Stories covered almost only by the pro-government outlets, or almost only by the critical ones. We count only who covered them, not how. A story lands here when its coverage is so one-sided that it would rarely happen by chance.
How we measure it
Now Archive Every blindspot we have found, each with the outlets' groups of the day it was found. The groups change with each monthly measurement.
How a story gets here Every outlet belongs to a group of its stance towards the government: pro-government, neutral, critical. For 54 outlets the group comes from the monthly measurement of headlines; for 26 outlets with few headlines it is the editors' estimate. We wait 24 hours from the first report, so the other side has time to write. We count which outlets covered it. If the critical (or the pro-government) outlets are so few that this would happen by chance less than 1 time in 20, it is a blindspot. We search the missing outlets' articles again: if we find coverage we had not linked, the story drops out. A model first judges whether the story says something political. Whatever slipped through, say a lottery draw, is hidden by the editors or, on their instruction, by a second model, always with a stated reason. When we say no outlet of a group has covered a story, we mean none of the outlets we are currently ingesting.
The methodology in full
Outlets by group 18 36 26 From the measurement of October 2026, except for 26 outlets with few headlines, where the group is the editors' estimate. It changes every month.
All outlets