Last updated on December 2nd, 2022 at 10:28 pm
Estimated reading time: 40 minutes
Ok, here goes. I’ve been meaning to write this for a while, but time and energy have not allowed it. But I’ve just been involved in some lively Twitter discussions, ultimately productive, so I’m going to try to share some thoughts around this issue of transgender females (male-to-female transitioning people) competing in women’s sport.
It’s a supercharged, emotive issue. No matter what one says, the “other side” will criticize it. That, to me, is why it has to be discussed from as many perspectives as possible, because so many conflicts require that the issue be held up and examined from as many sides as possible.
I’m going to try my best to offer the full 360 degrees on the issue of male to female (MTF) transgender athletes. So bear with me – my intention is to explain, not to convince. I have no specific ideology on this, just an opinion, and it’s an opinion that matters less than the logic of how we discuss this issue.
I want to offer my insights (for what they’re worth) and offer suggested next steps and predictions. This means I have to work around to the ‘dark side of the moon’ to see the issues, to ask questions and to play devil’s advocate. I cannot possibly do them all justice, but I hope it helps a little.
It’s a complex issue with many threads, so forgive me if I don’t link them all as elegantly as I hope to in the future. I’ve gone overboard with sub-headings, which I hope makes reading this easier (or skipping big sections possible), but here’s a first attempt at it.
[ribbon toplink=true]Problem definition: What needs to be addressed?[/ribbon]
It may seem absurd to you that a problem even needs to be defined here, but it does, for a few reasons. I want to try to explain those below, but if you want to just read the summary, it’s this:
Do transgender females who lower their testosterone levels into the female range, as required by current policy, continue to have performance advantages over biological women?
You may not realize it, but that relatively simple question contains within it about five important “loaded” concepts, which makes proper discussion of this issue difficult. So at the risk of labouring the point, we have to really unpack that question to try to understand the issues.
There is an argument here that we are wrong to even ask this question. That we should simply accept that men and women are different, and that crossing from one category to the other for the purposes of sport is inherently unfair, and can only disadvantage “biological women” and so should be prevented. I get that, and I can see the rationale for it. I just think it’s naive in this current climate, and I am also interested in the above question in an academic sense (which colours the way I engage with it, in the same way that women athletes will be influenced by their specific context on this matter, as will medical doctors, etc).
The reality, however, is that we must address that question above, because it has been given to us. We cannot simply dismiss it as irrelevant or the product of a PC-world (even though some elements, as I discuss below, threaten to become just that), because whatever happens here, there is going to be a significant legal challenge to sport’s transgender policies, and when it happens, the question above is going to be asked in courts, and thus must be debated.
So, point is, don’t shut the debate down just because you believe you’re theoretically right (this goes both ways, by the way)! You have your position, but appreciate that your position, whatever it is, is going to come under “attack”, so fortifying it might be a good idea!
And fortifying is best done with some evidence that turns your conceptual arguments into quantifiable ones – if CAS cases of the last decade, like Pistorius, Chand, Semenya, some doping violations, have shown us anything, it’s that the legal world will want to quantify the matter, not just hear impassioned conceptual arguments, however strong!
So let’s look at what two opposing sides will say, should it ever reach that point.
[ribbon toplink=true]The principle: Women’s sport is a protected class for good reasons. The boundary line can’t be broken[/ribbon]
Start with the fundamental principle. Why does women’s sport exist at all? Why do we crown a Fraser-Pryce AND a Bolt at the Olympic Games? The obvious answer is that it is a protected category that gives those who will not get the benefit of androgenization (testosterone) during puberty and adolescence to compete in sport.
Without it, the best 3000 athletes in track and field would be male, in pretty much every event (some events even more). What testosterone does to males in their teens does not guarantee that every male will be faster and stronger than every female (the best women are still way faster than most men, for instance), but it does give them such a huge head start and advantage that when we compare “like for like”, namely a group of humans competing for similar prestige at similar levels, men will totally outperform women.
In 100 athletes competing for university scholarships, the 100 best will be men. In 100 athletes racing for Olympic places and medals, 100 will be men. In 100 players vying for sports in their country’s national football team, 100 would be men. At any level, across any range, the top 100 (way more, actually, add a zero) an open competition between all humans would be won, without any exception, by those who benefit from testsosterone’s effects on muscle, skeleton, heart, blood and fat.
Women’s sport is thus protective. Protective over performance integrity for women, and possibly protective for safety in sports where that is a concern. For the former, the categories in Paralympic sports achieve the same outcome. Those who are more affected by a given disability are not required to compete against those less affected. For the latter, weight classes in combat sports do the same – heavyweights don’t fight against middleweights because it is both unfair and unsafe.
