify as a sport under Title IX; today, however, the activity is still too underdeveloped and disorganized to be treated as offering genuine varsity athletic participation opportunities for students. Second, I hold that, although cross-country, indoor track, and outdoor track are usually considered different sports, Quinnipiac may not count some runners who participate in each sport three times. Quinnipiac’s practice of requiring women cross-country runners to participate on the indoor and outdoor track teams, and its treatment of the indoor and outdoor track teams as, in essence, an adjunct of the cross-country team, are sufficient to show that some cross-country runners who participate on the indoor and outdoor track teams should not be counted under Title IX. Specifically, cross-country runners who were injured or red-shirted during the 2009-10 indoor and outdoor track seasons cannot be counted because their activity does not amount to genuine athletic participation opportunities. Finally, although I find, as a matter of fact, that Quinnipiac is no longer engaged in the same roster manipulation that was the basis for my preliminary injunction order, the University is still continuing to deflate the size of its men’s rosters and inflate the size of its women’s rosters. Although that roster management is insufficient to conclude that Quinnipiac violated Title IX as a matter of law, it supports the ultimate conclusion that the University is not offering equal participation opportunities for its female students. I. Findings of Fact Quinnipiac University is a private, coeducational institution located in Hamden, Connecticut. Quinnipiac is a member of the National Collegiate Athletic Association (“NCAA”), and is a Division I school, which means that it belongs to the NCAA’s most competitive athletic division. The University competes in the Northeastern Conference (“NEC”), to which ten other Division I programs belong. Historically, men’s and women’s basketball and ice hockey have been the school’s premier sports. Quinnipiac, however, sponsors seven varsity men’s athletic teams and 12 varsity women’s athletic teams, including, inter alia, women’s volleyball, women’s cross-country, women’s indoor track and field, women’s outdoor track and field, and women’s competitive cheer. During the 2009-10 academic year, 5,686 students were enrolled in the University’s undergraduate program. That year, 2,168 students, or 38.13 percent of the student body, were male, and 3,518 students, or 61.87 percent of the student body, were female. Based on the varsity rosters for the first day of teams’ competitions, 166 male athletes and 274 female athletes participated during the academic year. Using those first day of competition figures, in 2009-10 male athletes made up 37.73 percent, and female athletes 62.27 percent, of the University’s varsity athletes. If Quinnipiac’s numbers were accepted, the University would not be liable under Title IX because the school would be offering athletic opportunities for women in numbers proportional to the percentage of women in the school’s undergraduate population. To succeed, the plaintiffs must undermine the way Quinnipiac counts its varsity athletes. Plaintiffs seek to do so by raising three areas of factual disagreement. First, the plaintiffs claim that Quinnipiac’s roster numbers are inaccurate and the University’s policy of setting varsity roster targets leads to artificially undersized men’s teams and oversized women’s teams. Second, the plaintiffs claim that Quinnipiac’s female cross-country runners should not be counted multiple times for their participation on the indoor and outdoor track teams because the indoor and outdoor track teams are not true independent sports teams but serve, instead, as a more structured nontraditional season for the University’s female cross-country runners. Third and finally, the plaintiffs claim that the Quinnipiac competitive cheerleaders should not be counted because competitive cheer is not yet a legitimate intercollegiate varsity sport. I take up each of those factual questions independently. I then conclude my findings of fact by discussing Quinnipiac’s plans for the 2010-11 academic year. A. Roster management Following the 2008-09 academic year, Quinnipiac changed its practice with respect to setting roster targets for its teams — the practice that formed the basis of my preliminary injunction order. Rather than establish ceilings for men’s teams and floors for women’s teams, and implicitly encourage coaches to manipulate their rosters before and after the first dates of competition — the dates on which the University collected its roster data to prove its Title IX compliance — Quinnipiac established a new policy of setting roster targets. That policy required coaches to meet roster numbers prescribed by the University after consultation with its coaches. The roster target system was overseen by Dr. Mark Thompson, the senior vice president for student and academic affairs, and Jack McDonald, Quinnipiac’s athletic director. Thompson inaugurated the new roster management system in the summer of 2009. After reviewing the relevant Title IX and NCAA regulations, Thompson developed a set of preliminary roster targets based on Quinnipiac’s roster sizes in 2008-09, NCAA average squad sizes, and the teams’ operating and scholarship budgets. Thompson testified that the purpose of setting the roster targets was to ensure that all athletes were given genuine participation opportunities. Those preliminary numbers were revealed to Quinnipiac’s head coaches in a meeting held in July 2009, and were finalized on September 1, 2009, just before the beginning of the fall season. Issuing the initial roster target numbers in July 2009 meant that coaches were given their numbers after the recruiting season had ended and incoming Quinnipiac students had made their enrollment deposits. In other words, teams that had to find new players to meet their roster targets were likely forced to find those players on campus or among the small pool of graduating high school seniors and junior college players who had not already decided where to enroll for the 2009-10 academic year. At the July 2009 head coaches meeting, some of the coaches audibly expressed frustration with their preliminary roster targets. In between the issuance and finalization of the preliminary numbers, however, every coach met with Thompson to discuss whether the preliminary numbers were appropriate. For example, the women’s basketball coach requested increases in the team’s roster, Def. Exs. BZ & CA; the women’s ice hockey coach requested a reduction of the team’s roster, Def. Ex. CB; the women’s softball coach requested an increase of one player to the team’s roster; Def. Ex. CF; the men’s basketball coach requested an increase of two players to the team’s roster, Def. Ex. CK; and the men’s ice hockey coach requested an increase of two players to the team’s roster, Def. Ex. CN. Thompson agreed to all of those requests. Indeed, the only roster target change that Thompson appears to have denied was a requested increase of