The Dashboard.
484 stories live. Every chart on this page is clickable. A recorded statement is one post and a tested argument is one story built from many, so counts of the two aren't comparable.
How far it goes
How far each statement goes. Both tracks are in here, because this asks how bad, not who.
320 items. Click any part of it.
How the claims fared
Each claim read two ways, on the same scale. The gap between the two rows is the whole finding.
Read plainly
Read at its widest
154 claims carry a plain result and 153 carry a widest one, because a reading only gets a result where a check tests it. Click any part of it.
What it's about
The fault line or the investigation a story sits in.
182 items. Click any part of it.
Where
The country a card is set in. Plenty of carriers post about somewhere they don't live, so this follows the subject rather than the speaker.
166 items. Click any part of it.
Who it was aimed at
Never presented as a score against each other.
320 items. Click any part of it.
Where it sits
Fault lines, the weekly pass, and the long investigations.
484 items. Click any part of it.
When it was said
Recorded statements, by the month they were said.
320 items. Click any part of it.
The latest week
What crossed the line in the most recent pass.
Week 37367
367 items confirmed across the full pass, up from 314 the week before. Both weeks checked the same 1,034 accounts.
The rows count the 48 cards we carry from that week, chosen from the full pass.
Calls for violence
23 items →Calls for harm to come to a group, or celebrates harm that's already been done to them.
The Islamic State's spokesman told followers to keep attacking Jewish and Christian targets, and pointed at the dead as proof it works.
Who it targetsJewish people and Christians
He filmed himself on a New York street leading a chant about a massacre of Jews.
Who it targetsJewish people
A streamer with a million followers told them to identify every American Jew and plan to arrest them.
Who it targetsAmerican Jews
What these numbers don't say
These charts describe what we picked, not the world. We choose stories that are live and consequential, so the shape of the results is a fact about our queue. Some months run higher only because more of the weeks we've covered fall inside them, not because those months were worse. And neither track is ever presented as a score against the other.
Browse every cut · Why we believe our readings are correct · The full-instrument dashboard at Hobocode, with severity by narrative family, ignition watch, and the day by day weather.