Given that women’s sport has historically been discriminated against (and indeed, women in wider society have been, too), it is worth recognizing that their insistence to defend their category has historical merit that may go even beyond fairness and safety.
But purely for physiological reasons, it makes sense to defend the line that separates the classes. Without it, or with a broken line, almost half the world’s population lose their valued space in sport.
Quite how that line is defended is the detail of the issue.
[ribbon toplink=true]Blue corner vs Red corner: What are the two opposing theories?[/ribbon]
But before trying to summarize the positions, let me try to answer the question as simply as possible, and try to get to the point of the controversy:
The answer is “Transgender MTF athletes LIKELY retain some degree of biological advantage even after lowering testosterone levels”. Studies in transwomen pretty universally show that lean mass and strength decline by about 5% after 12 months of lowering testosterone. The initial typical advantage in mass and strength that biological males enjoy over females is between 20% and 100% (depending on task), and the composite outcome of mass and strength may be even greater (think kinetic energy = mass x velocity squared).
In other words, the studies that exist for strength suggest that about one-tenth of the male-vs-female advantage is removed. That is nowhere close to what would ensure fairness. It means that “some” above means “most of”.
However, we don’t know this for sure, and the size of the theoretical advantage of being biologically male that remains even after lowering the testosterone levels is a contentious issue. That’s because the studies done to date weren’t in athletes, so you can argue that a person whose starting lean mass and strength are higher has more to lose. Of course, you can also argue the opposite, which makes this an unstable target of debate. There is also the issue of what happens after 12 months. And there is the measure of strength used – isokinetic testing or grip strength may be good “one dimensional” assessments, but may not necessarily tell us what happens to power in a complex task like say, a rugby tackle. The difference in these variables may be even larger than in simple actions, but so too might the reduction of those differences with testosterone reduction.
What currently happens, policy-wise (see below for more, including exceptions to this regulation) is that a MTF athlete is allowed to compete only if they show that they’ve kept their T levels below a certain point (10 nmol/L for some sports, 5 nmol/L for others) for at least twelve months. This is aimed, in theory, at taking away the advantage enjoyed by biological males. But does it? Could some degree of advantage remain, particularly in sports where the benefit of testosterone isn’t removed by its reduction?
If this is the case, namely that some advantage remains even after lowering testosterone, then it would have massive implications for women’s sport, because it would theoretically allow a sub-elite male to compete as an elite female simply through declaration or self-identification, and compliance with the policy.
Why? Because in running events, if the normal male-vs-female performance difference is 10%, then a male who is say, 7% slower than the very best males (sub-elite in other words), but who loses only 3 to 4% after reassignment to the female category, would now be at the level of elite females. In strength based events, particularly involving the upper body, this starting difference is even larger – weight lifting world records are 25% to 35% better in men than women even when you correct for mass, which means that even 20% reductions in strength with T reduction would not remove the male advantage.
In effect, if the policy doesn’t remove the advantage, it could ‘flood’ the female sports division with new arrivals, who enter the group by virtue of a physiological advantage that is unavailable to those already in the group.
On other hand, proponents of allowing transgender MTF athletes to participate under the testosterone-lowering guidelines are saying that the performance ‘adjustment’ or impairment after reducing testosterone is large enough that this never happens. Their position is that a biological male will lose enough of their “androgen-derived advantages” and enter women’s sport at the same place they left men’s sport.
If this were true, then there’d be little to fret about, at least for performance, fairness and safety. It would mean that a 12 month period of compliance would be enough to ensure fairness, and that no biological male ever enters women’s sport with a theoretically insurmountable advantage. It is, according to its proponents, the current situation, as evidenced by the lack of dominant MTF elite athletes.
And so in summary (and simplification), these are the two models:
- Lowering T, as current policy requires, completely removes the male performance advantage, and so should allow participation of transgender MTF athletes without any opposition (blue corner). Many in this group have also argued that being compelled to lower testosterone creates a health problem for them, and should not be required (in varying degrees and with varying ‘fervour’, as we shall see)
- Lowering T, as current policy requires, only partly removes the male performance advantage, and therefore MTF athletes should not be allowed to compete because they will still have insurmountable advantages and the integrity of women’s sport may be destroyed (red corner). This would be particularly the case in sports where stature and size matter, but is also because once you’ve had the benefit of testosterone during development, it can never be fully taken away.
And OK, I’m editorializing a little there, but that’s the gist of it. And because that question splits the debate more or less into those two models, I believe one is compelled to ask how they might be evaluated and separated.