one women’s volleyball player by Robin Sparks, for whom, Thompson testified, Sparks never proved the need. Ultimately, Thompson received the input of each Quinnipiac coach before finalizing the teams’ roster target numbers. Every Quinnipiac coach, except for Germaine Fairchild, the women’s softball coach, and Sparks, signed letters confirming their acceptance and approval of their roster targets. See Def. Ex. BO. During the academic year, Quinnipiac’s coaches were permitted to make changes to their rosters. Any addition or deletion of a player, however, had to be approved in advance by Thompson. The plaintiffs allege two potential ways in which Quinnipiac’s practice of setting roster targets could result in a Title IX violation. First, the University may have manipulated its roster sizes by misidentifying student athletes, specifically, by erroneously discounting individual athletes from men’s teams and adding individual athletes to women’s teams. In other words, the plaintiffs claim that Quinnipiac’s varsity teams were not actually complying with the roster targets the University assigned, and the actual number of athletes was not captured by the roster target totals. And, second, the University may have set artificially low roster numbers for its men’s teams and high numbers for its women’s teams, thus creating de facto ceilings and floors for squads on the basis of sex, which had the effect of denying genuine athletic participation opportunities to some women athletes. 1. The total number of athletes competing at Quinnipiac The plaintiffs and Quinnipiac offered two competing theories for how the University’s 2009-10 athletes should be counted. Relying on a clarification letter issued in 1996 by the Department of Education’s Office of Civil Rights (“OCR”), Quinnipiac offered its squad lists for teams’ opening days of competition as evidence of the total number of athletes participating in the 2009-10 academic year. It did nothing more than put forward those opening day roster numbers for the purpose of demonstrating its compliance with Title IX. By contrast, the plaintiffs’ expert witness, Dr. Donna Lopiano, scrutinized Quinnipiac’s 2009-10 squad lists for which athletes were on teams’ rosters on the opening days of competition and/or were added to, or dropped from, rosters later in the season. Lopiano analyzed those squad lists to determine which students counted as athletic “participants” according to OCR’s definition of the term. She accounted for all potential participants according to the following eight criteria: (1) Athletes who competed during the academic year, regardless whether such competition took place during the sport’s traditional or nontraditional season; (2) athletes who did not compete because of injury but received athletic financial aid, such as partial or full scholarships; (3) athletes who were red-shirted during the 2009-10 year; (4) athletes who were listed on their team’s squad list for the first day of competition in the traditional season and practiced with their teams, even if they never competed in a game; (5) athletes who were on their team’s squad list for the first day of competition in the nontraditional season and practiced with the team, even if they were not on their team’s squad lists for the traditional season and did not compete; (6) athletes who did not appear on a squad list for the first day of competition in the traditional or nontraditional season, but who practiced with their team for a majority of the season and received at least one semester’s worth of athletic benefits; (7) athletes who, in Lopiano’s view, raised a “red flag” and either should not have been counted or, if not already included in the first six criteria, should have been added to the list of athletes; and (8) athletes who exhausted their four years of NCAA eligibility but who, in a fifth year, continued to practice with their team and receive the same athletic benefits as any other varsity athlete. PI. Ex. 149 at 2-3. In producing her figures for Quinnipiac’s total number of athletic participants, Lopiano made several additional categorical choices about which women should be included. Lopiano did not include the women’s competitive cheer team in her count of female athletes, on the assumption that women’s competitive cheer did not constitute a valid sport for Title IX purposes and, therefore, the team’s members were not “participants” to be tallied under the statute. In addition, Lopiano offered four alternative totals for female athletes participating on the cross-country and track teams. As I explain in more detail in the next section concerning Quinnipiac’s cross-country and track programs, there is substantial overlap between the women’s cross-country, indoor track, and outdoor track teams; indeed, every female cross-country runner was on the indoor and outdoor track teams’ 2009-10 rosters. Thus, those runners potentially may be counted as three distinct athletic participants under Title IX. In order to account for the various ways female runners can be counted, Lopiano made four calculations of the number of Quinnipiac women athletes: (1) the total number of participants on each women’s team, regardless whether runners are counted three times for participating on the cross-country, indoor track, and outdoor track teams; (2) the total number of participants on each women’s team, but counting participants competing on both the indoor and outdoor track teams only once for track and field; (3) the total number of participants on each women’s team, but counting cross-country runners and exclusive track runners once, regardless whether they participated on multiple teams; and (4) the total number of participants on all women’s teams, but subtracting the indoor and outdoor rosters completely. Id. at 4. Although each of those methods produces different totals for the number of female varsity athletes during the 2009-2010 year, in none of those calculations did Lopiano detect evidence of roster manipulation. Rather, she agreed that the athletes on the women’s cross-country, indoor track, and outdoor track squad lists were properly included as participants, insofar as a single athlete competing on all three teams participates in separate sports. Ultimately, Lopiano disagreed that individual runners who competed on more than one of those teams should be counted as more than one participant for the purpose of Quinnipiac’s Title IX compliance because, in her view, it was a ruse for Quinnipiac to increase the statistical representation of its female athletes. But because Lopiano’s disagreement centered on whether, categorically, female runners could be counted multiple times, and not on whether the runners actually participated on those teams, the distinctions between Lopiano’s various tabulations of the University’s female runners do not bear on Quinnipiac’s purported roster manipulations. Rather, Lopiano observed potential evidence of Quinnipiac’s inaccurate participant counting on other teams- — -namely, the men’s basketball, ice hockey, lacrosse, soccer, and cross-country teams, as well as the women’s field