Market internals, measured in the open.

Every trading day an automated pipeline refreshes our datasets and recomputes the whole catalog. As of today that catalog is:

• 73 live market tools, grouped by the question each answers: Internals, Risk, Tape, Stocks, Options, Valuation, Sectors, Sentiment, Global, and Crypto. From classic named signals — the Hindenburg Omen, Zweig Breadth Thrust, 90% up/down days — to the put/call ratio, VIX term structure, credit spreads, short interest and squeeze detection, options flow, intraday session gauges, and daily scanners with researched catalysts. See the full catalog.
• 125 economic indicators on the Economy hub — inflation, labor, housing, rates, money and credit, energy and the grid, the advertising economy — each with full history, an SPY overlay, and a dated reading of the latest release.
• 17 Manuals — deep references for the gauges people argue about (margin debt, the yield curve, CAPE, the VIX curve, M2, the Fed balance sheet…), with every famous episode computed from data rather than retold from memory.
• 8 Concepts — a typed knowledge graph of canonical definitions: exact formulas, worked examples, failure modes, and a dated live observation, each also served as machine-readable JSON for AI assistants.
• 7 base-rate signal studies — every time a classic trigger fired on SPY since 1993, what actually followed.
• 20 fund-manager portfolios — quarterly 13F holdings, parsed straight from SEC EDGAR.
• 17 research essays — Insights with a clear stance, live charts, and falsifiable claims — plus a research letter where the strongest findings ship as dated, scored calls.

The Market Monitor ties it together around three visitor jobs—monitor the market, find opportunities, and assess risk—while its Decision Desk preserves four horizon “clocks”: Tape (days), Swing (weeks), Cycle (months), and Valuation (years). Each carries a dated verdict, the base rates behind it, and the thresholds that would flip it.

The reference side of the site is organized by a deliberate division of labor, stated on every page it applies to:

• Concepts define. What a term means, exactly — the formula, a worked example, where it misleads.
• Manuals interpret. What a gauge has actually meant historically — every episode computed, forward returns graded, myths checked against the record.
• Tools measure. What the reading is right now, recomputed daily with its full history attached.

One subject can live on all three layers — the Zweig Breadth Thrust has a definition and a live gauge — and each page answers its own layer's question, then hands off rather than competing.

The core is our own database of daily OHLCV bars for roughly 5,500 US-listed symbols (SPY back to 1993, the broad universe back sixteen years), refreshed and verified after every close. Around it we compute directly from primary sources: SEC EDGAR XBRL filings (13Fs, share counts, platform revenue), FINRA (margin debt, short interest), FRED/ALFRED and the OECD (macro series with real release vintages), Cboe and OCC (volatility indexes and options clearing volume), the Treasury and Federal Reserve (yields, Z.1 balance sheets), Census, NAR, Zillow and Redfin (housing, attributed), the EIA (grid and energy), and Robert Shiller's own dataset (CAPE since 1881).

Two conventions run through all of it. Economic series are dated by release, not reference period — charts show prints when the market learned them. And where a series is a proxy or an estimate (a Redfin model, an ETF stand-in, a reconstructed archive), the page says so in plain sight rather than presenting it as ground truth.

Most market-internals tools are either locked behind a subscription or presented as a number with no explanation. We do the opposite: the readings are free, and each one leads with what it means in plain language and exactly how it was computed — including where we deviate from a classic specification and why that changes the reading.

Every study is edge-tested before it ships. When the test finds nothing, the page ships anyway and says so — several of our pages exist specifically to document that a famous indicator does not predict what its reputation claims, and more than one planned feature died in testing and is described honestly in the methodology of whatever replaced it.

The clearest way to explain what this site is: look at the words on its pages. Most finance media speaks the advice register — what to buy, when to sell, which ten stocks to watch. Our pages almost never use those words, and the difference is the product:

Those are decision-support concepts. Every page answers two questions — what is the market actually doing, and what followed similar conditions historically — and then hands the judgment back to you with the sample size attached.

The Trading Tools is built and maintained by Yuriy Matso, who reviews every published reading, study and essay. The daily pipeline is heavily automated — including AI-assisted research and computation, disclosed in how we use AI — but a person stands behind each page, and corrections are made on the page with the modified date bumped, never silently. Spot an error? info@thetrading.tools.

Start with the Market Monitor, today's Bubble Tracker reading, the Manuals, or read how it works and our methodology.

The Trading Tools is for informational and educational purposes only and does not provide investment advice. Nothing on this site is a recommendation to buy or sell any security. See the full disclaimer.