[ribbon toplink=true]A need for evidence: Is there compromise, and who should provide it?[/ribbon]
That’s where evidence comes in. And on Twitter, a number of you have said that even asking for the evidence is foolish, unnecessary, a waste of energy. I get what you’re saying, but I also maintain that when (not if) these policies are challenged, then being able to rely on a degree of evidence, rather than simply the concept of male vs female advantages is going to make a huge difference to how strong one side’s position is compared to the other’s.
Both sides, to some extent, need evidence. You can argue that the burden of proof lies with the MTF athletes who want to compete, saying that they should provide evidence that they have no additional or unfair advantage. But, the reality is that sports governing bodies create a policy, and that policy needs to be evidence based, so the obligation to provide support for the position cuts both ways.
It’s one for lawyers to grapple with, but if the Pistorius-carbon fiber blade issue and the DSD issue of the last decade has shown me one thing, it’s that those who implement regulation and policy will ultimately be asked to defend it. And when they are, the evidence will be crucial, because the legal system doesn’t want to rule on rhetoric alone! Thus, irrespective of whether one argues that transgender MTF athletes should or should not be allowed to compete as women, some evidence, logic and rationale is required.
First, in support of the blue corner, we know from the work of Joanna Harper (who I’ve interviewed on this site before) that when she transitioned from male to female, her performance dropped off by roughly the normal difference between men and women. In other words, she went from being a near-elite male, ranked say, in the 93rd percentile among men, to being ranked about the same, 93rd percentile among female. She has documented other cases showing similar reductions in other athletes, a case series that suggests that lowering testosterone impairs performance enough to ensure fairness to women.
This was for running performance, mind you. We don’t know how power events, or contact and team sports performance might be affected (much more difficult to measure). We know that the starting gap is considerably larger than in running events – 30% in weightlifting vs 10% in running, for instance. Throwing events and generally anything involving the upper body may be even greater. Punching power is 260% in men than women, for example.
Also, the Harper study was not a prospective, controlled study where training and other performance-affecting factors are controlled for (then again, how could they be perfectly controlled?).
However, this side does offer this evidence to diffuse some of the fear and anxiety – the effect of lowering testosterone is powerful enough that it would seem unlikely that there will be a tsunami of athletes who dominate everything from the 100m to weight lifting, at least any time soon. It is, however, not sufficient to rule out advantages that may exist.
[ribbon toplink=true]Absence of evidence and evidence of absence[/ribbon]
Speaking of that tsunami, the other “evidence”, for what it’s worth, offered by those in the blue corner, is that women’s sport has NOT been flooded or over-run by men who identify as women (MTF athletes, in other words). Their argument goes something like this:
If MTF had the huge advantage everyone claims, then Olympic and World level podiums would have been overrun by all those biological men who identify as women, because in the last few decades, there must have been hundreds of potential MTF champions. Yet none have won medals, so thus the advantage everyone fears is not real.
I find this line of evidence pretty weak, for a few reasons. The main one is that it assumes that the MTF athletes who have tried to enter women’s sport have had the necessary athletic potential or caliber to win, even after a 5% to 10% decline in performance. Remember, 10% is a huge gap, but a man who is 10% slower than other men is still a very good athlete.
Take the 100m event, for simplicity. The best men run 9.80s. The best women run 10.90s. But a man who is say, 7% slower than the Olympic finalists is still running 10.5s. That’s not “mediocre”, it’s very good. If that man suppressed his testosterone levels to compete as a woman, and lost 10% of his performance, he’d run 11.5s and not even qualify for the Games. He’d be way too slow. If however he lost only 5% of his “biological male advantage”, then he might make the Olympic final, but he still wouldn’t win a medal. For that, he’d need to lose only 3 to 4% of his original biological male ability.
So in order for an MTF athlete to reach the Olympic podium, with say, a 5% performance decline, he’d need to start out as a 10.36s runner, who’d end up running 10.88s. Or take a different event – an 800m runner who runs 1:51 as a man could end up running 1:57 as a woman, but only if they lost 5% of their performance thanks to reducing testosterone. At 10%, that man would need to be a 1:46 runner!
How many 10.36s and 1:46/1:51 male runners do you think have made the transition? I’d guess so few that this stat of “they’ve never won a medal and therefore they can’t have an advantage” is pretty meaningless – it says more about the rarity of the event than it does about a lack of advantage.
An advantage may still exist. In fact, if the performance of a biological male got worse by anything less than 10% after lowering testosterone, then the resultant MTF athlete would have a conceptually “unfair” advantage, and one that could be theoretically insurmountable. At 10% impaired, only truly elite men could win medals women’s sport. At a 7% loss of performance, you could be a medal winner even as a sub-elite male. But at 4% or less lost with lower testosterone, a sub-elite man could totally dominate women’s sport.
