The 20% Ceiling and the Counting of Names on Roster Pages: Is NCAA Swimming Being Measured by Passports or by Training History?

Late one April night in Nagoya, I opened the roster page of the University of Florida men's swim team and started counting. Twenty-four names. Fifteen of them had a hometown column pointing somewhere outside the United States: Budapest, Sofia, Johannesburg, Perth, and a couple of places I had to look up on a map. Fifteen out of twenty-four is 62.5 percent. I sat still for a moment, then did the arithmetic anybody can do in four seconds: twenty percent of twenty-four is 4.8. Round down, five spots. If a bill currently sitting in the United States Congress were passed with exactly that number, Florida's men's team would have to explain to ten of those fifteen young people that their place on the list no longer belongs to them.

I keep a notebook of name pronunciations. The habit goes back to a Nagoya Grampus versus Kashima Antlers match in 2026, when I mispronounced the winger Serginho's name three times in the first half and the whole commentary booth laughed. Yavuz Omer Aga is a name I wrote in that notebook recently, with a small exclamation mark in the margin. A Turkish athlete on an American university roster. The name itself is not the issue. The issue is elsewhere: what exactly makes us call a person "international," and is the way we count actually measuring the thing we think we are measuring?

A bill, a number, and a sport chosen as the test case

The proposed Protect College Sports Act sets a hard limit: no more than twenty percent of a college sports roster may be international athletes. In swimming and diving, that limit touches a reality that has existed for decades but has never been counted publicly, fully and systematically. The starting point matters: the twenty percent ceiling is not a performance rule. It is a composition rule. It does not say who swims faster. It says who is allowed to be on the list.

Swimming was not chosen as the test case because it imports the most talent in American college sports. It was chosen because it is the easiest sport to count. Roster pages are public, the lists are short enough for one person to read in an evening, and every line has a hometown field. A men's Power 4 swim roster usually holds twenty to thirty names. The ease of counting creates the illusion that the problem is equally easy to understand.

The data analysis was produced by Leslie Lucas, who runs College Swimming Consulting, and published by SwimSwam. One detail about the author belongs at the top, because it does not invalidate the numbers but can change how they are read: she is the mother of Cooper Lucas, a junior on the Texas men's swim team. A recruiting consultant who is also a current team parent. I saw no sign of bias in the figures presented, but I always note this kind of relationship in my book, next to the names and the meets, because my profession taught me that a data source and the person presenting it are two different stories.

The author herself set limits on her own conclusions: the bill has a long way to go before it passes. She called it an intriguing conversation, not a verdict. She also recorded that complaints about the international share of rosters are rising across college sports discussions and in the SwimSwam comment sections. Put those two statements side by side and you have the whole nature of this story: a policy question with low near-term legal probability, but very real reputational pressure, and public pressure always moves faster than law.

Reading every page, counting every line

The work Leslie Lucas did is technically simple. She opened each school's roster page, read the hometown column, sorted domestic from international, and divided. Because it is simple, it carries immediate persuasive power. The most striking figures sit at the top of the table, and every one of them belongs to the Southeastern Conference.

Florida leads with fifteen of twenty-four, or sixty-three percent. Auburn follows with thirteen of twenty-two, or fifty-nine percent. LSU has eleven of twenty, or fifty-five percent. Tennessee has thirteen of twenty-five, or fifty-two percent. Georgia and Kentucky both sit at fifty percent. At the other end, Duke is recorded with roughly one international athlete, a Turkish man. And the headline aggregate: thirty-one of forty-two men's Power 4 teams exceed the twenty percent threshold. Seven of the teams projected to finish in the top ten at the 2026 NCAA men's championships also exceed it.

I want to pause on the word "exceed," because it hides a small but important arithmetic problem. For a team with twenty roster spots, twenty percent is four people; to exceed it, that team needs five. For a team with twenty-five spots, the threshold is five, and the team needs six to exceed it. For thirty spots, the threshold is six and the team needs seven. In other words, a larger team can drop below the threshold without changing the nature of its recruiting at all, simply by adding a few domestic athletes to fill out the list. That is the kind of detail I jot down when I watch matches: a percentage always depends on its denominator, and coaches can adjust the denominator.

There is a second data-quality point. For the five leading teams, the analysis gives both numerator and denominator, so a reader can redo the division. For Georgia, Kentucky and Duke, the denominator is not given. Duke is described with an absolute number, not a percentage. That gap forces me to slow down. In commentary, a share without a denominator is an unfinished sentence. It may be right, but it cannot be independently checked.

