Plain English guides

The full 10-section guide — 13F filings, conviction signals, copy-trading myths, and the honest limits of smart-money data. ~15 min read.

Plain English guide to SEC Form 13F. What's in it, when it drops, what it does and doesn't show.

Step-by-step: open any 13F on EDGAR and know what every field means. With real Berkshire examples.

New stake, exit, add, trim — the read-the-delta methodology. Compare share counts not dollar values, weigh by portfolio share, and know which moves are mechanical noise.

The free, step-by-step way to pull any manager's holdings from SEC EDGAR — and the limits of the data before you act on it.

The $100 million threshold, who counts as an institutional investment manager, the 45-day deadline, and who is exempt.

The term comes from Buffett's 1984 'Superinvestors of Graham-and-Doddsville' essay. What it meant then, what it means in the 13F-tracking era, and how HoldLens curates its cohort of 30.

Officers, directors, 10%+ owners — the SEC Section 16 definition, the Form 3/4/5 rules, and how legal insider trading differs from illegal.

13Fs are long-only by design. What that hides — shorts, swaps, hedges — and why a visible long never tells you the whole bet.

The hedge fund edge explained without jargon. Why 85% of managers have none — and what the 15% have in common.

Why every 13F is six weeks late by design — and how to use lagged data without getting burned.

Which Buffett principles are actually transferable to a retail account, and which depend on structural edges you don't have.

Why mechanically copying Buffett's 13F underperforms the underlying portfolio.

How to tell a real bet from index padding. The −100..+100 scale explained.

Why every hedge fund performance number you read is probably an overestimate — and how missing dead funds distorts 13F signals.

Three SEC filings, three signals. The difference between a quarterly portfolio snapshot, an activist disclosure, and a passive big-stake filing — and how to read each one.

Two SEC filings, two completely different signals. Form 4 is a 2-day insider receipt; 13F is a 45-day institutional portfolio snapshot. Speed, scope, signal strength, and when to read each one.

The complete required-vs-voluntary matrix — 13F quarterly longs, 13D/13G 5% tripwires, Form 4 insider trades — with triggers, deadlines, and what never gets disclosed at all.

Original research — April 2026. We backtested our own ConvictionScore over 221 ticker-quarter pairs across 4 quarters. Result: r = −0.12, no predictive signal. Top-decile BUYs underperformed SPY by 5 pts; bottom-decile SELLs beat it by 24 pts. Why, and what to use 13F data for instead.

Both return capital. One is flashier; the other is more tax-efficient. The honest tradeoffs on tax, flexibility, and long-term compounding.

Where the real buyback numbers live in SEC 10-K, 10-Q, and 8-K filings — plus how to spot debt-funded financial engineering and stock-based-comp distortions.

Plain-English guide: when an investor crosses 5%, which filing they pick reveals their intent. Activist (13D) means board fights. Passive (13G) means index hold.

What short interest actually measures, how days-to-cover is calculated, and why high short interest is BOTH a squeeze setup AND a smart-money signal.

What the STOCK Act of 2012 actually requires, why disclosures show ranges (not exact amounts), how late filings are penalized, and how to read the disclosures.

What overlap actually means, how to measure it, and why owning VOO + VTI + QQQ + SPY delivers far less diversification than you think.

The −100..+100 metric synthesizing Form 4 insider transactions across the tracked-superinvestor universe. Methodology, math, what the score does and doesn't mean.

8-K material events scored on a unified scale. Item taxonomy mapping, weight calibration, and how to read the live feed.

The three SEC filing types HoldLens reads together. Why no single filing tells the whole story, and how the trilogy synthesizes into one ConvictionScore.

The SEC's public filing database — the canonical source for every 10-K, 13F, Form 4, 8-K filed since 1993. How to use it, what to ignore, where it falls short.

The 9-character identifier that uniquely names every North American security. Structure, math, why tickers aren't enough for 13F reporting.

The definitive proxy statement — what to read, what to skip, how to vote informed. Read the 1-page summary in 10 minutes.

The SEC's quarterly Official List defining what shows up on 13F. ~17,000 securities in scope; shorts, bonds, foreign equities, most derivatives all out of scope.

6 months for reporting issuers; 12 months for non-reporting. Volume limits, manner-of-sale, and three practical reading angles for tracking insider supply.

The cross-fund synthesis — 13 hedge fund 13F filings read together. How the AI trade split four ways (chips, power, Microsoft-out, Microsoft-in), where the value investors added, the most concentrated bet, and the lone net seller. Links every per-investor deep-dive.

Berkshire's most active 13F quarter since 2024 — Delta Air Lines re-entry, Alphabet Class C add, Macy's + NYT adds, full exits of Visa + Mastercard + UnitedHealth + Aon, Chevron + Constellation trims. EDGAR-reconstructable.

Pershing Square's biggest tech rebalance in years — brand-new Microsoft position at 15.3% of the $13.7B portfolio, near-full Alphabet exit (combined GOOG + GOOGL trimmed ~95%), full Hilton exit, Amazon + Restaurant Brands adds, Brookfield + Howard Hughes trims. Eleven holdings throughout. EDGAR-reconstructable.