Problem is, we know neither the number nor caliber of the athletes who have transitioned, nor the drop-off in performance they experienced, so we can’t evaluate these possibilities. At all. This is why the range of performance losses caused by a reduction in testosterone matters hugely to this debate. It’s all good and well to say that “On average, men’s performances drop by 10% when they lower their testosterone“, but if some drop by 15% (they’d disappear totally), and others by 5% or less, then women’s sport will have an integrity problem at some point, as soon as a decent enough performing man transitions.
Put differently, it’s theoretically possible that a man who can run 10.3s identifies as a woman, reduces their testosterone levels, slows down by 3%, and runs 10.60s, winning by far. We simply don’t know – perhaps different degrees of androgenization because of differing sensitivity to testosterone mean you’ll have extreme responders and non-responders to the lowering of testosterone. But the fact that it’s never happened doesn’t prove anything. This, to re-iterate the point, is why data matters.
It’s a classic illustration of that adage “the absence of evidence is not evidence of absence [of an advantage, in this case]”.
Aside from that, it’s also largely irrelevant. We should not care only if they’re on Olympic and World Champion podiums. The impact is likely to be much larger at the levels below that, where 99% of sport is played. That’s where a 5% drop in performance may have even more meaningful impacts, because it’s “the steep part of the curve”, where a mediocre biological male might leap into representative level, winning local events, or places on teams that compete at the next level up. Then it all comes down to the philosophy of how people value sport at participation and representative level, rather than elite levels, and it gets very messy indeed.
[ribbon toplink=true]The permanence of testosterone advantages – sports where the advantage never disappears[/ribbon]
And then finally, it’s all good and well to have that discussion for running or cycling where one can attempt to argue that the advantages will disappeare once testosterone is lowered.
But we also have a subset of sports where the advantage will never disappear. This is particularly true where anthropometry – think stature/height, limb length etc – are crucial for sports performance. Lowering testosterone may reduce hemoglobin, muscle mass, strength, power and cardiovascular capacity, and it may cause fat mass to rise, but it’s not changing the skeleton, and it arguably isn’t undoing a body type and much of the size/bulk created in part by testosterone.
In some of these sports (contact sports, specifically), there is also a huge welfare issue, and so for that reason, the transgender MTF athlete poses particular concern for sports like boxing, MMA, rugby, AFL, even basketball, netball and handball.
Quite how sports sort through this issue, I don’t know. Rugby, for instance, will need to be especially vigilant, because this is a situation likely to arise and may create welfare risks to other players. In one sense, this might make it ‘simpler’, because you can ‘discriminate’ (legally) if there are reasonable grounds to, and the protection of all players may be one such reason.
But then, if I gaze into my crystal ball, a legal challenge will say “Prove that smaller players tackling bigger players are at greater risk of injury”. Because you might think it obvious, but there’s no evidence for this (for example, scrumhalves, the smallest players on the field, don’t have the highest injury risks, and locks, the largest players, don’t have the lowest risk). So again, we see a sound conceptual argument, a good theory, but no hard facts to support it, and we’re back in that position again!
That’s a hugely complex issue, and does push one further towards caution, which is to say, exclusion, in this debate. I don’t know the way around this, but if sport is to fulfil its obligations to ensure fairness for its participants, then it has to. If that same sport has an obligation for safety, then this becomes even more significant and important.
[ribbon toplink=true]Short term conclusion, then onto the details for the interested![/ribbon]
That’s my short overview of the issue. I know I can’t do it justice, but I’m at 5000 words total in this piece, and I need to stop! My feeling is that while I’d love to find a compromise position that allows every MTF athlete to compete without questions over advantages inaccessible to others, that situation doesn’t currently exist. That’s why the questions must be asked, and it’s good to discuss this without the labeling and name-calling and nastiness and vitriol it can trigger.
I think the scale of the situation is not nearly as bad as people fear it may be for elite sports, because for many sports (running, swimming, cycling) I think the lowering of testosterone does affect performance to a large enough degree that we’re not talking about the entire men’s population, or even hundreds of men crossing over into elite women’s sport – the starting requirement is pretty stringent – you’ve got to start out as a pretty good male athlete to challenge elite women, even with smaller performance impairments.
But I respect that one can’t put a number to the problem’s size and say “XYZ is too many” in elite sport. In each case, one would be too many for a subset of athletes affected by any unfair advantage. I also think that the performance reduction may not be large enough to prevent the theoretical scenario I discussed above where some sub-elite men are able to win women’s events, and that individual variation and sporting variation will make it unfair in some instances.