One source, one method, one blind spot

My professional rule comes from the stumble at Toyota Stadium in 2026: never comment before watching at least ninety minutes of footage, and never read a name without verifying its pronunciation through three sources. Three sources, not one. That rule makes me read this dataset with caution, not because it is wrong, but because it is thin.

Every headline figure comes from a single analyst, using a single method, drawing on a single category of source: each school's own roster page. There is no second confirmation, no alternative method to cross-check against. That is not the analyst's failure. NCAA does not publish a nationality database for athletes, so a second source is hard to find. But readers should know what they are reading: a primary dataset, transparent in its method, not yet cross-verified.

And the author identified the biggest gap in her own method. The hometown field on a roster page records what the athlete declares, and it does not say where that athlete was trained. The example given is Kaii Winkler, an athlete raised and trained in the United States who competes internationally for Germany. Counted by the hometown field, Winkler is international. Counted by training history, he is domestic. One person, two opposite classifications, and both defensible.

If the flaw leans in any direction, it leans toward inflating the number rather than shrinking it. The count of international athletes may be higher than the count of foreign-developed athletes, because dual nationals who grew up in the United States, trained in the United States and attended American high schools, but chose to represent a family's country of origin, all land in the international column. Nobody knows the size of that distortion, and because nobody knows, it cannot be ignored in a policy debate built on the number itself.

The paradox named 2026

One detail in the dataset made me read it again and again, and I believe it is the most important element of the whole story, even though it appears only as a passing comparison.

In the 2026 season, the international share of incoming freshmen in men's and women's swimming and diving was below twenty percent. Today, measured across full rosters, thirty-one of forty-two men's teams exceed twenty percent. These two figures do not directly contradict each other, because they measure different things at different times. Put side by side, though, they open an explanation the original analysis does not explore.

Picture a roster as a bathtub and freshman recruiting as the faucet. The thirty-one-of-forty-two figure measures the water currently in the tub. The below-twenty-percent figure from 2026 measures the flow through the faucet. If the flow is below twenty percent while the water level is above twenty percent, then the excess in the tub is a legacy of earlier recruiting classes, not the result of current recruiting. And if that is right, time alone, without legislation, would bring the tub down. Four years is the length of one college cycle. A roster can turn over almost completely in four years.

I should be explicit that this is my inference, not the original analysis's conclusion, and it depends on an unverified assumption: that the 2026 freshman figure is a valid proxy for current recruiting flow. At least three things could break that assumption. First, one year's freshman data can be distorted by one year's admissions policy. Second, some schools may have pivoted harder toward international recruiting after 2026. Third, freshman intake and international transfers from other programs are not counted in the same frame. Even so, it is a question anyone writing a law ought to answer before writing the law.

The mechanics of a twenty percent ceiling: the arithmetic of eleven spots

If the cap were applied as written, look at the concrete subtraction. Florida has fifteen international athletes on a twenty-four-man roster. The cap allows 4.8, so four or five depending on rounding. The gap is ten to eleven spots. Auburn has thirteen of twenty-two; the cap allows 4.4; the gap is nine. LSU has eleven of twenty; the cap allows four; the gap is seven. Tennessee has thirteen of twenty-five; the cap allows five; the gap is eight. For Georgia and Kentucky at fifty percent, the gap is roughly a third of the roster.

This is why I call it a non-marginal rule: it does not touch edge cases, it touches the programs that are winning. A twenty percent ceiling in men's Power 4 swimming is not housekeeping. It is a large-scale restructuring of personnel at exactly the programs producing America's fastest college swims.

The impact would not be evenly distributed, and this is the part I think is least discussed. A swim roster is not a uniform block. It is a collection of event clusters: sprint freestyle, distance freestyle, butterfly, backstroke, breaststroke, individual medley, and relays. A roster spot is rarely a generic spot; it usually corresponds to an event group a coach is trying to fill. When a program has to cut ten spots, it does not cut in tidy proportion. It cuts where it can, which usually means the groups with the fewest domestic athletes waiting behind them. The result can be a team that does not lose strength evenly, but is hollowed out in specific event groups.

I sat in a commentary booth in Kazan in 2026 when Kylian Mbappe ran 37 kilometres per hour and I shouted until my throat hurt, and I learned something that day: at the elite level, a team's strength is not its best athlete, it is its fourth-best. A relay needs four people. Four, not three. A swim team that loses one athlete in the right event group is like a football team losing a full-back: nobody notices in the first half, everybody notices in the eighty-fifth minute.

One more technical detail, rarely mentioned: the sport is called swimming and diving. If the twenty percent cap is computed on the combined roster, the swimming percentage may be higher than the reported figure, because at many schools the diving squad is almost entirely domestic. A team of twenty-five with six divers means a denominator of twenty-five, but the excess sits almost entirely in the remaining nineteen. That is the kind of detail you only notice if you have read relay sheets, and it shows how a per-head cap may not map onto the competitive reality it intends to regulate.

