For four quarters, at least, Graham Mertz, the quarterback for the University of Wisconsin, lived as if dreaming. A lanky red-shirt freshman, Mertz took the field for his first start as a Badger, in the Big Ten Conference’s season opener, on Friday, October 23rd, against the University of Illinois, and completed his first pass attempt, a gain of three yards. His second attempt gained twelve. His third, a pretty touch pass, went for a touchdown. Mertz completed his fourth attempt, his fifth, his sixth. He followed a conservative strategy, mostly screens and short passes, but grew more daring. He rolled out on a play fake and threw a touchdown off his back foot. He whipped another over the middle, a precise dart into the end zone. With less than a minute to go in the first half, he launched a fifty-three-yard touchdown pass to a streaking wide receiver. By then, he was fourteen for fourteen, with four touchdowns—he threw the ball seventeen times before he missed a target, and only then because the ball was dropped. (Graciously, he blamed himself.) He finished the game twenty for twenty-one, with five touchdowns. Patrick Mahomes, the best quarterback in the world, tweeted his amazement. J. J. Watt, the three-time N.F.L. defensive player of the year, posted a video of a man using a flamethrower, in tribute. Wisconsin won, 45–7.
In a normal season, the freshman quarterback would have looked up to see eighty thousand people cheering for him, chanting and singing and calling his name. But this was not a normal season. The cold metal bleachers were bare, and the cheers that followed each of his touchdowns were fake. When the song “Jump Around” played before the fourth quarter, the small cluster of cardboard cutouts that had been installed in the stands did not jump. When it’s empty, Wisconsin’s Camp Randall Stadium has about as much charm as an old airport concourse. But a touchdown is still a touchdown, even if no crowd makes a sound. “He was smiling cheek to cheek after the first touchdown,” the tight end Jake Ferguson said of Mertz afterward. “He knew, and everyone in that huddle knew, that we were rolling.”
Two days later, Mertz tested positive for the coronavirus. Then it was reported that Wisconsin’s backup quarterback had tested positive as well. By midweek, the number of positive tests within the football program was up to twelve, and included the head coach—and the team’s next game, against the University of Nebraska, was cancelled. A week later, the number of cases on the team was up to twenty-seven. Two players for Illinois also tested positive, although it was unclear if they were infected during the game or somewhere else entirely. Wisconsin’s next game, against Purdue, was called off.
Purdue vs. Wisconsin was one of ten top-tier college-football games that was postponed or cancelled that weekend. The Badgers returned to the field a week later, on November 14th, routing Michigan, but fifteen other games were not played as originally scheduled that weekend, including four of seven games that were to have featured teams from the powerhouse Southeastern Conference (S.E.C.). So far, more than eighty games have been cancelled, and several conferences are only a few weeks into their schedules. But the season has rolled on, even as coronavirus cases have skyrocketed across the country. If we could say definitively, in precise and dramatic detail, that the former was connected to the latter, then presumably universities would face public pressure to stop the games. But this is probably the wrong way to think about life during a pandemic. And the approach that we have taken to college football likely reflects our broader failure to grasp what it takes to shut down the virus.
Back in August, the Big Ten announced that it was postponing its football season until the spring. Players had begun returning for workouts in June, and team after team around the country had suffered outbreaks of SARS-CoV-2. There were reports of athletes suffering serious long-term effects—an anguished Facebook post written by the mother of one player, which described her son’s struggle with myocarditis, an inflammation of the heart muscle that can follow a viral infection, shook many within the sport.
The Big Ten is the oldest conference in the country and includes some of the largest and most respected universities in the world. The decision to call off the season was hugely costly—the conference was in the middle of a multi-billion-dollar television deal—and hugely unpopular. But the concerns prompting it were grave, and many people expected the rest of the so-called Power Five conferences to follow suit, as the Pac-12 did, just hours later. The season appeared to be on the brink of cancellation.
