by Will Boisvert
January 8, 2025
A recent study found evidence that male athletes who identify as women and undergo testosterone suppression have little advantage over female athletes. In reality, it’s just evidence that scientists will call fat, transgender couch potatoes “athletes” to serve a political agenda.
The controversy over trans-identified men in women’s sports has heated up, with senators raking NCAA chief Charlie Baker over the coals for allowing male athletes to play on women’s teams, and collegiate female volleyballers and LPGA golfers organizing to get rid of them.
There haven’t been many scientific studies investigating how much of an advantage trans-identified male athletes have over female athletes. But a paper published last April in the British Journal of Sports Medicine claimed to do just that, and came up with a startling conclusion: trans-identified men have little advantage over women, and even score worse on some tests of athletic ability.
The paper, “Strength, power and aerobic capacity of transgender athletes: a cross-sectional study,” by Blair Hamilton and six coauthors, had several results. The most eye-opening one is that trans-ID men can’t jump: their average “countermovement jump”—squatting, then jumping—had a height of 36.4 centimeters, substantially lower than the female athletes’ 40.7 centimeters. The study also found that trans-ID male athletes have “decreased lung function” compared to women, and worse cardiovascular fitness. The only measure in which trans-ID men clearly bested women was handgrip strength. The media trumpeted the study as a significant contribution, one that undermined transphobic assumptions that trans-identified men have an unfair edge over women in sports.
But a careful reading of the paper should prompt skepticism. Many of the metrics on which the women outperformed the trans-ID men were “relative” physiological indices that factored in the women’s smaller size and weight, like “relative average power to fat-free mass.” Those aren’t very important in competition: a woman might get more watts out of each kilogram of her mass in a laboratory, but her total power can still be overwhelmed by a trans-ID man with a lot more kilograms. The ladies beat out even the study’s cohort of regular male athletes on relative average power, but the men would still easily overpower them on the playing field.
But there’s a deeper question here, as with all such studies: is it really making an apples-to-apples comparison between the study groups? Or are there other factors, besides the trans-identified men’s testosterone suppression and cross-sex hormone treatments, that account for their poor performance relative to women?
When we delve into the data, we do see such factors—important characteristics that differ sharply between the different groups, and that call into question the paper’s premise that it was studying comparable cohorts of “athletes.” The stats suggest that, while the women in the study were indeed serious athletes, the trans-ID men were probably not athletes at all. Instead, they look very much like overweight, out-of-shape couch potatoes—and thus not representative of how real trans-identified male athletes would perform against women in competitive sports.
The study looked at four groups of athletes: men who identify as women (called “transgender women” in the study) and were taking drugs to reduce their testosterone to the very low levels normal for women; women who identify as men (“transgender men”) and were taking testosterone supplements; and comparison groups of regular female and male athletes (called “cisgender women” and “cisgender men”).
So let’s take a look at how physically comparable these groups are to each other. Here’s Table 1 from the paper, listing participant characteristics:
Two disparities that immediately jump out are the marked differences in body mass index and weight between the transgender women and the cisgender women (and cisgender men). At 22.5 and 23.6, respectively, the BMIs for cisgender women and men are well within the normal range, while the transgender women’s average BMI of 26.2 is in the overweight range. The transgender women are of course much heavier than the cisgender women, but they are also 7.5 kilograms—about 17 pounds—heavier than the cisgender men, even though they have the same average height. In short, the transgender women are chubbier than the other groups.
In fact, they are pretty fat, as we can see from another table showing the amount of fat and lean mass in their bodies:
The average portion of fat in the mass of transgender women is 31.5 percent. That’s a lot of fat; it qualifies as borderline obese. The fat gap between the transgender women and the cisgender women is as large as the gap between the cisgender women and cisgender men. In absolute terms, the transgender women are packing 10 to 11 more kilograms than the cisgender women and men—that’s 23 extra pounds of fat. With all that blubber weighing them down, it’s no wonder that the trans-identified men jump about as well as beached whales.
