Published 2026-09-12. The instrument, described before you trust its readings.
Every fact on this page traces to a primary disclosure: SEC EDGAR filings (10-K, DEF 14A), JPX/TDnet filings in Japan, and the companies’ own releases. We do not screen on news coverage, social sentiment, price action, or anyone’s narrative about management. If it isn’t in a filing, it isn’t in the screen.
1. Structured risk-factor extraction. Each company’s filings are read into a fixed rubric — business risks, bitcoin-strategy risks, securities and capital structure, preferred-stock terms, governance and control — so companies are compared on the same axes instead of whichever risks happen to be vivid that quarter.
2. The frozen dilution screen. One rule, written down before it meets the evidence, testing whether insiders can dilute public holders in ways the filings already disclose. A warning triggers when primary disclosures alone let a reasonable reader compute either:
(a) Size — officer/insider-linked potential shares (options, RSUs, SARs, warrants held by directors, officers, and employees as a group) at or above 10% of fully diluted shares outstanding; or
(b) Structure — any adjustment, ratchet, or anti-dilution provision that automatically grows the insider pool as future financings expand the share count — i.e., the insider percentage is contractually protected while public holders dilute.
The earliest warning date is the earliest disclosure date at which the triggering fact was publicly computable — using only information available as of that date. No hindsight.
It is not a buy or sell recommendation. It is not a rating of the treasury strategy — note that the Metaplanet flag fires sixteen months before the company’s bitcoin pivot, while it was still a hotel company doing a turnaround recap. It is a governance screen: it tests compensation and dilution structure, not business quality.
Owned stock is excluded from the numerator everywhere. A founder holding 40% of the vote in owned Class B shares is a control fact, not a dilution fact — the screen measures potential future dilution from insider-linked instruments, and the distinction has held up in practice (see the discrimination test).