StockDrifts LogoStockDrifts

Research & Methodology

Methodology, data analysis, and the thinking behind StockDrifts signals.

Original Research

Ten years of insider buying: what 113,117 Form 4 purchases returned, and where the returns came from

113,117 insider purchases from 2016 to 2025, scored on 12-month returns against a size-matched index: what copying insiders earned, and when it paid.

Original Research

When CEOs bet big on their own stock: liquid names paid, thin ones did not

19,978 CEO purchases, 2016 to 2025: how the size of the bet and the stock's liquidity changed the 12-month return, plus this quarter's big CEO bets.

Original Research

Cluster buying: what happened when several insiders bought the same stock

12-month returns by how many insiders bought the same stock within 30 days, 2016 to 2025, in liquid and thinly traded stocks, plus this month's clusters.

Original Research

The insider buy/sell ratio: the weeks insiders called the market, and the crashes they missed

The weekly ratio of US insiders buying to selling since 2016, and what the S&P 500 did in the year after insiders bought most. With this week's reading.

Original Research

The microcap trap: falling knives, insider selling and the rules that did not pay

Why insider buys in thinly traded stocks lagged, when buying the dip was a falling knife, what microcap insider selling meant, and five rules tested.

Original Research

Only 21% of active 13F funds beat the S&P 500 over five years

3,769 active 13F filers scored against the S&P 500 over five years, through the 2022 bear market. Size, concentration, holdings count and position sizing, with nine charts and every fund downloadable.

Original Research

Does a top 13F track record persist? About a third of it does

5,004 funds split into two 18-month halves: how often top-quintile records repeated, who fell to the bottom, five-year streaks versus chance, and what a three-year record predicted next.

Original Research

Q2 2026: where the top-rated 13F funds are overweight, and where they are not

Q2 2026 13F filings: 763 four- and five-star funds against 3,363 bottom-rated funds by industry, sector, market cap and single stock: semiconductors, tobacco, disk drives and power producers on one side, REITs and insurers on the other.

Original Research

What the best 13F funds are losing on, and what they are up eightfold on

24,746 positions at 763 top-rated funds measured against their estimated purchase price: the most widely held losers, the biggest unrealized gains, big bets below cost, and where the crowd paid more.

Original Research

What the best 13F funds owned before the AI crash, and whether they saw it coming

763 top-rated funds' June 30, 2026 books frozen through the July 29 semiconductor low: what they held, what they sold in Q2, and whether it helped.

Methodology

How StockDrifts rates 13F funds: the star rating, the universe and the limits of the data

The definitions behind every StockDrifts 13F study: how the one-to-five star rating is computed, why 12 quarters are required, who is excluded, and what Form 13F cannot show.

Methodology · Updated

Korea Insider Signal Tiers — Cross-Sectional Alpha from 56K+ DART Filings

170,720 share-count changes across 56,930 Korean DART filings, ranked by cross-sectional alpha into signal tiers. See which insider buys actually pay.

Methodology · Updated

Japan Insider Signal Tiers — 12-Month Forward Returns from 101K+ Filings

101,000+ Japanese large-shareholding filings scored on fixed 12-month forward returns. See which EDINET insider signals predict returns, and which don't.

Methodology

How We Rate Japan's Insider Investors — Our Methodology

How StockDrifts rates every EDINET filer 1–5 stars, using a composite of 5-year and 3-month returns across 50,000+ filings. The full methodology, in detail.