Justice As An Explicit Roll Of The Dice: How Many Men Are Rapists?

One of the things we heard constantly during the Brett Kavanaugh confirmation hearings was the old saw that false reports of rape amount to "2 to 8%". Some while back, Scott Greenfield some while ago published a fine review of the statistics of false reports, which found a disparity in terminology that left us with this:

True-ish: 35.3%
Definitively False: 5.9%
Inconclusive: 58.8%

But the thing I wanted to look at was the reverse of the "2 to 8%" figure (actually, 2-10% per Greenfield). If we're supposed to "believe" accusers on the basis of a low reported false accusation rate, why isn't it equally valid to decide guilt based on the number of genuinely innocent men? That is, shouldn't also "believe" men who claim innocence, on exactly the same evidentiary grounds? But how innocent? Let's try to figure out the percentage of the male population that are rapists. We start with Uniform Crime Report data. This counts reported rapes since 1960; the data tool provides data to 2014. So, using the "Legacy Rape" data*,

• Sum the number of reported rapes from 1960-2014: 4,002,079
• Divide by the 2014 population (318,857,056): 1.26%
• Multiply by 2 (because men are half the population):  2.51%
• Adjust for the RAINN estimated 69% non-reporting rate: 3.64%

The above assumes a 1:1 ratio of rapes to rapists, which contradicts the repeat offender theory of David Lisak and others. It also assumes that the number of rapists is not in part diminishing, because the oldest rapes in that count includes some that might have been committed by men now dead. This is important, because it artificially raises the male rapist percentage. Nonetheless, if we were to say that the number of false rape reports is a reason to "believe" the accusers, the even smaller percentage of male rapists in the general male population (versus the high end of false accusers at 8% or 10%) is an even stronger reason to acquit.

Of course, both of these are stupid; we don't roll the dice to determine guilt or innocence, but look at the individual circumstances of the charges and the presented evidence. N.b., I also didn't investigate actual convictions, because the Bureau of Justice Statistics website is presently broken.

---

*"Legacy rape" corresponds well to men raping women; the "Revised Rape" tallies include "penetration, no matter how slight, of the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the consent of the victim", which would obviously include prison rape.