That said, neither side has good evidence for their position. One side has sound theory, and a biological concept so obvious that it shouldn’t even need explaining (but I try to below, because some people believe in the physiological equivalent of a flat earth, so do read on if you’re interested).
However, this side also doesn’t have great evidence to turn that concept into policy that can withstand an extreme legal challenge. Is the biological concept of male vs female advantage enough, without evidence? I’d be more reassured with actual data to support it, maybe that’s just me!
What I’d love to see is the performance comparison from hundreds of athletes, at a range of levels, before and after they transition. Let’s see what the range of decline is with lowering testosterone, and let’s try to understand whether it’s reasonable to expect a typical drop of 10% for those sports separated by 10%, and 30% in the strength events, and whether some transwomen drop by much more, and some much less (which would be the problem scenario).
Ideally, you’d do this in a controlled manner, where you are reassured that training and other life factors (and let’s acknowledge – there are plenty in play in these cases) are not driving the changes more than lowering the testosterone. But this may be impossible, hence the need for hundreds if not thousands of data points (also not realistic, I know).
So that’s my bit. What follows are more detailed thoughts about the issue. If you have reader fatigue, stop now. But if you’re keen and interested, I have a few more specific things to say, specifically about why the issue is so loaded and difficult for many to process and discuss. I also want to share with you my fears about how people are unwittingly or deliberately trying to rewrite biology to eliminate testosterone.
So do read on. Or don’t. Thanks for getting this far.
Ross
[ribbon toplink=true]What is the current situation, and how did we get there?[/ribbon]
When we look at this question:
Do transgender females who lower their testosterone levels into the female range, as dictated by current policy, continue to have performance advantages over biological women?
There are a handful of important “between the lines” issues worth addressing. Let’s take a look at some of them.
First and foremost, there seems to me to be a misunderstanding around what the current situation is.
A few weeks ago, a rapper called Zuby wanted to make a tongue-in-cheek point about the issue, and so he identified as female and broke a women’s deadlift world record “without even trying” (his words). Notwithstanding issues around record ratification and validity, it got media coverage, and triggered substantial discussion about the transgender issue in sport. But, and here’s the first issue we encounter in this maze, what Zuby did doesn’t truly reflect the situation across most sports.
Why not? Because most sports have regulations that compel MTF athletes to at least lower their testosterone levels for 12 months before they are eligible to participate (note: some organization and some countries (Canada, for example) have no such regulation – they have no requirement to lower testosterone, which means that a person CAN effectively just self-declare and compete like Zuby did. This is crazy, and I fully support sport’s autonomy to regulate itself to prevent this from happening with regulation. It’s the ‘worst-case scenario’ for sport if this goes widespread)
The policy in place by most sports is modeled on one proposed by the IOC, which says that in order to be eligible to compete in women’s sport, the athlete has to provide a couple of things:
- A written and signed declaration that her gender identity is female
- Proof that her testosterone levels are below 10 nmol/L, and have been for at least 12 months prior to competition (the latest IOC guidelines have suggested 5 nmol/L, as was reported last year, but that is yet to come into effect).
- She has to then keep her serum testosterone levels below 10 nmol/L as long as she wants to compete in women’s sports.
What is not required is:
- Legal recognition of the person’s gender identity as the person’s sex
- Surgical anatomical changes
So, because Zuby did not lower his testosterone levels for 12 months prior to his deadlifting demonstration, his example doesn’t really illustrate anything about the transgender issue (unless of course, you’re talking about the transgender issue that wants to “delete” testosterone from biological reality, or of those organizations that have no testosterone reducing requirement, but we’ll get to that later).
The fact that Zuby’s demonstration got the coverage it did in the context of the transgender debate suggests to me that a lot of people think that the way it works currently is that you simply declare yourself a woman and then compete. This is not true (with the afore-mentioned exceptions), though there is also legitimate concern that it may yet become that way. The current situation, however, is not quite as “fear inducing” as that, because of that requirement to lower testosterone levels.
So, rather than showing the folly of the transgender issue, what Zuby did, basically, was to provide a conceptual example of how men outperform women, and expose the stupidity of people who say that women have no inherent biological performance disadvantages (yep, they’re out there). No more, no less. But, like Zuby, you already know this, even if you have a passing interesting in sport. The best men are, in running events, 10% to 14% faster than women.
If Usain Bolt runs 9.80s, then the women’s champion is running around 10.85s (10.7% slower). When the London Marathon for men is won in say, 2:03:37 in a few weeks’ time, we can expect a women’s time of 2:17:13 (11% slower). This is accepted, universally, as the difference between men and women.
