SOURCES & METHODOLOGY

Everything on this site comes from public, primary sources — filings, official feeds, and market data — compiled by our own in-house process. No third-party middlemen, no paid data services, no black boxes. Here's exactly where each piece comes from.

Our data sources

SEC EDGAR — 13F filings

Quarterly 13F-HR filings from the investors we track: Berkshire Hathaway, Scion Asset Management (Michael Burry), Duquesne (Stanley Druckenmiller), and Pershing Square (Bill Ackman). We pull holdings (name, CUSIP, value, shares) straight from each filing's information-table XML and compare against the prior quarter — only new positions become briefing ideas. 13Fs are quarterly and lagged: a filing shows what was held at quarter-end, not today. Search 13F filings on EDGAR.

House Clerk — financial disclosures

Periodic transaction reports (PTRs) from the House financial disclosure index. We read the official filing index and link each idea to the actual filing PDF — because we don't parse PDFs, we never invent trade details that aren't in the index. Disclosures are filed weeks after the trade: they are records, not real-time signals. House Clerk disclosures.

Senate eFD — financial disclosures

Senator transaction reports from the Senate's electronic financial disclosure search, covering the last 30 days of filings. Same honesty rule as the House: we link to the filing itself and invent nothing. Senate eFD search.

Yahoo Finance — market data

Index, crypto, FX, rates, and commodities quotes: S&P 500, Nasdaq, Dow, Bitcoin, Ethereum, EUR/USD, GBP/USD, USD/JPY, AUD/USD, the 10-year Treasury yield, gold, and WTI crude. Price and day-change are computed from the previous close. Refreshed with every briefing run. Yahoo Finance.

CFTC — Commitments of Traders

The weekly Commitments of Traders report, public CFTC data: leveraged-money net positioning in yen futures versus open interest, with the week-over-week change. Best effort — if the feed is down, the briefing says so instead of guessing. CFTC Commitments of Traders.

Central-bank calendars — curated in-house

Upcoming rate-decision dates for the Fed, ECB, Bank of Japan, and Bank of England. This is the one list we maintain by hand rather than scrape, refreshed quarterly, and it's labeled curated in every briefing so you know the difference.

ESPN — sports scores

The Scoreboard pulls live games, schedules, and final scores from ESPN's public scoreboard feeds (NFL, college football, MLB, NBA, NHL, MLS), refreshed about every minute. Scores are as reported by the feed — we don't alter them.

Our proprietary methodology

Collecting the data is only half the job. The other half is our own process for turning raw filings and quotes into a readable briefing — built in-house, owned by us, and kept as a trade secret. In plain terms, here's what it does:

1. Extract. Every morning our pipeline fetches each source directly — the House and Senate disclosure indexes, the latest 13F filings for our tracked managers, market quotes, and the CFTC report. Each source is checked independently, and the briefing names every source as ready or failed — a failed source is marked failed, never silently filled in.

2. Sort. Raw filings become research signals: new 13F positions versus the prior quarter, congressional trades linked to their filings, day-movers in FX computed from the quotes, positioning extremes from the COT data. Nothing is invented — if a detail wasn't in the source, it isn't in the briefing.

3. Compile. The signals are assembled into the Money Briefing and the FX Playbook with the same disclaimers you'd see on the page: every idea is labeled a research signal, never financial advice, and filing lag is stated outright.

The pipeline runs daily around 7am ET, so the briefing carries the prior day's close. The exact way we weight and rank signals is ours — that's the proprietary part. What it will never do is fake a number: invented figures are the one thing the method is built to make impossible.

Sources change; when we add or remove one, this page is updated. Questions about our data? Reach us through the Links page.