hockey, ice hockey, lacrosse, soccer, and softball teams. For all of those teams, Lopiano performed three calculations based on her eight criteria for defining athletic “participants,” as described above. First, in what she dubbed “Count A,” she added each team’s participants who met criteria (1) through (4). Next, in “Count B,” she took the totals from Count A and included any other participants who met criteria (5) through (8). Finally, in “Count C,” she subtracted any athletes whom she designated a “red flag” under criterion (7) and who, as far as she could tell, did not actually participate on their teams for Title IX purposes. She then compared those counts to Quinnipiac’s opening day squad lists, the University’s chosen method for establishing its numbers of male and female athletes. In relevant part, Lopiano calculated the following numbers of athletic participants on Quinnipiac’s men’s teams: COUNT COUNT COUNT QU MEN’S TEAM ABC COUNT Basketball 17 18 18 17 Ice Hockey 31 31 31 30 Lacrosse 43 43 43 41 Soccer 24 27 27 24 X-Country 13 16 16 13 Total 170 176 176 166 PI. Ex. 151. When one takes all of the potential participants included in Count B, none of which are excluded in Count C, and compares them to Quinnipiac’s count, the difference is ten male athletes. Thus, according to the plaintiffs, Quinnipiac undercounts its male athletic participants by up to ten students by relying on squad lists for the opening days of competition. For the women’s teams, on the other hand, the plaintiffs contend that Quinnipiac overstates the number of its female athletic participants. Lopiano calculated the following numbers of athletic participants on Quinnipiac’s women’s teams: WOMEN’S COUNT COUNT COUNT QU TEAM COUNT Field Hockey 26 25 24 Ice Hockey 26 26 20 26 Lacrosse 28 30 30 29 Soccer 29 29 28 27 Softball 21 21 21 20 Total 170 172 164 166 Id. According to the plaintiffs, then, Quinnipiac overstates its female participants (excluding the cross-country, outdoor track, and indoor track runners) by two. During trial, however, Quinnipiac demonstrated that the University engaged in minimal undercounting of male athletes and no overcounting of female athletes, and that the discrepancies between the plaintiffs’ and defendant’s calculations were attributable to facts about individual athletes of which Lopiano was unaware. Take, for instance, the six women’s ice hockey players whom Lopiano discounted and who are responsible for the net overstatement of Quinnipiac’s female athletic participants. Lopiano identified those six players as “red flags” and subtracted them from her “Count C” because she believed they were not legitimate players and were cut after their season ended due to a Quinnipiac policy prohibiting women’s coaches from cutting players mid-season in order to maintain the appearance of Title IX compliance. See PL Ex. 147 (email to Jack McDonald explaining why six cuts were made after end of women’s ice hockey season). Rick Seeley, the Quinnipiac women’s ice hockey coach, gave uncontradicted testimony that Lopiano’s belief was false. Seeley was never informed by Quinnipiac’s administration that he could not cut players during the season; at most, he was instructed to contact Mark Thompson before cutting any of his players. Furthermore, the 2009-10 women’s ice hockey team roster shows that five of the six players Lopiano identified as “red flags” competed in games during the season, Def. Ex. EW, at D 0859, and Seeley confirmed that the remaining “red flag” was the last player on the bench and the only team member, other than a red-shirted player, not to play in a game that season. Nonetheless, Seeley gave undisputed testimony that even the last player on the bench, like all of the women ice hockey players Lopiano identified as “red flags,” was on the team’s squad list on the first day of competition, practiced with the team, and received athletic benefits like any other varsity athlete. Moreover, Seeley said that all of the six players at issue quit on their own accord, or were cut for disciplinary and/or other legitimate, team-based reasons. Quinnipiac’s and Lopiano’s counts were further reconciled by the testimony of Tracey Flynn, the University’s Title IX compliance officer. The one-player difference between Quinnipiac’s and Lopiano’s counts for the men’s basketball team was explained by a student who transferred to the University mid-year; under NCAA regulations, he was not allowed to join the team immediately and, therefore, was not counted by Quinnipiac on the basketball team’s roster. The one-player difference between Quinnipiac’s and Lopiano’s counts for the men’s ice hockey team stems from a player who was cut before the first day of competition, but rejoined the team in January 2010; although, according to the team’s season statistics, that player did not play in any games, he did practice with the team and arguably should have been included in Quinnipiac’s tally. Quinnipiac’s counting should therefore be increased by one male athlete. The men’s lacrosse team’s two-player difference appears to represent a genuine disagreement between the plaintiffs and defendant about how to count the athletes. Records of Quinnipiac’s internal change-in-status log show that two additional players joined the team in January 2010, before the first day of official competition; those same records also indicate that two different lacrosse players quit at approximately the same time, before the first game of the traditional season. Def. Ex. EN. As email correspondence shows, the two players added to the men’s lacrosse team were substitutes for the two players cut. Def. Exs. CR-CS, CU-CV. Quinnipiac’s opening date of competition data is therefore accurate insofar as it includes the correct number of players on the men’s lacrosse roster. Unlike Lopiano, however, Quinnipiac does not account for the players swapped; instead, the University counts only the roster spots filled on, and not before, the first day of competition. A similar player swap explains the three-player difference between Quinnipiac’s and Lopiano’s counts of the men’s soccer and cross-country squads. Three players were added to the men’s soccer team roster in spring 2010, after the traditional season had closed, which is why they appear in Lopiano’s Counts B and C. Lopiano did not note, however, that three players left the team between January and March 2010, and it appears that the three who were added were supposed to fill the places of those who left during the nontraditional season. The same exchange appears to have occurred on the men’s cross-country team: three players were added to the team’s spring, nontraditional roster to make up for three outgoing seniors. Because all of those player swaps happened before the commencement, or after the conclusion, of the teams’ championship seasons, I will accept Quinnipiac’s count of its male athletes, with the exception of the one men’s hockey player who must be added to the University’s tally. In sum, the facts show that Quinnipiac’s count of female athletes — without considering the participants on the University’s women’s