The thing is it only takes one person to kill you.
How would you begin to gather statistics and estimate risk for something like this, for example?
When the video from the Proud Boys’ founder made its way into my Google alerts, I didn’t know who he was at first. I was so distracted by the strangeness of someone making a takedown video about me that I didn’t think to look into who was actually behind it.
...
I took a closer look at the video. It had 24,000 views already. Some of the commenters were talking about the city I lived in. I was alone in my house, sitting in front of my computer, thinking, Oh no. Maybe this isn’t a funny story. Maybe this is a scary story.
The FBI tracks hate crimes. That could be a starting point. Various federal agencies provide protection and investigation of threats for political/judicial figures. It seems their analysis of what percentage are credible, and which ones are attempted, could provide good data.
We also have to view the risk as seen by the other commenter. They might not be outraged about this due to them not being a target by not becoming notorious on the internet and engaging with hate groups. This is largely the same as someone in a city saying the city is safe, but never going to the bad area. Is it a problem - yes. Is it one that directly affects them - no. Is it something they will take action on - it depends on their other priorities.
What do you mean by pseudoanonymity? You can't hide the fact that you're a woman on the internet if you're writing about the experience of being a woman, for instance. And the author here got her share of anti-semitic and anti-trans hate, despite not actually being either of those things.
As for hiding the fact, you don't have to write in first person, you could write about a fictitious person, or could change identifiable information and disclose that. It clearly doesn't matter what you say, people will assign what they assume anyway, ie trans or jewish.