China Political Censorship

• The index is the percent of tested political-criticism sites that are censored in China. A higher index means more blocking.
• Data source: OONI web_connectivity tests of political-criticism URLs (Citizen Lab category POLR).
• Index = censored measurements / usable measurements, over the last 3 days.
• A measurement counts as censored if it matches a known blocking signature (confirmed) or shows a censorship-like anomaly (suspected).
• A measurement is not usable if it failed because of a problem on the test probe.

OONI aggregation API (web_connectivity).

• Coverage depends on where and when OONI's volunteer probes run. The index reflects the measured sample, not every connection. The 3-day window smooths this.
• One category is a small part of a country's tested URLs. A category with few measurements in the window is noisier than the country-level index.
• China counts confirmed and suspected blocks. The signal is broader than a confirmed-only signal, but noisier: anomalies occur at a low baseline even on open networks.
• Categories come from the community-maintained Citizen Lab test lists. About 2% of measurements have no category; they are excluded from this market.

Prices are quoted in index points.

• 1 point = 1 % = $1.00.
• This keeps position value, fees, and PnL in dollars.
• The traded price moves in fractions of a point.

The traded price and the index are different numbers. The index is the measured value. Trades set the price, and the price moves freely around the index.

• Funding pulls the price back toward the index.
• The gap between the mark price and a 1-hour average of the index sets the funding rate.
• Price above index: longs pay shorts.
• Price below index: shorts pay longs.