cross-country, indoor track, and outdoor track squads, or the members of the competitive cheer team — is not overstated. Rather, the evidence shows that the women’s ice hockey team, the only team that the plaintiffs put forward as artificially keeping more players on its team than was necessary or desirable, provided participation opportunities to all of the players on its roster. The women’s ice hockey players who left the team did so after the season and for legitimate reasons; indeed, the plaintiffs put on no evidence that any women ice hockey players were forced off because of the University’s roster management. The facts suggest that the men’s teams were arguably slightly undercounted. But eight of the ten identified players who were purportedly undercounted were athletes who joined teams to replace players who were cut or quit. Thus, there is no evidence, other than the one player added to the men’s ice hockey team mid-season, that Quinnipiac men’s coaches felt pressure to add more players to their teams following the first date of official competition. Rather, the evidence shows that the men’s coaches, like the women’s coaches, followed the direction of the University’s athletics department and maintained their rosters at the prescribed sizes. That conclusion was confirmed by Flynn, who testified that the suspicious trend of men’s teams adding players, and women’s teams dropping players, after the first date of competition — the trend that she had observed and testified about at the preliminary injunction hearing — did not take place diming the 2009-10 academic year. Indeed, when pressed by opposing counsel on cross-examination, Lopiano could not identify any examples of the University adding or cutting players after the first date of competition in order to manipulate the school’s roster numbers. 2. The sizes of Quinnipiac’s men’s and women’s teams relative to national and conference averages Even if Quinnipiac’s men’s and women’s teams complied with the University’s revised roster management program, and ceased their former respective practices of adding and cutting players after the first date of official competition, the question still remains whether Quinnipiac’s mandatory roster numbers are appropriately set and afford athletes genuine varsity participation opportunities. At trial, the plaintiffs introduced data showing that the Quinnipiac men’s rosters were slightly smaller than national and conference averages, while women’s rosters were slightly larger than those averages. The plaintiffs introduced the following chart representing the difference between the size of Quinnipiac’s men’s teams and the size of average NCAA Division I and NEC teams: MEN’S TEAM QU COUNT NCAA AVG. NEC AVG. Baseball 31 33.3 35.88 Basketball 17 15.4 14.09 lee Hockey 30 27.7 28 Lacrosse 41 45.6 51.4 Soccer 24 28.5 25.6 Tennis 10 10.2 11.22 X-Counta'y 13 15.4 16.45 PI. Ex. 151. Those data reveal that four of the seven men’s teams — the baseball, lacrosse, soccer, and cross-country teams — had rosters that were smaller by approximately two to four players than the national average, and that five men’s teams were smaller than the average Northeastern Conference squad. By contrast, only the men’s basketball and ice hockey teams — the University’s premier programs — had teams larger than the national and conference averages. The plaintiffs also introduced evidence showing that Quinnipiac’s women’s teams were larger than the national and conference averages. They introduced the following chart showing the disparity between women’s teams and the national and conference averages: WOMEN’S QU TEAM COUNT NCCA AVG. NEC AVG. Basketball 18 14.5 13.82 Field Hockey 24 22.7 22.2 Ice Hockey 26 23.3 27.5 Lacrosse 29 27.2 23.44 Soccer 27 26.2 24.8 Softball 20 19.6 19.2 Tennis 10 9.4 8.8 Volleyball 12 14.8 12.56 X-Country 18 16.9 13 Indoor Track 30 37.9 29.91 Outdoor Track 30 37.2 30.55 Id. Those figures show that, with the exception of women’s volleyball, indoor track, and outdoor track, every Quinnipiac varsity women’s team was larger than the national average Division I women’s team. Quinnipiac’s teams were larger than the national average by approximately one to four players. The chart also shows that the University’s indoor and outdoor track squads were comparatively much smaller than the typical women’s track and field team; both had seven to eight fewer players than the average Division I team. The Quinnipiac track rosters were consistent in size with the NEC averages for those sports, however. But other Quinnipiac women’s teams, such as basketball, cross-country, and lacrosse, were bigger than the average NEC squad by four to five players, a larger disparity than between those squads and average NCAA Division I rosters. Indeed, except for the track teams, volleyball, and ice hockey, every University women’s team was bigger than the average conference team. Quinnipiac challenged the evidence that its men’s teams were smaller than average, and its women’s teams larger than average, by introducing data of the variation between the size of NEC teams. For example, NEC women’s indoor and outdoor track teams ranged from 18 to 56 athletes, and the conference’s roster numbers reveal that the Quinnipiac women’s indoor and outdoor track teams were not only consistent with the mean size of their rivals, but that the University’s teams were near the median of the conference’s indoor and outdoor track team sizes. See PL Ex. 103 (setting forth 2008-09 roster sizes of NEC teams, and showing that, of the twelve teams in the conference, Quinnipiac had the sixth-largest women’s indoor track team and the fifth-largest women’s outdoor track team). But other Quinnipiac women’s teams are demonstrably bigger than their conference competition when the 2009-10 Quinnipiac rosters are compared to the conference’s 2008-09 rosters, as set forth in Plaintiffs’ Exhibit 103. The University’s 2009-10 women’s basketball team had 18 members; that was three more members than the next biggest NEC team in the preceding season. Id. Similarly, the Quinnipiac women’s cross-country, lacrosse, and soccer teams would all have been the third-largest NEC team for their respective sports. Id. That stands in contrast to the men’s teams, which, with the exception of the cross-country and basketball teams, all had relatively small rosters when the University’s 2009-10 roster numbers are compared to the NEC’s 2008-09 tallies. That comparison shows that, of ten competing schools, the men’s tennis team would have been smaller than five other NEC teams; that, of eleven competing schools, the men’s soccer team would have been smaller than six other NEC teams; that, of nine competing schools, the men’s baseball team would have been the second-smallest team in the NEC; and that, of six competing schools, the men’s lacrosse team would have been the smallest team in the conference. Id. Altogether, the evidence indicates that Quinnipiac’s men’s rosters tended to be smaller than the national and conference averages and its women’s rosters tended to be larger. That general finding does not hold true for all