[ribbon toplink=true]Meet the androgens – the purported source of the advantage, and the target of regulation[/ribbon]
It has also been universally been accepted that a key (though not sole) driver for this male-vs-female difference is the hormone testosterone, along with its “family” of androgenizing hormones.
“Androgen” literally means “male making”, and what these hormones do, at least from the perspective of sports performance, is to drive what are called secondary sex characteristics. They include lean mass, strength and power, reduced fat mass (thus power to weight ratio), stronger bones, larger hearts, increased hemoglobin mass, skeletal structure, and more than a few I’m neglecting to mention lest I write a thesis here.
There are other androgen effects, like deeper voices and facial hair (plus there are effects on primary sex characteristics, like the genitalia, which is where the argument starts entering DSD territory, but let’s leave that for now), but medals aren’t won for those. On the other hand, a combination of lean mass, strength, reduced fat, power to weight, oxygen carrying capacity, and cardiovascular advantages add up to a pretty substantial benefit.
Part of the compelling argument for how testosterone drives performance differences between men and women comes when you look at what it does to boys and girls at the moment where it suddenly increases in boys. Take a look at the graph below, which shows the gap in % between the boys and girls running world records from 100m to 800m, aged 5 to 19.
It should be pretty obvious what you’re seeing – boys have a small advantage early, it then diminishes to the point of being non-existent by the age of 10 (in fact, the girl’s 200m WR (24.43s) is faster than the boy’s 200m WR (24.69s, and the 400m is not far off it.
Note also that almost by definition, WR holding children are precocious, and so what you see below is earlier development than in the normal population, where the same curve exists, just shifted right a little)

But then after the age of 10, things change. The gap suddenly starts appearing, and by the age of 14, the average gap for these events is 10%, and by 19, not a single one is less than 10%. That’s the performance gap that necessitates separation of men and women for sport, and happens because of physiological differences that begin to emerge at puberty.
Inferring cause and effect here is tricky, but the separation co-incides with puberty, where the Leydig cells in the testes to release 3 to 10 mg of testosterone daily. In girls, in the absence of testes, no such spike occurs, and testosterone instead comes from a mix of adrenal glands, ovaries and the conversion of precursors in the liver, kidney, muscle, fat and skin.
However, it never reaches anything like it does in men, whose testosterone levels will range between about 7.7 and 30 nmol/L, whereas in women the typical range is between 0 and 1.7 nmol/L (excluding, of course, certain controversial conditions, and others like PCOS).
This testosterone is then responsible for virilization, which includes the biology mentioned above that confer a performance advantage of at least 10% to men.
The exact impact made by testosterone on performance is trickier to dissect. Some academics and a few ‘rights campaigners’ have argued that testosterone is a poor predictor of performance among men and women, that it must have a minimal role because women with high testosterone don’t outperform women with low testosterone, and the same is true in men. This is a very lazy intellectual argument, but let’s park it for now and get to it later.
[ribbon toplink=true]Separation into categories and why self-identification simply cannot be allowed[/ribbon]
Back to the story – as a result of a clear and obvious difference between men’s and women’s performances, that we separate men and women into two categories. That way, we can crown BOTH a men’s 100m champion and a women’s 100m champion. We can recognize equality, despite also acknowledging difference.
What is implicit in this argument is the following:
Simply allowing self-identification as female is not good enough. We cannot allow a situation where all it takes is for someone to say “I’m identifying as female” in order to enter women’s sport. That would be analogous to what Zuby did (and what some organizations allow, and some other people are fighting for), and it would result in the disappearance of women from sport.
That’s because the performance gap of 10 to 14% you saw earlier is so large that literally thousands of men fit into it. If even 10% of these men, who aren’t quite elite, top-of-the-pile-male athletes, were allowed to enter women’s sport with no biological ‘alteration or intervention, then the very best women could quickly be relegated to outside the top 100, even the top 1000 in some instances.
[ribbon toplink=true]The misrepresented value of testosterone, and the drive to delete it from biological sporting importance[/ribbon]
That is a situation that is so obviously untenable it doesn’t even bear discussion. Yet, somehow, it happens, all too often. And my great fear in this debate, and as a consequence of the DSD debate, is that there’s going to be a “creep”, a gradual slide towards basically ‘deleting’ testosterone from biological sporting importance.
I read what some academics are saying, admittedly sometimes in the context of DSDs (a group different for a few important reasons – physiology, ethics, medicine etc), and I think “Come on, please be careful, because the moment you start to lump testosterone in with all the other genetic advantages, then women’s sport is in big, big trouble”.
Take for instance the common argument, that goes something like this “There’s no evidence that testosterone is responsible for performance differences, because men with high testosterone don’t always beat women with low testosterone. Therefore, why should we regulate testosterone?”
