The Presidential Fitness Test is a core memory for many Americans, especially if you’re a millenial or Gen X. For decades, we participated in an annual ritual that included, among other things, a mile run.
The standardized test was phased out in the 2010’s, but it’s been back in the news since the president announced last year that he wanted to bring it back. Nothing much has actually happened on that front, but it did lead me to do some research into the history of the test.
For our purposes, the most interesting thing about the test is that the reference scores were established by a research study in 1985. They represent one of the best representative measurements of what normal people can achieve when running.
A common hypothetical among runners is, “Do you think anyone can run X:YZ?” Insert a time and a distance, and people have varying opinions about what’s possible for normal people – given appropriate training and time. Some are optimistic (i.e. anyone can go sub-3:00) while others are lot more conservative (i.e. not everyone can BQ).
Analyzing race results gives us some indication of what’s common for runners, but this isn’t necessarily generalizable to everyone. People who run races choose to run races. So it’s not necessarily representative of the population as a whole.
Today I wanted to dig into the data behind the Presidential Fitness Test and think about – what’s an achievable goal for (most) regular people?
The History of the Presidential Fitness Test
The Presidential Fitness Test traces its roots back to the 1950’s. The United States was at war in Korea and President Eisenhower had concerns about the fitness of America’s youth to serve in the military. He set the federal government to work finding ways to promote and improve fitness.
This would lead to the development of the test and a series of national surveys. In 1958, 1965, and 1975, researchers collected data from across the country to measure how fit students were. These early tests included some of the familiar events from later on – but they did not include a mile run.
In 1980, the American Alliance for Health, Physical Education, Recreation and Dance developed a new protocol for testing students’ fitness – the Health Related Physical Fitness Test (HRPFT). This test included the mile run as an option for measuring young peoples’ aerobic capacity. Based on a sample of around 12,000 students, they also included norms to rate performances.
Based on this data, the 85th percentile (top 15%) of 16 year old boys could run 6:12 and the median time was 7:11. For girls, the 85th percentile was 8:13 and the median was 10:45.
At the same time, the federal government issued a report titled, “Promoting Health / Preventing Disease: Objectives for the Nation.” This set out certain targets for improving physical fitness, and a study was commissioned in 1984 to measure progress towards those goals.
This study, the 1984 National Children and Youth Fitness Study, surveyed a a sample of 8,800 students about their physical activity habits and then put them through the same battery of physical fitness tests. This study referenced the 1980 HRPFT and described it as having a “sample of convenience.” In other words, the sample wasn’t properly randomized to ensure that it was representative. So that 1980 baseline could be taken with a grain of salt.
The 1984 study didn’t produce a detailed table of results for the mile run, but it did offer an average time: 7:44 for boys and 11:03 for girls. It’s unclear whether this average is a mean or a median. If it’s a median, then these results are slightly slower than the results from the 1980 HRPFT. But if it’s a mean, they’re more or less the same.
The more interesting contribution from the 1984 study is the data on students physical activity levels. About 80% of students reported being active through some community organization outside of school and about 60% of students reported having appropriate physical activity year round. They also listed the most common forms of activity – and we’ll talk about that more later.
The 1985 National School Population Fitness Survey
That sets the stage for the next major iteration of the Presidential Fitness Test. In 1985, the President’s Council on Physical Fitness conducted the next (and final) major survey of America’s youth, building on previous studies in 1958, 1965, and 1975.
A panel of experts reviewed the components of the earlier versions of the test, and they recommended several changes. One of these was the inclusion of the mile run – which would become a permanent part of the Presidential Fitness Test from that point on.
Then, they went to great lengths to recruit a truly randomized sample of students across the United States. They broke the country down into regions, randomly selected school districts from within those regions, and then randomly selected classes and schools from those districts. Within each selected class, each student was expected to participate. The result was a sample of ~19,000 students that was broadly representative of the youth of the public school students in the United States.
A few small groups were excluded. First, it only included public schools, so private school students weren’t tested. Second, it only included students enrolled in gym class. So some individual students would have been excluded for physical reasons. But the sample is still broadly representative of able bodied American youth.
The students completed nine tests, including the mile, and the results were compiled into a set of benchmarks broken down into 5% increments.
The 85th percentile was 6:06 for boys and 8:23 for girls. This benchmark was the one used to award the Presidential Fitness Award.
The median time, which would make a student eligible or a participation award, was 7:04 for boys and 10:31 for girls.
It should be noted that there is a significant gap here between the boys and girls times. In part, this could be a product of the times. It was only a decade into the implementation of Title IX and this was only one year after the first women’s marathon in the Olympics (1984). That 1984 study on students’ physical activity habits also suggested girls were less likely to be engaged in sporting activities. So women’s sports – and running in particular – still had some catching up to do.