With all that blubber weighing them down, it’s no wonder that the study’s trans-identified men jumped about as well as beached whales.
And some of the transgender women are even fatter than those averages suggest. Individual BMIs weren’t published, but we know it because of the standard deviations. The standard deviation, denoted by the +- symbol in the tables, is a measure of how widely scattered the data are about the average (or “mean” in stats-speak.). A low SD indicates that the data points fall close to the mean, while a high SD says that they tend to be widely spread out. The standard deviations are quite large for the transgender women, meaning that some of the BMI’s and fat mass numbers are much lower than the mean—and some much higher.
The standard deviations—small for cisgender women and men, large for transgender women—also tell us something crucial about the training regimens of the different groups, and thus whether they are comparably serious athletes.
Serious, dedicated athletes tend to be similar in their training intensity. They exercise and practice very hard and often, and they watch what they eat. As a result, excess fat melts off and they all converge on a lean, muscular body type—which means their BMIs should be close to each other.
The cisgender women and men in the study show this pattern. The small standard deviation in their BMI stats indicates that each individual athlete has a BMI very close to the healthy mean. That suggests that they are all training hard so that their bodies have converged on the fit, lean, muscular ideal. In fact, their BMI means and standard deviations, 22.5 +- 1.9 for the cisgender women and 23.6 +- 1.8 for the cisgender men, are similar to those of elite athletes. For example, this study of elite female Norwegian athletes found an average BMI of 21.7 +- 2.4, compared to 24.0 +- 4.2 for the adult female population as a whole.
But the study’s transgender women don’t fit that statistical stereotype. Their BMI mean and standard deviation are 26.2 +- 6.0— an overweight mean with an extremely large standard deviation. Those numbers indicate that the individual transgender women in the study have a very wide range of BMIs and body types. Their BMI standard deviation is over three times larger than that of the cisgender women and men, and wider than we find in the general adult population, which includes grossly obese people and stick-thin anorexics. Whatever their training regimen, it hasn’t made the transgender women converge on a lean, fit body type.
To see how striking the disparity is, here are graphs of “normal distributions”—the bell curves also known as Gaussian distributions—with the same BMI means and standard deviations as the study groups. (The x axis is BMI values and the y axis is probability density, which equates to the percentage of the population that would have that BMI; thus, 0.1 = 10 percent, 0.2 = 20 percent.)
The svelte, sharply peaked, red bell curve represents the cisgender women’s BMIs, and the green curve the cisgender men’s. The broad, shallow, blue curve, the one that spreads out like a dollop of lard on a skillet, represents the transgender women’s BMIs.
To put numbers on those shapes, in a normal distribution about 68 percent of the individual values would lie within one standard deviation of the mean, and 95 percent within two standard deviations. For the cisgender women, some 68 percent of individual BMIs would be in the range from 20.6 to 24.4, and 95 percent would lie between 18.7 and 26.3. Thus, almost all the cisgender women would be in the normal range of BMIs, as athletes typically are.
Not so with the transgender women. They have an average BMI of 26.2, in the overweight range, and a standard deviation of 6.0. That means 68 percent of their BMI values in a normal distribution would fall inside a wide spread from 20.2 to 32.2, the latter number being well into the Class 1 obesity range. Some 32 percent would fall outside this range, some of them weighing into Class 2 and even flirting with Class 3, the major leagues of obesity. The blue curve is so broad that it’s bound to harbor many fatsos in the high-BMI right tail (and perhaps some 98-pound weaklings in the low-BMI tail on the left). What it’s very unlikely to represent is a predominance of fit athletes.
The paper does present individual data points on athletic performance, and here we see an anomalously large range for the transgender women, with bizarrely unathletic outliers. In the countermovement jump, for example, the lowest score for a transgender woman was 19 centimeters, 7.5 inches, while the lowest for a cisgender woman was 31 centimeters. What kind of athlete can’t jump eight inches off the ground?
