Wow, got busy for couple of days and the discussion just took off! The full text isn’t free, so I can’t link to it. I got it off Elsevier; they were promoting a new feature that allows patients (as opposed to professional researchers) get certain non-public-domain for only a $5 processing fee per article and they gave my account a $5 credit to check out the feature (maybe they think I’m a doctor who can spread word to patients? I do a fair bit of light medical journal research from time to time but I a layman). And it’s accompanied by all the usual language limiting distribution.
The numbers are what they are, the motivation for the study was to compare younger v. Older men because younger men had issues with condom breakage and slippage and generally complained about the things and older dudes were fine. Result that erect sizes are about the same (young guys a bit wider), though old guys had larger flaccid figures. However, the old guys all had ED, and the authors posited that larger flaccid may be due to loss of elasticity in tunica related to ED (noting that it could actually be caused by the cessation of erections).
That’s why I posted figures only for the 111 -member younger cohort with no ED and who achieved erections naturally.
I can quote from the article later re BP, but like I said, with regard to erect size, the author not only expressly mentions that the fat pad was maximally compressed , but also provides an illustration and criticizes other studies that measured fat pad thickness apart from the penis length. So that bit is petty certain.
So what of the results? The sample size is only 111, so the confidence interval would be on the wide side, but they nonetheless provide information. Why did they max out at 7.5? Well, in a random sample of 111 men, taken from an infinite population (for the purposes of the statistical calculation, same result if 80 million or 80 trillion), one only has a 63.38% chance of encountering at least one (that is, one or more) man that has a feature that exists in 1 of 111 men (0.9%) in the wider population. To get to 99%, you would need to sample more than 500. That being said, approximately 12.69% of people who have had 15 male partners would have run into a 1/111 (0.9%) penis, and that’s a lot of people. And a 0.9% penis, according to the study, is 7.5 or larger.
The closer you focus on the extreme ends of any normal distribution, the less accurate the extrapolations.