But even at the fastest end, there was a huge gap. The best boy in the sample ran 4:46 and the best girl in the sample ran 5:58. I checked the results for the 1985 Meet of Champions in New Jersey, and the best high school boys ran 4:12-4:15 that year. Meanwhile, the best girls ran 4:46-4:50. Those times are roughly equivalent according to age grading.
If you convert the girls and boys times to performance level percentiles through age grading, there’s a consistent gap between them. At each point in the distribution, the girls times are around 10 points lower than the boys.
If we assume this was simply a product of the times and a modern sample of girls would perform on par with the boys, we can simply adjust their times to equal the same performance level pecentiles as the boys.
In this graph, the middle line represents those adjusted times for the girls. Now, the times are much closer. The 85th percentile for girls would be 6:54 and the median would be 8:00.
What Does This Mean for What People Are Capable Of?
Consider a few things here. The median results from the 1985 study were about 7:00 for boys. After adjustment, this would be ~8:00 for girls. Based on the 1984 study, about half of students aren’t active year round – and the students who are active year round are much more likely to perform near the top of the distribution.
So while there’s certainly some genetic component to who performs well (and especially really well), I think it’s a reasonable assumption that anyone – given enough physical activity and training – would be capable of reaching that median (with the caveat that they’re able bodied and not don’t have some kind of physical impairment that prevents them from running).
Let’s start from the assumption that most boys are capable of running 7:00 or faster and most girls are capable of running 8:00 or faster. Let’s further assume that adults in their 20’s, given adequate activity and training, are at least equally capable of running these times. After all, the impacts of aging shouldn’t set in and slow people down until their late 30’s or 40’s.
What would this equate to in marathon times?
One method we could use is to convert the mile times to marathon times using the VDOT calculator. For men, that would be equivalent to about 3:50 and for women it would be about 4:15.
Another method we could use is to use the age graded performance level percentile to yield an equivalent marathon performance. Given a median PLP of 54.46%, that would equate to a 3:40 for men and about 4:00 for women.
For the marathon, especially, this assumes that a person commits to sufficient training. It’s a lot easier to fake your way through a mile than it is to fake your way through a marathon. But this data suggests that 3:40 to 3:50 is a reasonable base level goal for any young, able bodied man and 4:00 to 4:15 is reasonable for women.
This shouldn’t be interpreted as a ceiling or an argument that faster times – like a BQ – are out of reach for many people. I think that’s an open question, and it would take additional research beyond the data we have to say how broadly achievable that is. But it should be interpreted as a floor, and anyone who is out there with a 4:30 or 5:00 PR should feel confident that they can do better.
How Have Things Changed Since 1985?
Of course, it should be noted that this is based on data from 1985 – and that was 40 years ago. Things have changed since then and those changes could plausibly hold people back.
The 1984 study offers some really interesting insights into how active kids used to be. Based on that survey, about 80% of kids were involved in some kind of activity outside of school. A more recent survey, from 2022-23, found that less than 50% of students today participate in some kind of community or scholastic sports.
The 1984 study also listed out the most common forms of physical activity outside of school. The #1 activity for boys and the #2 activity for girls was bicycling. I doubt this was some kind of cycling race. It was likely kids using bikes to get around. That’s just not something you see much of anymore.
Another interesting finding is that two of the top five activities for girls were “disco / popular dance” and roller skating. Boys, on the other hand, had traditional sports (basketball, football, baseball) rounding out their top five activities. Both included swimming, and I’m going to assume that this meant swimming at the community pool or the lake – not being part of a swim team.
This is a reminder that leisure time – hanging out with your friends – used to include a lot more physical activity. Going to the disco or a skating rink is a physical activity. Siting around on Facetime is not.
A final big change is that children – and adults – are more likely to be overweight today than they were in the 1980’s. In the 1980’s, 5-10% of American adolescents were obese. By 2017-18, that number had risen to over 20%. This could be related to the changes in physical activity noted above. Or it could have other drivers. Regardless, it’s another thing that would prevent people from reaching their running potential – because additional weight necessarily slows people down.
So What’s the Bottom Line?
When you put all this together, it’s likely that a random sample of adolescents today would be slower than they were in 1985. And a random sample of people in their 20’s might likewise underperform. But the reasons for that – weight and activity – are easily addressed.
Based on this data from the Presidential Fitness Test, I think a reasonable baseline goal for an able bodied young person is around 3:45 for men and around 4:10 for women. Faster times are certainly possible for many (most?) runners with more rigorous training. But I’d wager that these times are well within reach given a) a person is a healthy weight, b) they have been reasonably active for at least a few years, and c) they follow a sensible training plan.
What do you think – is that setting the bar too low or is it about right? And do you have any fond memories of the Presidential Fitness Test from when you were a kid?