So it seems clear that the trans-ID men in the study are not comparable to the female athletes in factors like fatness and BMI that have a major influence on athletic performance. That by itself makes the study deeply flawed. It’s comparing firm, crisp apples to soft, bloated oranges that are genetically engineered to ooze animal fat.
The authors don’t say much about these large disparities between the study groups, or how they are consistent with their premise that the groups have comparable levels of athleticism. They hint that some of the trans-ID men may have overdone it on the estrogen supplements, leading to high estradiol levels that might hormonally dispose them to pack on fat. Still, unless the trans-ID male athletes were exclusively composed of sumo wrestlers and jockeys, it’s hard to see how any athletic regimen worth the name could produce such a wide, fat range of body types.
An explanation for the conundrum may lie in the criteria for deciding who counted as an athlete and got into the study. The researchers say that the participants were all competitive athletes, with 36 percent competing in endurance sports, 26 percent in team sports and 38 percent in power sports. (They don’t say anything further about what the sports were, or how many of the different cohorts played each sport.) But none of them were competing at an international or national level, and, with an average age in their thirties, they were likely not in collegiate sports. At best, they were mainly competing in amateur sports clubs or the occasional 5K charity run.
But the criteria for inclusion in the study don’t even require that perfunctory level of competition:
“Participants and eligibility criteria. Participants were required to participate in competitive sports or undergo physical training at least three times per week. Following written consent, participants were asked to record their last four training sessions and self-rate their training intensity for each session on a scale of 1–10 (10=maximum intensity). The mean of the four sessions was recorded to represent the athletes’ training intensity.”
In other words, the “athletes” in the study may not have been athletes at all. Maybe they had to say that they were competitive athletes in some specific sport, but it doesn’t sound like they had to actually compete; they just had to regularly exercise (or say they did). Some of them might have been serious athletes who compete in organized sports and train vigorously. But, as far as I can tell, if you just rode a stationary bicycle three times a week—or at least told the researchers you did—you qualified as an “athlete.”
And the assessment of the subject’s “training intensity” seems to have been based on self-reporting, with no attempt to ensure that different groups had objectively similar workout regimens. The groups all had the same mean intensity score of 7, but self-reported intensity doesn’t seem like a great comparative metric. An obese man might consider a brisk one-mile walk an exhausting workout, while an athletic woman might consider a 5K run merely relaxing.
So why did the training regimens of the cisgender men and women in the study produce a narrow range of very athletic body types, while the transgender women’s training regimen produced a broad range of unathletic body types? A possible answer, one that’s fully consistent with the study’s methodology, is that the transgender women had no training regimen and weren’t athletes.
I emailed the study’s lead author, Blair Hamilton, now a researcher at Manchester Metropolitan University in England, and the corresponding author, Yannis P. Pitsiladis, a professor at Hong Kong Baptist University, with questions about the disparities between the study groups and how they came about. I noted the differences in BMI means and standard deviations. I asked if there were any objective recruitment standards to ensure that the different study groups were comparably athletic. I asked whether the cisgender categories were predominantly competitive athletes while the transgender women were predominantly not athletes but just self-reported exercisers, which is a possible outcome of the participation criteria. I asked if there were differences in the sports that different groups played: maybe the cisgenders were mainly doing track and field while the transgender women were disproportionately darts players. I wondered why the trans-ID men were so fat if they were supposed to be athletes, and whether that undermined the study’s conclusions that competitive trans-ID male athletes in general have little advantage over female athletes.
I haven’t heard back from them.
So the mystery remains. I won’t say that the researchers deliberately selected good female athletes and lousy trans-ID male athletes—or non-athletes—in order to confirm the transgenderist doctrine that there is little difference between them in sports. I have no direct evidence of that, just the strange fatness and BMI disparities. I will note, though, that Blair Hamilton, the lead author, is a man who identifies as a woman and plays goalie on women’s soccer teams . It’s conceivable that the researchers had an ideological stake in the outcome of the study.
