I think that is simply the limit of the adbot AI. This is a political site, Joe is a politician, this is a good place to put an ad for Joe. Kinda like putting an ad for Catholic Charities USA on the sidebar of a news story on a priest being charged with child molestation. No human input on the ad placement.
But if you click on the ad just to find some way of telling those people you are praying for a slow painful death for Joe, you are telling the adbot it is doing a good job. It can’t distinguish between positive clicks and negative clicks – all clicks are good.
“I get to step outside there, touch a tree, get a dog and kiss my girl,” he said on his release. A smiling Dewey also told reporters he was not angry about the injustice, asking, “What good would it do me?”
Another one for the files of people who have had significant portions of their lives stolen from them yet harbor no ill will toward their oppressors. Why does this appear so universal? Selection bias? Sample size too small? We only hear about the forgiving ones?
The strange conclusion of this phenomenon is that we only need to be thoroughly and unjustifiably abused by the system to come to terms with it.
On the other hand, if a bunch of readers who hate Arpaio click on the ad, the pro-Arpaio group responsible for the ad has to pay Radley for each click. Think of this way: You have the power to take a small amount of money from Arpaio accolytes, and give it to Radley.