Roster spots and scholarships are two different stories

A confusion recurs in American college sports debates: treating a roster spot as if it were a scholarship. A Division I men's swim program operates with the equivalent of roughly 9.9 full scholarships while carrying twenty to thirty athletes. Most of the roster is not on full funding, and many are walk-ons paying their own way or receiving minimal support.

Two consequences follow. First, a twenty percent cap attacks the right to be listed, not the flow of scholarship money. Administratively, that is a very different instrument from capping international scholarships. Second, in a period when American college athletic budgets came under pressure after 2026, when dozens of swim programs were cut or reduced, international athletes paying international tuition, in full or in part, became a non-trivial line in some programs' finances. I offer that as a hypothesis, not a finding, because the original analysis provides no tuition or scholarship data. But anyone who has looked at a college athletic budget knows those two lines are never separate.

The SEC is not America in miniature

The conference concentration is the most telling detail after the aggregate. All five teams above fifty percent are in the Southeastern Conference: Florida, Auburn, LSU, Tennessee, Georgia, Kentucky. That is not a random distribution. A national cap, if enacted, would apply to every conference on paper, but in practice it would land hardest on exactly one.

Several explanations are plausible, and I will only offer ones with a reasoning basis. First, coaching networks. In a small international sport like swimming, information about young athletes travels through relationships between American college coaches and foreign club coaches. A conference with more coaches experienced in international recruiting holds an information advantage. Second, facilities and climate, which matter to athletes arriving from Europe and Australia. Third, a success effect: a program that once brought in international athletes and helped them shine becomes a destination recommended by word of mouth inside the athlete community itself.

There is a more structural reading: the NCAA is operating as a destination market in the global talent supply chain, not only as a developer. In the traditional model, a young foreign athlete trains at home and comes to America to finish. With cases like Winkler, the flow is more complex: athletes developed in America from childhood, competing under another country's flag. America is simultaneously an importer of talent and an exporter of results to other national federations. A rule that looks only at passports cannot distinguish these two flows running in opposite directions.

Why swimming, and not tennis or track and field?

My profession makes me obsessive about cross-sport comparison. Whenever a sport claims to be special, I look for another sport with a similar structure and test the claim.

In American college sports, the international share in tennis has long been very high, often cited as among the largest in the NCAA system. Track and field is similar, with a community of athletes from Kenya, Jamaica, Europe and Asia. Men's and women's soccer are among the most international sports of all. If the criterion is "the sport most internationalised," swimming is not at the top of the list.

So why is swimming the sport chosen to tell this story? Three reasons come to mind, and all three are professional rather than athletic. First, swimming data is easy to read. Second, swimming has a specialised, close-knit media ecosystem in which a roster analysis can spread fast. Third, swimming is expensive to run, which puts it in the firing line whenever universities cut programs. A sport in the budget's crosshairs easily becomes a sport in the policy crosshairs.

This leads to a methodological warning: if a national rule applies to one sport only, it does not create fairness across sports, it creates a new kind of unfairness between the sport that is covered and the sports that are not. And if it applies to all sports, the scale of the problem is entirely different from what is being presented.

Legal and enforcement questions: a definition that does not exist

The most important enforcement question is the one usually skipped in debates about principle: who counts as an international athlete?

The hometown field on a roster page is not a legal definition. It is an administrative field entered by the athlete or a school's communications staff. A legally binding rule cannot rest on a field like that. It needs a standard definition: nationality at the time of enrollment? Place of birth? Where high school was attended? Where the athlete was primarily coached during their developmental years? Years of residence in the United States? Each choice produces a different number and a different set of affected athletes.

In Winkler's case, the four criteria usually cited all place him in the American column, while his international sporting passport places him in the German column. A rule written around passports would classify an athlete developed entirely within the American system as a person to be restricted. This is what I call a definitional error, and in my experience covering sports disputes, definitional errors cause more long-term damage than numerical ones.

Beyond definition, there is the question of authority and timeline. A federal bill affecting a group's access based on nationality would almost certainly face legal challenge, stretching over multiple sessions. During that period, what operates is not law but anticipatory recruiting behaviour. A coach does not need a law to start reconsidering strategy; knowing a law is being discussed is enough to change behaviour.

And if a cap were applied, enforcement would most likely occur at the admissions funnel rather than by cutting current athletes. With a four-year roster cycle, filtering at intake could bring the share below the threshold relatively gently. That is why I consider grandfather clauses not a minor concession but an indispensable part of any rule that wants to survive legally.

