There’s some controversy about the “reader context” reaction, but I think the reaction is factually correct and Hasan was mistaken. Here’s a screenshot from the FBI statistics; they naturally have the usual set of imperfections, for instance excluding the many cases where “the offender age, sex, race, and ethnicity are all reported as unknown,” but my sense is that they are the best data we have.

So while the numbers of white-on-white homicides and black-on-black homicides were virtually identical, the rate of black-on-black homicide was almost 5 times the rate of white-on-white homicide (since the black population in 2019 was indeed apparently about 1/5 that of the white population).

And of course it makes sense to compare rates (numbers divided by population) rather than raw numbers in this situation: If, for instance, someone told you that the city of Jonesville had 1/5 the population of the city of Smithville, but had the same number of homicides—saying nothing about race, but just transposing the numbers to geography—your reaction would probably and rightly be “Wow, Jonesville has a vastly more serious homicide problem than Smithville.