I ran a fairly straightforward analysis of track and field performances across junior ages in different track and field competitions.
The raw analysis looks like this for international records. Above the line is male advantage, below the line is female advantage.
This pattern its repeated across national and state-level competitions. You can see that for almost all events at all ages, boys hold advantage over girls.
Where female advantage is detected, this is easily explained.
At 10 years old, girls grow ahead of boys, and catch up/overtake them briefly in running.
The female advantage in discus at 15-16 years old is because girls throw lighter implements.
The distance drop off as boys move to the 2 kg discus is obvious.
But actually, while these data are good for getting a handle on the magnitude of advantage, I came up with a slightly different question to ask of them.
With help from @johnarmstrong5, I came up with a null hypothesis: if there is no difference between boys and girls pre-puberty, the frequency of boys and girls "winning" should be around 50/50.
So I collapsed the performances as wins or losses. See below for international records, scored as wins for the boys above the line and wins for the girls below the line.
Let’s have a think what hormone categories looks like. And let’s assume that @neiltyson is considering a high/low T category. This has also been proposed by @AliceDreger
The proposal only works if you don’t deny evolution and sexual selection. Remarkably, there are academics who argue there is no biological basis for why males run faster than females. While it is plausible ongoing underinvestment in female sport means female athletes have not yet reached their full potential, it is frankly ridiculous to think this can explain the entirety of the performance gap.
See Sheree Bekker et al for more details on why, because one time, this one female figure skater won a medal, Usain Bolt should be allowed to race against females.
The proposal only makes sense if we recognise that the action of T on a body gives advantage in sport. This is by no means universally-accepted. Many humanities types argue T is not a key part of sports performance, citing males with low T and people registered as female with high T. Even though both phenomena are explicable by factors like illness, doping and male DSDs, still this argument persists.
See Veronica Ivy, Katrina Karkazis et al for why we should pretend that the stupidly high prevalence of weightlifting males with low T is not because they have just finished an off-period jacking up.
Why male advantage in sport is not a social construct: height.
Height is a key difference between males and females. What is nature v nurture? What does that mean for sport?
Bigger skeletons are most obviously driven by longer bone growth. Key bones like those in your thigh (“long bones”) grow from their end to get longer, making you taller.
The site of bone lengthening is called the “epiphyseal plate” or “growth plate”. Here, cells divide/enlarge, making new tissue that pushes the bone ends apart. This tissue calcifies and is replaced by bone, leading to lengthwise growth.
Let’s set a concrete example: the 10 second barrier (100m sprint).
Wiki - allowing for small errors - tells me that around 200 male sprinters have broken it. We know, of course, that no female sprinter has been close (Flo Jo record 10.49s).
For the following, I’m going to ignore the premise that humans might be close to biomechanical limits over a 100m sprint. It’s just an illustration.
If we follow world record progressions, we see trends (not just in sprinting, the graph below is from a swimming event).