In any case, it’s a bad study. The lax and subjective eligibility criteria do not ensure that the different study cohorts are comparable groups of athletes. In fact, it’s clear that they are not comparable because of the stark differences in the confounding variables of BMI and fat mass, which are very relevant to the study’s tests of athletic performance. It’s shocking that the paper passed peer review despite these glaring flaws.
All of this makes the study irrelevant to the question of whether it’s fair to let trans-identified men compete in women’s sports. Competitive sports attract serious, competitive athletes. Real transgender athletes aren’t overweight sporadic exercisers. They train hard, optimize their diets, and fine-tune their hormone levels.
For a realistic example of a serious trans-identified male athlete competing in women’s sports, we can look at Blaire Fleming , the male player on the San José State University women’s volleyball team. His presence provoked a rash of forfeitures when opposing teams refused to play against him. Here he is on the court:
Fleming is tall, lean and fit, with a BMI that’s undoubtedly in the low 20s. He’s noticeably more muscular than the women. He doesn’t have 23 pounds of superfluous flab weighing him down. He’s a serious athlete who trains hard, not a casual exerciser like the trans-ID men in the BJSM study could have been.
But Fleming’s success in women’s volleyball was due not just to his training regimen but to his inherent physical advantages as a man. At six-foot-one, he’s taller than all but four of the women on the nineteen-player Spartan squad. His muscular legs give him a great jump, and his muscular arms let him hit the ball hard. Those athletic gifts won him the regular starting spot as the outside hitter, which goes to a tall player who can jump high and smash the ball downward at the other team. It’s the main scoring position, so Fleming led the Spartans with 314 kills last season, far outdistancing Nayeli Ti’a’s 228 . Fleming would not have stood out on a men’s team, but his male physique made him the dominant player on a women’s team.
He could not have dominated had he been born female. Fleming’s six-foot-one height puts him in the 90th percentile of men . A woman in the 90th percentile of females would be five-foot-seven, shorter than all but four women on the Spartan team and shorter than all the regular starters. A female version of Fleming would have more fat, less muscle and less jumping prowess. Female Fleming would never have gotten the outside hitter spot; at best she would have been a benchwarmer, and likely would not have made the team at all. Some other woman would have won Fleming’s athletic scholarship, and Nayeli Ti’a, five-foot-eleven, would have been the starting outside hitter and led the Spartans in kills.
Most of the Spartan women are in the 99th percentile of females in height, wingspan and musculature, as is typical of Division I volleyballers. Yet, under an identical training regimen, a mediocre, 90th percentile, trans-ID male athlete easily outclassed them, despite years of testosterone suppression. That’s a true apples-to-apples comparison, and it demonstrates an inescapable reality of sports: in any serious competition among well-trained athletes, trans-identified men have inherent, unfair advantages over women.
The BJSM paper’s off-the-wall results are evidence of nothing more than the politicization of science and the shoddiness of peer review. But they also speak to the fundamental problem with the transgenderist project—it’s drive to replace reality with wishful thinking. The trans-identified men in the study may believe themselves to be women and athletes, but the reality is quite different.
(Note for statistics geeks: The bell-curve graphs above don’t come from data on individuals in the study, which weren’t published. I drew them using a computer program on the Desmos graphing calculator website , inputting the numbers from the study for BMI means and standard deviations. Also, the bell curves are normal distributions, meaning that the BMI values are distributed symmetrically about the mean. But real-world BMIs don’t quite follow a normal distribution. Instead, they follow a “log-normal distribution,” meaning that the logarithms of the BMI values are distributed symmetrically about the mean. The bell curves of real-world BMI values thus look slightly asymmetrical, with a bigger right-hand tail. Below is a graph of real-world BMI values for the general adult population, which I got from a different paper . I used the normal distribution because the study doesn’t have data on the logarithms of the BMI values, so I can’t graph it. In any case, the log-normal BMI curves are pretty close to the shape of the Gaussian normal distribution curves that I drew, so I think the latter give a good visual sense of the stark disparities between the BMI stats in the BJSM study.)