sports. If the University’s 2009-10 roster count is compared to the NEC’s data for the 2008-09 season, the men’s cross-country team would have been bigger than seven other competing teams, and the men’s basketball team would have been the biggest in the conference; meanwhile, the women’s volleyball, indoor track, and outdoor track teams would have been near the NEC’s median size. But, aside from those exceptions, the plaintiffs are correct that the University appears to set its men’s rosters at relatively small numbers and its women’s rosters at relatively high numbers. The disparity between the University’s rosters and the national and conference averages may not be great — with the exception of women’s indoor and outdoor track, no Quinnipiac team is more than four players above or below the national average. Nonetheless, the plaintiffs have proven that some disparity exists. B. Indoor and outdoor women’s track and field In the 2009-10 season, Quinnipiac fielded four running teams: a men’s cross-country team, a women’s cross-country team, a women’s indoor track team, and a women’s outdoor track team. All four of those teams are coached by Carolyn Martin, who was promoted to the head coaching position at the end of the 2008-09 academic year; previously, Martin had served as an assistant coach to the women’s cross-country, indoor track, and outdoor track teams, and was a former student and varsity athlete at Quinnipiac. As head coach, Martin has two paid assistant coaches and two unpaid, volunteer coaches. Martin is the only Quinnipiac coach responsible for overseeing four separate teams. In fact, only one other University coach' — -Mike Quitko, who coaches the men’s and women’s tennis teams— coaches multiple teams, and he only oversees two squads. As head coach for the men’s cross-country, women’s cross-country, indoor track, and outdoor track teams, Martin is responsible for, among other duties, training her athletes, preparing them for meets, managing the teams’ budgets and operations, and recruiting players. McDonald testified that it was common for schools to appoint one coach for all of their running teams, and then have those coaches delegate responsibilities for specific teams to assistant coaches. In other words, the coaching and management of Quinnipiac’s varsity running teams was a standard practice among NCAA schools. The NCAA recognizes cross-country, indoor track and field, and outdoor track and field as different sports for purposes of athletic eligibility. 2009-10 NCAA Division I Manual (Def. Ex. EP) § 14.2.3.3, at 138. Furthermore, those sports have different rules of competition and their own championships. Women’s cross-country races are five or six kilometers long. Races can be run on all kinds of surfaces and can be either flat or hilly; in a single race, a course is likely to take runners across a variety of terrains. Track and field, by contrast, offers races of various lengths, ranging from short sprints, to middle-distance contests, to long-distance races of a mile or more. And, although it is true that tracks can differ in some ways, such as in their lengths, their surface material, whether they are indoors or outdoors, and the degree to which they are banked, most tracks are similarly constructed as an unbroken oblong loop. In addition to the track races are the field events, which can be distilled into two types: jumping events, such as horizontal contests like the long jump or vertical contests like the high jump; and throwing events, such as the shot-put, discus, and javelin. Cross-country offers no field events; a cross-country meet, rather, is a single race that all of the competitors run at the same time. In a cross-country race, a team generally may enter as many runners as it wishes, although only its top five finishers will score. A team’s final score is the sum of those five runners’ places, and the winning team is the squad with the lowest total score. Teams may, as a matter of strategy, enter more than five runners in a single meet. Entering more runners hedges against the risk that an individual entrant might have a bad race or injure herself on the course and finish in a slower time. A team may also enter a large number of runners with the hope of “displacing” their opponents’ runners. Because a team’s score is determined by adding the places of its top-five finishers, having a sixth runner finish before an opponent’s fifth runner will increase, or “displace,” the opponent’s final score by one more point. That will result in a higher opponent’s total score, and a greater likelihood that the team will beat that opponent. The cross-country scoring system is markedly different than the scoring system used at a track meet. At a track meet, a school enters athletes in a variety of events, and a team’s total score reflects the cumulative scores of its individual members. Many track meets are not scored at the team level, but only provide individual competitors the chance to excel in their respective events. At those meets, there is less incentive to enter many entrants in the same competition. Martin did testify, however, that a track team may include more distance runners than sprinters for events in which a team score is compiled. A sprinter, according to Martin, is able to compete in multiple races at a track meet because a short race is less physically taxing than a distance competition; for instance, a sprinter could compete in both the 100 and 200 meter dashes, but a distance runner is unlikely to run both the 5,000 and 10,000 meter races. Therefore, a track team may elect to carry fewer sprinters than distance runners in order to be as efficiently competitive as possible. Quinnipiac cross-country meets and practices are run on campus, but the University has neither an indoor nor an outdoor track for competition. Instead, indoor and outdoor track athletes must compete off campus at a local high school. The University’s indoor track team practices on a track in the school’s recreation center, but the outdoor track team practices off campus. According to Samuel Seemes, Quinnipiac’s track and field expert witness, it is not rare for a university to have a track team without also having a track on campus, and several other college and university track teams practice off campus and are unable to host home meets. Quinnipiac also does not enter any athletes in field events because it lacks the coaching and facilities for athletes to train for those events. According to Martin and Seemes, it is not uncommon for schools to enter athletes only in track events; both witnesses recalled other universities that enter their track- and-field participants in races only, to the exclusion of field events. Neither Martin nor Seemes, however, was able to state whether the sizes of Quinnipiac’s track-only track and field teams were similar to the rosters of other universities whose track and field teams only competed in races. In 2009-10, the Quinnipiac women’s cross-country team had 18 members, and its indoor and outdoor track teams had 30 members; the men’s cross-country team had 13 members. All