Methodology

TickerDrift ranks NYSE and NASDAQ common stocks by company news article counts from the official Finnhub company-news API — a media-attention proxy, not Google search counts.

News Count index

For each ticker and completed UTC day, the index is the number of company-news articles returned by Finnhub for that symbol on that day. Days with no articles are stored as 0 so day-over-day pairs remain comparable.

Normalization method: finnhub_news_count_v1. Query configuration is stored as source=finnhub;endpoint=company-news;symbol=TICKER.

Day-over-day change

• absoluteChange = currentIndex − previousIndex
• percentageChange = ((currentIndex − previousIndex) / previousIndex) × 100 when previousIndex > 0
• If previousIndex is 0: percentageChange is null; UI shows “New activity” or “Insufficient baseline”
• Missing observations are never treated as zero (explicit zeros are stored)

Comparability and exclusions

Rankings include only observations that share a rankable normalization method (primarily finnhub_news_count_v1), the same query configuration, and two completed equivalent daily periods.

Excluded from percentage rankings:

• Missing or rejected observations
• Stale observations (beyond the freshness window)
• website_rescaled imports (if any)
• Zero-baseline pairs (no finite percentage)
• ETFs and inactive securities (unless configured)

Publish gate

Daily ingest starts at 23:59 America/New_York: first the NYSE/NASDAQ common-stock universe is refreshed from Nasdaq Trader, then news counts are ingested. The previous published board stays live until every eligible ticker for the new day is covered (automatic publish), or via an admin publish action. Incomplete drafts are not shown on the site.

Data providers

Primary: official Finnhub company-news API. See docs/FINNHUB_SETUP.md and docs/DATA_SOURCES.md. Optional providers remain: Google Trends (alpha), Wikipedia pageviews, Reddit, CSV import, and DEMO for UI testing. We do not scrape Google Trends, Yahoo Finance, Wikipedia HTML, or Reddit.

Disclaimer

News article counts do not necessarily indicate investment performance. This is not investment advice.

Tie policy

Ties break by higher absolute change, then higher current index, then ticker alphabetical order.