A few weeks ago, I read a piece by Katrina Karkazis, in which she implored us to “stop talking about testosterone”, and downplayed its role in the performance differences between men and women. She used the example of Usain Bolt to illustrate this – with his supposedly high testosterone levels, he can apparently “only” run a 2:07 800m, considerably slower than many women. This, I suppose, was offered up as evidence that testosterone is but one of many factors influencing performance, and so we shouldn’t dwell on it as a differentiator between men and women. Read it for yourself:

That’s such an intellectually lazy, foolish, or dishonest argument it’s hard to know where to start. First, it’s a gigantic straw man, because nowhere in any biological or physiological model for sports performance would having high T levels be enough to compensate for the cardiovascular, metabolic, biochemical, anthropometric, respiratory, neurological and musculoskeletal differences between a good 100m runner and an 800m runner. Clubbing seals is easy when you pick an imaginary seal.
But more to the point, what these academics miss is that in sport, there is an implicit assumption that the very best athletes have the necessary combination of those attributes (cardio, metabolic, neural etc) in order to be in their niche as elite athletes (I call this the efficient market hypothesis of elite sport), and then the difference is STILL 10%!
Why? Because Bolt is “androgenized”, whereas his female counterpart is not. In all other respects – enzymes, neural system, metabolism, respiration – the best men and best women will be incredibly homogenous, to the point of being indistinguishable from one another. But thanks to one differentiator, Bolt is 10% faster.
Finally, I don’t know if she chooses to ignore it or is just ignorant of it, but Bolt’s 2:07, assuming it’s true, is still faster than 99.x% of women, but only say 95% of men. It’s just such a stupid comparison or attempt to illustrate that testosterone doesn’t matter. If anything, it makes the point that it does!
[ribbon toplink=true]The dangerous path of equating testosterone to other genetic advantages – let’s all compete in one race, and let the “genetically fortunate” win everything. Bye bye women[/ribbon]
It’s also a really dangerous one, because once you embark on that theory that testosterone is just like other genetic advantages, then the end result, the final implication, is that it (and remember, “it” is a proxy for male vs female sex categories) should not be regulated.
After all, if you’re going to dismiss the value of testosterone by comparing its impact to that of height in basketball, or long arms in swimming (I saw this from an academic last week, not for the first time), then why should testosterone be regulated, but height and long arms ignored?
Answer: nothing should be regulated. So we end up in a position where people are basically arguing that “genetic advantages” should be left alone. And if that is the case, then the greatest genetic advantage out there would be “being male” or the Y-chromosome, or testosterone, and every woman in elite sport would disappear because 5000 or more men would beat the best woman.
But hey, they have genetic advantages, right? Lucky for men. Sorry, women, maybe next time.
They’ll say “of course the separation of men and women is important”, but then must not realize that they’re undermining the basis for that separation with their weak, lazy or dishonest argument. Unless they have a better suggestion for why men outperform women than to say “it’s complex” (we already know it’s complex. Nobody said it was easy.
The reason some academics make this mistake, by the way, might be because they interpret the regulatory focus on testosterone to mean that authorities are saying that testosterone is the only factor that matters. They haven’t realized that it’s simply the best proxy for “male performance” by virtue of its significant effects on physiology.
Authorities could have created regulation around the presence of the Y-Chromosome, or the SRY gene, or presence of testes, but each of those would come under even more attack (justifiably, I might add). So instead, we have testosterone, but then weak arguments like the one above, or the one about height in basketball being the same as testosterone in all sport are used to try to undermine what is a compromise position. The real issue here is androgenization – being male – and the biology of performance that this creates. Testosterone is just the symptom.
Ultimately, this “testosterone doesn’t dictate better athletic performance” (Karkazis’ words above) argument ends in the deletion of testosterone, which means there is no basis on which to separate men and women, unless this group of physiological crusaders have a better suggestion for transgender cases (Again, I exempt DSDs, because I more closely agree with them on that matter. I do, however, think that when experts speak on these DSD matters they need to be very clear that there are differences, because too many people are conflating transgender and DSD, and expert views that don’t clarify this pour petrol on fire)
[ribbon toplink=true]Testosterone and sports performance within men and women, and why no relationship doesn’t mean no effect[/ribbon]
The reason I’m labouring this point about testosterone “not mattering” is because I’ve heard it so often. People are, either purposely or by accident, diminishing testosterone as a source of difference between comparable men and women. I put ‘comparable’ in there because it’s a pretty important concept in this debate – Bolt is not comparable to an 800m runner, so it’s a stupid comparison to make. He is comparable to a 100m runner, so that’s where you should evaluate the different. DSDs, by the way, are a little different, and a lot trickier, for a few reasons, only one of which is that their androgen benefits are not as clear cut as those of men compared to women. But let’s leave that for another day.