of the women’s cross-country runners participated on the indoor and outdoor track teams. In other words, 60 percent of the indoor and outdoor track teams consisted of athletes who ran cross-country in the preceding fall season. The remaining 12 members on the women’s indoor track team also participated on the outdoor track team. Thus, every participant on the women’s cross-country, indoor track, and outdoor track team competed on at least one other squad. It is not unusual for cross-country runners to participate on track teams, and for track participants to run in the indoor and outdoor seasons, however. Both Martin and Seemes testified that runners often participate in more than one season. In fact, Martin said that many runners expect to compete across seasons. She testified that she was having difficulty recruiting male cross-country participants because the University did not offer them track opportunities in the winter and spring seasons and would not sponsor those runners to compete individually in any winter or spring track meets. Under NCAA regulations, a Division I cross-country team must compete in at least six contests over the course of a year, with at least five athletes in each contest. NCAA Division I Manual (Def. Ex. EP) § 20.9.4.3, at 311. Indoor and outdoor track teams must compete in at least six meets in each season with a minimum of 14 participants, id., and may not compete in more than 18 meets combined, id. § 17.24.5.1, at 271. The women’s cross-country team ran eight races in the 2009-10 season, including the NEC championship and the NCAA regional finals; the women’s indoor track team entered runners in eight meets, but only met the minimum 14 participants to compete as a team in six contests; and the women’s outdoor track team entered runners in seven meets, but only met the minimum 14 participants in six of them. Martin testified that, during the indoor and outdoor track season, all of the athletes listed on team rosters practiced, even if they were not competing in an upcoming meet. Injured players were responsible for participating during practice, too, although their practice time consisted of physical therapy, strength training, conditioning, and other rehabilitative exercises. All of the athletes were expected to attend team meetings and each roster member received coaching, equipment, warmup gear, and shoes; only athletes who were competing, however, received racing uniforms. Track team participants were also entitled to athletic benefits, such as travel and per diem costs for away meets, as well as scholarship aid. There are four full-tuition scholarships available for all of the women’s cross-country, indoor track, and outdoor track runners. Those scholarships may be divided among participants on the teams. In 2009-10, the four scholarships were distributed among women who participated on all three of the women’s running teams. Martin did not extend any scholarship funds to athletes who ran only on the indoor and outdoor track squads. Historically, Quinnipiac’s running program has prioritized distance running over sprinting. Thus, the women’s cross-country team has been the predominant squad relative to the indoor and outdoor track teams, and the distance track events have been emphasized more than sprinting contests. Spreadsheets prepared by Martin illustrating the races Quinnipiac track runners entered during the 2009-10 seasons reveal the dominant role that distance running played in the University’s women’s running program. See Def. Exs. AJ & AR. The following chart was compiled using those exhibits. It shows the number of track events that participants entered in the women’s indoor and outdoor track seasons, and identifies those events as sprint, middle-distance, or distance competitions. A sprint is defined as any race equal to or shorter than 400 meters; a middle-distance race is defined as any race greater than 400 meters but shorter than 1,500 meters; a distance race is defined as any race greater than or equal to 1,500 meters. If an athlete ran an event more than once at a meet — e.g., if an athlete ran in a qualifying heat and then ran in the finals for the event — participation in that event for that meet is counted only once. SPRINT MIDDLE-TEAM EVENTS EVENTS -DISTANCE DISTANCE TOTAL EVENTS EVENTS Indoor track 26 48 62 136 Outdoor track 24 31 81 136 Combined track 50 79 143 272 The chart shows the teams’ heavy emphasis on distance running. In the indoor track season, 45.6 percent of all races entered by Quinnipiac runners were distance events. At the typical indoor meet, however, only about one-third of track races— not including field contests — are distance events. In the outdoor track season, distance events ballooned to 59.6 percent of races entered by Quinnipiac runners, even though distance competitions continued to constitute only about one-third of outdoor track events. When the two seasons are taken together, 52.6 percent, or slightly more than half, of all the track events Quinnipiac’s women runners entered qualified as distance contests, in comparison to 18.4 percent sprint events and 29.0 percent middle-distance events. In addition to the number of distance races entered, it is also clear that the cross-country runners are major contributors to the indoor and outdoor track teams. Using Martin’s spreadsheets, Def. Exs. AJ & AR, and isolating the runners who participated on the cross-country team, see Def. Ex. AB (women’s cross country squad list), reveals that cross-country participants ran in 74 races in the indoor season, which constituted 54.4 percent of all of Quinnipiac’s events, and in 68 total races in the outdoor season, which constituted 50 percent of the school’s events. WOien the seasons are combined, the cross-country runners accounted for 52.2 percent of the teams’ total races entered. The statistical evidence of the teams’ emphasis on distance running is consistent with Martin’s description of the operation of the indoor and outdoor track programs. Martin testified that there is a self-reinforcing effect to focusing on distance running. Quinnipiac has traditionally been a strong school for distance running; the University has won the last five NEC women’s cross-country titles. But Martin said that, even with improved distance runners, her indoor and outdoor women’s teams would be unlikely to finish better than third in their current makeup; moreover, she admitted that the teams would be unlikely to ever win a conference title without competitors in field events. According to Martin, it therefore makes sense for her to continue emphasizing cross-country in order to maintain the University’s success, rather than commit to competing in field events for which she currently lacks infrastructure, coaching, and funding. Female runners at Quinnipiac are unique in their competition for different teams across seasons. McDonald confirmed that, except for women’s runners, very few, if any, University athletes participate in multiple varsity sports. Lopiano agreed that such competition across sports was rare in college athletics except for runners; she