It turned out that there is no more unity in college football than there is in the rest of the country. Not all medical experts agreed about the riskiness of playing. Some parents of players protested the Big Ten’s decision. Players sued. Commentators railed about the decision on sports radio. A player labor movement merged with, and then morphed into, a movement to play the season. The President of the United States mocked the conference on Twitter. The rest of the Power Five—the Big 12, the Atlantic Coast Conference, and the S.E.C., which, collectively, are home to fourteen of the last fifteen national champions—were not inclined to defer to the Big Ten, and announced that they were going ahead with games. Football brings in more than a hundred million dollars in revenue to top schools, and, although the financial impact for smaller programs is less clear, athletic departments generally depend on money from football. Programs had not stashed away revenue for a rainy day or a pandemic. They had spent it—on other sports, expensive coaches, and lavish facilities.
The season began without the Big Ten. Some players got sick, others tested positive without showing symptoms, and still others were held out while contact tracing was being conducted. Several big games were cancelled, but enough were played to make it appear that something like a full season would eventually be completed. A couple of weeks after the nation’s first game had been played, the commissioner of the Big Ten, Kevin Warren, announced that the conference’s member schools had taken another vote: the Big Ten would resume playing football at the end of October. The Pac-12 soon released a similar statement, as did smaller conferences that had also put off playing. All but a handful of teams across the country would participate in the season.
Big Ten university presidents said that the reversal was prompted by new medical advice. The conference had enlisted a private company to provide daily rapid tests to all teams; experts had provided stringent health and safety guidelines. There was evidence that heart complications might be rarer in athletes than had been feared—and doctors and researchers had developed a screening protocol for detecting myocarditis.
There was truth to all of this, but it was not the whole story. While some medical experts believed that you could find a safe-enough way to play, most epidemiologists and public-health professors I spoke to were not in favor of it. What had really changed between August and September was that college football had come back without the Big Ten, outbreaks had happened, and there had been little outcry. By the time that the Big Ten resumed play, in late October, Notre Dame, Florida, Cincinnati, and several other ranked teams had all had coronavirus clusters. The L.S.U. head coach, Ed Orgeron, casually mentioned that “most” of his players had contracted it during the summer. (He seemed to frame this as a plus, because the team presumably wouldn’t need to worry as much about the virus as the season went on; two months later, the team’s game against Alabama was postponed after more members of the program tested positive.) Jamain Stephens, a defensive tackle at a Division II school outside of Pittsburgh, died of complications from COVID-19. But football fans kept tuning in. Teams are not generally forthcoming about infections, both out of concern for the privacy of their players, and, at least according to the University of Oklahoma football coach Lincoln Riley, because it gives them a competitive advantage. Only rarely did stories of players struggling with the illness, or even being hospitalized, appear in the press. COVID cases and mandatory quarantines were described by coaches and analysts as an annoyance or a misfortune roughly on par with a sprained ankle. Television contracts were being honored. Money flowed to schools and to coaches, if not to players. The playoffs were on the horizon. Fans went to bars. Families and friends gathered for game day. The lower stands of stadiums at many schools were filled with people. You could turn on the television every weekend and see some semblance of normalcy.
The introduction of daily rapid testing had seemed like a real advance—and it was loudly touted by the Big Ten. The argument was that widespread, frequent use of tests with quick turnaround times could allow for a return to more or less normal activity, even if those tests were not as accurate as those that take longer to produce results. The tests wouldn’t stop people from catching the virus, but, theoretically, they could identify infected people before they’d interacted much with others. (The Pac-12 is also using daily rapid testing. Players in other Power Five conferences are tested less often, typically using the more accurate P.C.R. tests, while teams in less prestigious conferences have differing—and sometimes less stringent—testing regimes.) The results of the approach would potentially have ramifications far beyond football. “This is one of the first really high-quality, real-world tests of that concept,” Dave O’Connor, a professor of pathology and laboratory medicine at the University of Wisconsin, told me.
It didn’t quite work. For the first month, testing went smoothly, but Wisconsin’s outbreak occurred anyway—and other programs conducting daily antigen testing had outbreaks, too. At the University of Maryland, twenty-three players tested positive in a two-week span; several Pac-12 teams had to cancel games because they did not have the minimum number of scholarship players available.