Who actually gets hit

In every quota debate, one group is turned into a number and disappears as a subject: the people inside the quota.

This bill would not affect "international athletes" as an abstraction. It would affect people aged eighteen to twenty-two who have signed multi-year commitments to a college program, moved to another country, and built their entire academic and athletic plan around a list bearing their name. Their transfer options are limited, both procedurally and in time. The most exposed are not new freshmen but second- and third-year athletes committed to long-term arrangements with few alternatives.

I once wrote about a young swimmer who left her family at seventeen for another country. I remember her saying the hardest part was not the training, but the first night in the dormitory, when she realised nobody in the corridor spoke her language. The athletes on these rosters are in those rooms too. They are not a policy indicator. They are young people who bet their lives on a system that is now weighing whether to remove them from the list.

The contrarian angle: the ceiling measures the wrong quantity

This is where I want to be direct about what I believe is most true in this story.

The debate has been framed as a debate about nationality. What is actually at stake is place of development. A twenty percent ceiling based on passports does not measure where an athlete was trained; it measures the flag that athlete carries onto the podium. The two overlap in most cases, but they diverge in enough cases to make a policy built on them systematically imprecise. And a systematically imprecise policy hits the wrong people systematically.

The second counter-intuitive point lies in the 2026 number itself. If the recruiting flow in men's swimming has been below twenty percent since four years ago, then today's excess is a stock, not a rising trend. I know that is less appealing than a story about a rising wave, and media prefer waves to stocks. But if it is right, writing a law to stop a wave that has already receded misallocates policy resources.

The third counter-intuitive point: scarcity does not create athletes. A twenty percent cap does not put more young Americans in the water swimming a 53-second hundred butterfly. It only shortens the list. If the bottleneck in American swimming lies in the cost of youth club swimming, access to pools, early specialisation and academic pressure, then an international recruiting cap solves none of those. It moves the problem from having many fast people to having fewer fast people. In a system I have followed for years, limits mostly redistribute scarcity rather than create abundance.

And the last point, perhaps the one that bothers me most: a debate about nationality is obscuring a debate about the model. American college sports is in a period of self-examination about what its system is, whom it serves, and whom it pays. A nationality cap is a simple answer, easy to state in one sentence, easy to put into a bill, easy to make people feel the problem has been handled. It handles none of the real problems.

On numbers that tell stories, and numbers that lie

Over years in this trade I built a rule: before every number, I ask whose pain it is telling, or whose shining moment.

Florida's sixty-three percent tells one story. The thirty-one-of-forty-two figure tells another. The below-twenty-percent figure from 2026 tells a third, and the three do not reconcile. When three numbers describe one phenomenon and refuse to agree, it usually means we are measuring three different quantities and calling them by one name.

I am not writing to conclude that the data is wrong, or that the bill is right. Leslie Lucas's analysis is a valuable contribution, exactly as its author called it, an intriguing conversation. Its value lies in putting a number on the table instead of letting feeling lead. In an era when every sports debate begins with feeling, someone sitting down to count roster pages is an act of discipline. I respect it.

But I must also say what any data writer should say: a methodologically transparent dataset can still be used to prop up a wrong conclusion. Counting by the hometown field is enough to describe a picture, but not enough to become the basis of a legally binding rule. The gap between the two is the gap athletes will fall into if someone decides to write law from a spreadsheet.

What I carry away after closing the spreadsheet

I closed the spreadsheet around three in the morning. Outside the window, Nagoya was quiet in the way big cities are quiet at that hour. In my notebook there are three names: Yavuz Omer Aga, Kaii Winkler, Cooper Lucas. Three young people, three different stories, all of them eligible to become examples in a debate that will never ask them.

If the twenty percent ceiling passes, college swimming will look different. If it does not pass, recruiting behaviour may already have changed. If the flow has been below the threshold for four years, things may be adjusting on their own without intervention. All three scenarios can be true at once, in different programs, in different states. That is the nature of a system of forty-two swim teams and thousands of people.

What we should do is not decide who stays. It is to define properly what we are counting, publish the data for anyone to check, and separate the question of training origin from the question of flag colour. A system confident enough to say it wants to create opportunities for its own young people should start by creating opportunities at home, not by withdrawing opportunities from people who came from elsewhere. Because if American pools do not have enough fast young swimmers, lowering the ceiling will not make them faster. It will only make America swim slower along with the rest of the world.

And the question I leave with you, after all these numbers: when you look at a list of twenty-four people and see fifteen lines pointing to hometowns abroad, are you seeing a passport problem, or a problem about a country that has stopped raising fast swimmers in its own neighbourhoods?