said, however, that at most schools women’s participation in cross-country, indoor track and field, and outdoor track and field presents no Title IX problem because schools offer the same opportunities for male distance runners. But that is not the case at Quinnipiac, where neither indoor nor outdoor track is offered for men, and where male cross-country runners are not permitted to represent Quinnipiac as individual entrants at indoor and outdoor track meets. Finally, at her deposition, Martin described the track teams as adjuncts of the cross-country team, and further noted that cross-country runners were required to participate in the indoor and outdoor seasons. In other words, as a condition of competing on the cross-country team, runners also had to participate on the indoor and outdoor track teams. That was true even for cross-country athletes who were injured or red-shirting a season; they, too, had to join the indoor and outdoor track teams, even though they had little to no hope of competing during the season. During the 2009-10 indoor track season, three cross-country runners did not compete at all because of injury; one runner only competed in one meet because of injury; and one runner did not compete because she was red-shirted. See Def. Ex. AB (2009-10 cross-country squad list); Def. Ex. AJ (indoor track competition spreadsheet prepared "by Martin). And during the 2009-10 outdoor track season, five cross-country runners did not compete at all because of injury, and one runner did not compete because she was red-shirted. See Def. Ex. AB (2009-10 cross-country squad list); Def. Ex. AR (outdoor track competition spreadsheet prepared by Martin). The evidence showed that such a requirement was unique to cross-country runners, and no other Quinnipiac varsity athlete was expected to participate on a second team in order to compete in his or her favored sport. Furthermore, no evidence was introduced showing that such a requirement was common among other Division I cross-country programs. C. Competitive cheer Following the 2008-09 academic year, Quinnipiac decided to create a new women’s varsity sport: competitive cheer. Competitive cheer is an outgrowth of traditional sideline cheerleading. Competitive cheer teams use many of the moves and techniques that sideline cheer squads have developed over the decades, and their routines look like more athletic and aerobatic sideline cheer orchestrations. But whereas sideline cheerleaders primarily work to entertain audiences or solicit crowd reaction at other teams’ games or school functions, competitive cheer teams strictly engage in sport. Participants do not perform for a crowd’s approval or involvement — they compete to win. In order to distinguish their activity from their sideline roots, cheer teams do not attempt to elicit crowd response; generally do not use pom-poms, megaphones, signs, or other props associated with traditional cheer-leading teams; do not wear skirts and sleeveless or cropped tops, but wear uniforms consisting of shorts and jerseys, much like what women’s volleyball players don; and emphasize the more gymnastic elements of sideline cheerleading, such as aerial maneuvers, floor tumbling, and balancing exercises, to the exclusion of those activities intended to rally the watching audience. As I noted in my preliminary injunction ruling, competitive cheer is an athletic endeavor that “could be easily described as ‘group floor gymnastics.’ ” Biediger, 616 F.Supp.2d at 295. Despite its athletic elements, however, competitive cheer is not recognized as a sport by the NCAA. Nor does the NCAA recognize competitive cheer as an “emerging sport,” a provisional designation that allows a university to count the activity toward NCAA revenue distribution and minimum sports sponsorship requirements. 2009-10 NCAA Division I Manual (Def. Ex. EP) § 20.02.25.1, at 298. Furthermore, the Department of Education has not recognized competitive cheerleading to be a sport. Indeed, schools reporting their athletic participation data to the Department of Education under the Equity in Athletics Disclosure Act (“EADA”) are instructed to report their cheerleading team rosters only if they have received a letter from OCR determining that their cheerleading squads are legitimately engaged in sport. U.S. Dep’t of Educ., Office of Postsecondary Educ., The User’s Guide for the Equity in Athletics Act Web-Based Data Collection (2009) (PI. Ex. 110), at 19, 21. Jeff Webb, who testified at trial, is at the center of competitive cheer’s history and maturation. Webb is the president of Varsity Brands, Inc., an athletic equipment manufacturer that caters to cheerleading teams. Through its subsidiary organizations the National Cheerleading Association (“NCA”), Universal Cheerleaders Association (“UCA”), and the United Spirit Association, Varsity Brands also holds competitions among scholastic cheerleading teams and private, “all-star” cheerleading teams. Webb’s involvement with the sport of cheerleading began in the early 1970s, when he was a student cheerleader at the University of Oklahoma. Shortly after graduating, he started UCA, which would eventually turn into Varsity Brands, with the intent of creating a business that would sell cheerleading equipment and offer training camps for cheerleaders. To help promote his business, Webb began holding competitions for cheerleaders, the first of which took place in 1980. Soon, those competitions began to be televised, and cheerleading began to be recognized as a form of competition. Webb testified, however, that he never imagined that his competitions would establish a new sport; rather, he understood his competitions as a publicity vehicle for his startup business. After forming UCA, Webb formed the NCA as a new competition format for colleges and universities; the NCA held its first competition in 1990, and holds a national championship in Daytona Beach, Florida, every year. That Daytona Beach championship is, today, the marquee competitive event for cheerleaders. Webb said that, although technically different than the UCA, the NCA competition rules were essentially identical to those of the UCA. Today, the two organizations mainly differ in terms of the teams that compete in their competitions: UCA events are reserved for sideline cheerleading teams, while both sideline and competitive cheer-leading teams may compete in the NCA format. In the NCA college competition, teams compete in two segments: a minute-long “spirit” exercise judged by the enthusiasm the team generates in the crowd, followed by a two- to two-and-a-half-minute team performance to showcase the team’s athleticism and gymnastic skill. The NCA welcomes all college and university teams to participate in its competitions, and has divisions to distinguish schools based on their size and ability; the organization, however, does not have a progressive or playoff system to winnow a final champion. Besides hosting competitions among cheerleading teams, NCA events also feature competitions among school dance teams and mascots. As was true with the UCA’s competitive events, Webb did not believe