When I spoke to O’Connor, in early November, his lab—in collaboration with county and state public-health officials and the C.D.C.—was trying to figure out how the disease had spread within the Badgers football program. The lab was looking for genetic signatures in the team’s positive samples, and was using what O’Connor’s colleague, Thomas Friedrich, called shoe-leather epidemiology to find clues about where the virus had originated. Had there been lapses in social-distancing protocols? Would a more sensitive test have made a difference? Could the outbreak be traced to one source, or was there more than one simultaneous infection? Did the efficacy of daily rapid testing depend on lower rates of community transmission?
O’Connor told me that he’d been partial to the view that, with “really fast turnaround testing, even with lower sensitivity tests, you could enable a safer reopening of schools, a safer reopening of the economy.” And yet, he said, “You can still have an outbreak.” The rash of positive tests on the football team had reminded O’Connor and Friedrich of the need both for data and for humility. When I asked them, over Zoom, if the investigation had changed their views on whether it was advisable to play football this year, Friedrich gave me a wry smile. “My personal view is that it wasn’t advisable in the first place,” he said.
“I started a little more optimistic than Tom on the ability of testing to make it safe,” O’Connor said. “I didn’t think it was advisable, but for another reason. We are in a state where there’s an entire culture around football.” Friedrich chimed in to agree. That culture is intensely social. The bigger problem, as they saw it, wasn’t whether the protocols could protect the players. It was that those protocols wouldn’t protect the rest of us.
This past summer, a group of researchers at Washington State University started thinking about what a football season might mean for the city of Pullman, where the university is based. At that point, Washington State, which plays in the Pac-12, planned to begin its season in September; the question was whether fans would be allowed to attend the games. The perils were obvious: every public-health expert was advising against large gatherings. But what was the precise level of risk? Washington State’s athletic department was running an operating deficit of nearly a hundred million dollars, and around half of its usual revenue came from football, including ticket sales—but the considerations weren’t only financial. Although Washington State has one of the smallest stadiums in big-time college football, its fan base is passionate. Game days are big days in Pullman.
The researchers used mathematical modelling to simulate how visitors coming to a campus community very much like Washington State’s to attend a football game would affect the levels of the virus in the community. The scope of their inquiry was relatively narrow: they did not have anything to say, for instance, about whether a visitor might become infected and bring the virus back to her own community—a risk that “could lead to multiple outbreaks around the country,” as Jill Weatherhead, a professor of infectious diseases at the Baylor College of Medicine, told me. The Washington State paper has yet to be peer-reviewed, but what the researchers found was sobering. In simulations in which a small number of visitors with COVID-19 mixed lightly with a campus community experiencing an uncontrolled outbreak, cases rose in the community by twenty-five per cent. When visitors with a high prevalence of the coronavirus mixed heavily with a campus community where the virus was previously under control, the local increase was more than eight hundred per cent.
The Pac-12 and the Big Ten ultimately decided not to let fans—with the exception of players’ families—into stadiums at all. But the rest of the Power Five, along with several smaller conferences, did. On those campuses, and in those college towns, we don’t yet know whether things played out the way the researchers’ model projected—for one thing, most of the schools are admitting fewer fans than the model estimated. And we may never precisely learn the results of these gambles. Two weeks before the S.E.C. football season began, Oxford, Mississippi, where Ole Miss is situated, had the highest daily infection rate—eighty-five per one hundred thousand people—of any college town whose school was playing football. (Health officials generally recommend the suspension of nonessential activities at anything above twenty-five.) When the team kicked off the season, nearly fifteen thousand fans were in the stands; more than two and a half million people watched on television, many of them doubtless gathering to watch at bars or inside homes. Katie Taylor, an epidemiologist with Mississippi’s department of health, told me, in an e-mail, that the department had not “at this time directly tied cases to outbreaks or clusters associated with football-related activities.” She added, “However, we know that many of these settings, especially gatherings for football watching, can lead to transmission.” The University of Alabama saw a massive surge in cases when students returned to campus. The number of cases had dropped by the time the football season began, and it has not risen precipitously since. But cases in Alabama as a whole have been rising steadily all fall. Karen Landers, an officer in Alabama’s public-health department, told me, “Attendance at sporting events and also attendance at parties relating to sporting events has contributed to outbreaks.” Department policy prevented her from providing specifics, she said.
