The issue isn't whether this number is precisely correct, it's that there is simply no concerted effort by the activists to pursue it. Just like how advocates for the invasion of Iraq were perfectly happy to allow Americans in 2002 to dramatically inflate their own sense of danger from terrorism, activists have an "ends justify the means" mentality of not being concerned with data unless it supports their cause, independent of whether said data reflects truth.
The article links to the Washington Post's database, but a separate project that is run by activists who are relatively data literate is this one:
https://mappingpoliceviolence.org/
I've dug through their dataset, and it's not bad. They tend to be a little too biased in the sense where they default to assuming unarmed or "allegedly armed", assuming that police reports of suspects having a weapon aren't trustworthy, which is understandable, but unscientific.
Also, no flag for "resisting arrest". An unspoken reality in these activist circles is that getting killed by police when NOT resisting arrest is a relatively rare event.
One of the stark things that stands out when looking through these databases is how the media's narrative bias has made non-Black victims of police shootings invisible. Felons with a violent history who are killed by police while resisting arrest get more media attention if their Black than do Latino victims of police shootings that were completely unjustified. The average HN reader looking at this comment can't name a single Latino person killed by police in the last 5 years. This is absurd and unjust, because there's no way that every Latino killed by cops is done so in a just manner, and without media pressure, the cops who killed them won't be brought to justice. Likewise for the single most horrific police killing I've seen (other than Floyd), Daniel Shaver. The cop who killed him got off with an early retirement and full pension.