that his NCA competitions would eventually establish a sport that would, some day, be recognized by athletics organizations such as the NCAA. Instead, he envisioned NCA competitions as a further promotion of his cheerleading supply business. Indeed, testimony at trial revealed that the NCA’s scoring system was intertwined with the promotion of Varsity Brands. During the “spirit” portion of the competition, cheer-leading teams are awarded points for using props, such as pom poms, sold by Varsity Brands; the more props a team uses, the more points that team receives. Webb testified that he was not averse to competitive cheerleading eventually becoming an independent sport. Still, as a sideline cheerleading purist, Webb wants competitive cheer to be distinguished from traditional sideline cheer out of concern that competitive cheer will threaten his competitions — for instance, competitive cheer might annex certain moves and routines from sideline cheer — and lead to confusion about the difference between the two forms of cheerleading. Webb did not think that the rise of competitive cheer would hurt his business, however. On the contrary, he welcomed its emergence, believing that the creation of new competitive cheer teams at high schools, colleges, and universities would increase demand for Varsity Brands’ products and services. Previously, Quinnipiac had a women’s sideline cheerleading unit that performed at men’s basketball games and, once or twice a year, entered competitions against other schools. Those competitions included participating in the NCA championship in Daytona Beach. For the 2009-10 academic year, however, a new cheerleading team was created for the exclusive purpose of competition. Mary Ann Powers, the coach of Quinnipiac’s sideline cheer unit, was asked by the athletics department to head the new competitive cheer program in March 2009. She accepted the offer and assumed her coaching role in the summer of 2009. In addition to Powers, two paid assistant coaches and one unpaid volunteer coach help manage the team. Before Powers was hired, Quinnipiac released an advertisement for the position of competitive cheer coach. Among other prerequisites, applicants were supposed to have experience recruiting athletes and familiarity with NCAA and NEC rules and regulations. When Powers accepted Quinnipiac’s offer, she was not qualified to recruit and was not familiar with the NCAA or NEC rules. Indeed, she was not cleared to recruit any athletes off-campus until the spring of 2010, when she passed the NCAA’s recruitment examination for coaches. The competitive cheer team was assigned an initial roster target of 30 players, a number that initially disappointed Powers, who had hoped to have 32 or 36 athletes. At the very least, she hoped to have a roster number that was a multiple of four, which represents the number of competitive cheer positions and the number of players necessary to perform certain cheer routines. Thompson told her that she would have to make do with 30 athletes for the 2009-10 season, but said that he would consider requests for more players in the future. Although Powers was unable to recruit players off campus for the 2009-10 season, she said that her on-campus recruiting was sufficient to field a competitive team. Powers testified that there was very strong interest in the competitive cheer team among undergraduates, and that incoming and prospective Quinnipiac students contacted her to express their desire to participate. New students who were interested in joining the competitive cheer team were sent a questionnaire about their cheerleading and/or gymnastics experience. Powers also requested that interested students submit videos demonstrating their cheerleading skills. For the first season of the competitive cheer team, Powers elevated 16 students from her former sideline cheer unit. According to Powers, those students had participated on the sideline cheer team with an eye towards eventually competing exclusively; she testified that those members relished the opportunity to travel to the NCA championship in the spring and wanted to compete like other varsity athletes. After adding the veteran sideline cheer participants, there were 14 spots to be filled. Thirteen were freshmen, and the remaining player was a graduate student and former gymnast who still had a year of NCAA eligibility- Quinnipiac allotted six full-time scholarships, worth $47,500 each, to the competitive cheer team. In the team’s first year, Powers only extended one and one-half scholarships, and distributed them among ten members in amounts ranging from $3,000 to $9,000 a year. Powers testified that she was cautious in how she assigned her initial scholarships because she wanted funds to be available for players in future classes. The competitive cheer team had an initial budget of $50,000, which covered the costs of travel, lodging, equipment, and other operational expenses. Ultimately, however, the team spent $130,000 during its inaugural season. That increase in costs is attributable, in part, to team trips to compete in Georgia and Florida. Although the NCAA does not recognize competitive cheer as a varsity sport, the Quinnipiac competitive cheer team still followed applicable NCAA rules, such as requiring all participants to be cleared by Quinnipiac’s medical staff before competing and following practice time restrictions. Team practice, which focused on strength conditioning, skill development, and improving the team’s coordination and timing, was limited to eight hours in the off-season and 20 hours during the season, competitions included. Per NCAA rules, team members were also allowed to engage in individual practices not mandated by the team. Powers testified that her athletes received benefits on par with those received by other varsity teams, such as equipment; access to facilities, trainers, and strength and conditioning coaches; the opportunities to participate in “power hour” study halls and in the University’s “positive play” community service program; and publicity about the team’s news and successes. The competitive cheer team also acknowledged its own best athletes with team awards, and one participant won a university-wide scholar-athlete recognition. The competitive cheer team, however, did not receive locker room space, which the University extended to other varsity teams. The Quinnipiac competitive cheer team also had to buy separate catastrophic insurance coverage because, as a non-recognized sport, it was not covered by the NCAA’s insurance program. In addition to forming a new competitive cheer team, Quinnipiac joined and helped establish a nascent intercollegiate competitive cheer organization, the National Competitive Stunt and Tumbling Association (“NCSTA”), with seven other schools: the University of Maryland, Fairmont State University, the University of Oregon, Azusa Pacific University, Baylor University, Fort Valley State University, and Ohio State University. Five of the NCSTA schools — Quinnipiac, Maryland, Fairmont State, Oregon, and Azusa Pacific — fielded varsity competitive cheer s