Back to the transgender issue, picking up an issue I parked earlier, there are others who argue for a trivial role of testosterone. One is Rachel McKinnon, who often points out that testosterone is a poor predictor of performance within women, and within men. This too is a misleading representation, whose motives I don’t know.
Let’s take height and basketball to illustrate. I’m sure we will all agree that being tall is an advantage for basketball, right? But if we went to the NBA, or the NCAA champs on the go, and we lined every one of those players up in order from tallest to shortest, and then we made a list of best players to worst players, even accounting for playing position, those two lists would show little overlap. In other words, within the NBA, height will have a poor predictive value for performance. Is height then irrelevant?
Or take VO2max, which nobody should refute is crucial for endurance sports success. I’ve seen studies where the relationship between 10km or marathon running time and VO2max is as high as r = 0.9. A very good predictor. But if we took the elite men or women from the last decade of marathon majors, it would be impossible for you to tell me who was fastest based on VO2max. You might as well throw darts at a list of names to rank them. That is, VO2max is a poor predictor of performance within a group of elite runners. Does this make VO2max irrelevant?
Of course not. Why? This is the classic “range” problem – when you are looking for the predictive value of something, call it X, then its predictive power disappears when you look at a homogenous group who all have X – basically, you filter it out by selecting for it! In this case, other factors take over, and X starts to look less important.
In fact, X was THE CRUCIAL FACTOR that got you into the conversation. It’s like you have to have the ticket to get into the VIP lounge at the club, but once you’re in there, then your designer clothes and your dance moves make the difference to your social standing, because everyone in there already has the ticket!
The ticket, to labour the point, is testosterone. Or rather, to put my spin on it, the ticket is androgenization, testosterone’s effects. And when you try to find the effect of testosterone on performance, if you look at it in a group of similarly androgenized athletes (men, or women, for example), then you will obviously find little to no effect, because the way you’ve asked the question has already filtered it out! (this, ultimately, is what undid the IAAF’s attempts to find it in their study that we critiqued recently)
This is a classic and oft-repeated argument in this transgender debate. But here’s the thing – the absence of an effect of testosterone within men, or within women, does not mean that testosterone has no effect between men and women. Those advocating for transgender females inclusion in sport without any need to lower testosterone are being dishonest when they offer this argument, so beware.
[ribbon toplink=true]A simple target for a complex issue[/ribbon]
All of this brings us to the regulation. Sports authorities are in an unenviable position here. They are trying to be inclusive (generally, a good thing), when they may not actually need to, but in so doing, have opened themselves up to criticism over how they’re managing the attempted inclusion, because inclusion may create exclusion. The great irony is that they’re then accused of discrimination, but it’s actually their attempts to be inclusive, in an imperfect way, that allows this criticism. The other irony is that their desire to be inclusive could be landing them in a position of less inclusion than being exclusionary.
As I’ve said, we cannot simply allow self-identification. Therefore sports must impose some degree of adjustment on MTF athletes, for the sake of the majority. I appreciate that this line is “arbitrary” (we haven’t even discussed why the desired testosterone level is 10 or 5 nmol/L, and how that might be challenged). However, it’s an important line, and authorities have, understandably, tried to use testosterone to set it. With its bimodal distribution, and totally separate reference range, it’s the best candidate to achieve the desired degree of compromise.
In less accommodating moments, I find myself thinking, why should sport even accommodate those who switch? And I see many of you making this argument. But I would like to, and sport would like to, and what’s why we are where we are. Roads to hell, good intentions, and all that.
Testosterone is, for now, where the ‘game’ is at – transgender MTF athletes DO have lot lower it, currently, so the fear doctrine about men simply identifying and competing like Zuby is misplaced. But equally, whether that’s enough is uncertain, and we are in a time where people are chipping away at the male-female divide, and this has real consequences for the integrity of women’s sport.
It’s why we must have this discussion, and I hope it can be had in a civil way.
That’s all from me, my thoughts are now jumbled. Maybe in coming weeks, I’ll post excerpts from this one at a time. It’s a big issue, and I didn’t even scratch the surface.
Ross
PS: Thanks for reading, I’m sorry for the length, but this is a big issue and I wanted to touch on as many relevant issues as I could. Thanks as always for your support, and if you’re so inclined to leave a small donation, it would be enormously appreciated (but won’t have any bearing on me trying to write future articles!)
You must be logged in to post a comment.