The Percentile Trap: How Misused Statistics Skew Fairness in Sports and Occupational Therapy

Lately, the conversation around gender in athletics, education, and therapy has gotten a lot more complicated. Questions about fairness, biology, and statistical interpretation are at the heart of some heated debates—whether it's about competitive advantage in sports or access to essential services like therapy. While these issues might seem separate, they both hinge on a common problem: how statistics are used (or misused) to justify decisions.

I recently came across a post from an athlete who was assigned male at birth but is now competing in women’s sports. This issue connects directly to occupational therapy because it highlights how statistical reasoning—whether in competition or clinical settings—can shape real-world opportunities. The athlete argued that their transition had led to an ‘equitable’ change in performance by comparing their high school results in men’s competitions to their current college results in women’s competitions. The key claim? Their relative standing within their gendered category had remained the same. At first glance, this sounds reasonable, but a closer look at the numbers reveals a serious flaw in logic.

The athlete based their argument on percentile rankings rather than raw performance data. The problem is that percentiles are relative measures, meaning they don’t tell us whether someone has gained or lost an advantage—only where they stand within their new comparison group. A shift in percentile ranking doesn’t prove that performance is now "fair;" it only shows that the athlete has moved from one statistical landscape to another. The missing factors are competition depth and biological differences that persist after hormone therapy. Men’s fields tend to have a deeper pool of competitors with more extreme high-end performances. So, an athlete shifting from a high percentile ranking in men’s competition to a similar ranking in women’s competition doesn’t prove fairness—it just shows that they are still outperforming most competitors in the new category. The real question should be: how does their raw performance compare to the top female athletes? If they are still performing at a level beyond what is typical for elite women, then the claim of an ‘equitable’ transition is weak.

Ironically, while this athlete misunderstood statistics in a way that supported their position, I once spoke to a parent who had a much clearer grasp of fairness. The parent was frustrated because their daughter was denied therapy services based on gendered scoring norms in a standardized test. Unlike the athlete, who used statistics selectively to argue fairness, this parent saw a real and unfair outcome in how the numbers were applied. The school’s Committee on Special Education (CSE) was supposed to determine eligibility based on functional need, but in practice, the CSE chair defaulted to rigid percentile cutoffs. Girls, who tended to score lower in a specific domain, were denied therapy at higher rates, while boys with identical raw scores were qualifying for services.

This reveals a deeper flaw: gender-specific percentiles don't necessarily reflect real-world needs. If a girl and a boy score the same but only the boy qualifies for therapy because his score places him lower in the male distribution, that’s not fairness—it’s an arbitrary statistical decision. These norms were designed to reflect performance differences between genders, not to decide who should or shouldn't receive services. But when statistical norms become gatekeepers for access, they stop being just descriptive tools and start becoming barriers to care.

The irony here is striking. The parent, who was simply advocating for their child, had an intuitive grasp of how statistics were shaping real-world fairness, while the athlete—despite being deeply invested in fairness—failed to see how their own statistical reasoning masked an advantage. This highlights a larger issue: people often trust numbers without questioning how they are used. It’s easy to cherry-pick stats to fit a narrative, but it takes real critical thinking to step back and ask whether the outcome makes sense.

For occupational therapists, this isn’t just an academic debate—it’s a daily ethical challenge. How we interpret statistics affects real people. If we lean too heavily on percentile rankings and gender norms, we risk making decisions that aren’t actually fair.

+++

What Should OTs Do?

1. Look beyond percentiles. Raw scores and real-world function are what truly matter. Percentiles can be useful, but they shouldn't dictate eligibility on their own.

2. Advocate for functional benchmarks. Therapy should be based on what a person can or cannot do, not just their statistical placement within a gendered norm set.

3. Challenge outdated test norms. If gender-based norms lead to unfair outcomes, OTs should push for change. We shouldn't let statistical tradition override clinical reasoning.

4. Develop statistical literacy. Numbers are powerful, but they can be misleading. Understanding how statistics shape outcomes is a professional responsibility.

At the heart of occupational therapy is the mission to help people engage meaningfully in their lives. That mission shouldn't be compromised by rigid statistical norms that don’t reflect real-world function. Whether in athletics or therapy, fairness isn’t just about following numbers—it’s about understanding them.