Chastity interest over time

How this data is collected

Each chart is a separate signal, collected independently and shown on its own axis — we deliberately do not blend them into a single index, because they measure different things on incomparable scales.

Community activity counts new discussion threads per month on a large, continuously active English chastity forum, by each thread's real creation date. It is a single comprehensive archive, so there is no per-batch ingest confound; the current (incomplete) month is excluded, and any known archival bulk-import months are quarantined from the series. It is our longest fully-dated signal, but it is one community — an activity proxy, not a population count.

Wikipedia pageviews come from the Wikimedia pageviews API (data begins July 2015), counting monthly views of the chastity-related English articles. Podcast coverage counts podcast episodes that mention chastity each month, discovered through Apple Podcasts search and counted from the shows' RSS feeds by each episode's publish date. It is a coverage proxy — how often the topic is discussed, not how many people listen — and because discovery matches show metadata and feeds expose only recent episodes, coverage is partial and the early years are undercounted. Search interest is uploaded from Google Trends CSV exports; Trends values are scaled 0–100 against each term's own peak, so they show shape, not absolute volume, and are not comparable between terms or against the other charts. Reddit subscribers are the current subscriber counts of the main chastity subreddits; continuous monthly tracking began September 2026, with a few sparse earlier points recovered from the Internet Archive. Market supply counts the devices and brands we actively track — a prospective, supply-side proxy that grows with our catalogue, not a historical market census.

Devices first seen in our catalog counts, for each year, the devices whose earliest known date falls in that year — the earliest store-publish date (Shopify/WooCommerce), Amazon "date first available", or, where no platform date exists, the earliest Internet Archive capture of the product page. Dates come mostly from authoritative platform or Amazon sources (91.4%) with the remainder (8.6%) archive estimates, and cover 82.4% of the real catalogue. Before counting we exclude non-product junk rows, collapse duplicate colour/size variants to a single earliest-dated record, and drop store-migration batches — a single date shared by more than ten products from one store, which marks a re-platforming or bulk scrape rather than real activity (74 such batches removed). The series starts in 2022, the earliest year the dates are defensible.

Read our full methodology →