Duquesne Family Office diversification pivot — Natera conviction deepened to 18.1% of the $3.38B portfolio (single largest position), new positions in YPF (Argentina oil), Alcoa, ST-Microelectronics, BBB Foods (Mexico discount retail), NewAmsterdam Pharma; AMZN + GOOGL + TEVA + CPNG trimmed out of top 12. AUM contracted 25%; holdings rose to 68. EDGAR-reconstructable.

Appaloosa Management's clean China-to-US-tech rotation — Amazon doubled to #1 holding at 15.2% of the $5.93B portfolio, Alibaba trimmed from #1 to #6, Uber added back at 7.7%, Micron deepened to 9.5%, new SanDisk + Corning positions; JD, KraneShares China ETF, Qualcomm, American Airlines + Whirlpool trimmed out of top 15. Holdings concentrated from 38 to 31. EDGAR-reconstructable.

TCI Fund Management's most concentrated stance in years — Microsoft cut from 17.1% to just 2.6% (top-3 holding to residual), GE Aerospace deepened to 34.4% of the $39.2B portfolio, Visa to 23.5%, Moody's to 16.0%, Canadian Pacific to 9.3%; GE + Visa alone now 57.9% of the entire book. Alphabet added (combined GOOG + GOOGL to 8.3%). Just 9 holdings. EDGAR-reconstructable.

Munger's protégé cuts his 15-year Bank of America anchor 71% by share count (from ~28.6% to 4.6%) and opens four new positions: Moody's, MSCI, Tencent Music, and H&R Block. Crocs added 41%. The book stays concentrated — Alphabet ~45% combined, Pinduoduo, Berkshire, and East West Bancorp (a bank he kept) held unchanged. EDGAR-reconstructable.

Tiger Global rotates down the AI stack — Nvidia added to 9.2%, TSMC up 49% to 8.2%, Applied Materials up 85%, Lam Research held — while halving Microsoft (−54% to 4.1%) and trimming Take-Two, Apollo, Block, Reddit, ServiceNow, AppLovin. Meta added; MercadoLibre new. Alphabet stays #1 at 13.4%. 53 holdings. EDGAR-reconstructable.

Viking Global pushes Visa to the #1 holding (+59%), builds a quality-industrials and healthcare-tools cluster (Danaher, Fortive, Thermo Fisher +110%, new Air Products, Lennox +153%), and adds aggressively to Carvana (+162%) and Tesla (+47%) — while trimming Microsoft (−28%), Alphabet (−10%) and TSMC. New Apple position. A diversified 77-name long-short book. EDGAR-reconstructable.

Lone Pine bets on the physical infrastructure of AI — Vistra (top holding) and Talen Energy (+41%) for datacenter power, plus ASML, Carpenter Technology (+38%), and new Teradyne, Corning and MasTec for equipment and materials — while halving the TSMC foundry (−54%). AppLovin +88%. A concentrated 36-name growth book. EDGAR-reconstructable.

The Margin of Safety author makes Amazon his top holding (+47% to 12.7%), adds Alphabet (+9%) and Ferguson (+27%), and opens new positions in Aon, Visa and Teleflex — while trimming Willis Towers Watson and Liberty. A concentrated 22-name value book. EDGAR-reconstructable.

The most concentrated 13F we track — just three positions. Warrior Met Coal (39.9%) and Alpha Metallurgical (28.1%, added) make metallurgical coal 68% of the US book; Transocean (32%) trimmed 25% and Valaris exited. A pure deep-value commodity-cyclical bet. EDGAR-reconstructable.

The 'English Warren Buffett' trims almost every top US position at once — Marriott, Stryker, Waters, Visa, Alphabet, Pfizer, Interactive Brokers and ADP all reduced, with MSCI (−61%) and Rollins (−58%) near-exited. A uniform pullback across a 34-name quality book. EDGAR-reconstructable.

Maverick trims broadly — Carpenter Technology (−61%), MasTec (−65%), Live Nation (−73%), CRH, Somnigroup, TSMC and the top Amazon holding all reduced — while opening brand-new positions in Meta, Alphabet and bitcoin miner Hut 8. A rotation out of cyclicals into mega-cap tech. EDGAR-reconstructable.

Berkshire's 1988-89 KO purchase — $1.3B invested, today ~$28B position, untouched for 37 years. Trade mechanics + 13F-traceable accumulation timeline.

Berkshire's 2016-onward Apple accumulation — built into the firm's largest-ever equity position. Q1 2016 entry · Q2/Q3 2024 partial trim · still Berkshire's #1 holding. Full 13F-traceable timeline.

August 2011: $5B preferred + warrants for 700M BAC shares at $7.14 strike. The cleanest example of Buffett's 'structured private investment' template, alongside Goldman 2008 + GE 2008. ~$13B paper gain at 2017 warrant exercise; top-3 Berkshire holding through 2024.

Q1 2009: Appaloosa Management bought ~$2B of Bank of America + Citigroup + AIG common stock at distressed prices (BAC ~$3, C under $1). Returned ~$7B as the implicit TARP+CPP government floor proved Tepper's nationalization-tail-risk thesis right. ~$4B personal payday — second-highest single-year hedge-fund individual payout in modern history at the time, behind only Paulson 2007.

Scion Capital's 2005-2008 subprime CDS trade returned ~489% net. The canonical case study in why 13F-based research has structural blind spots.

Pershing Square's six-year activist short campaign — entered 2012, closed at a loss in 2018. What the public-record trail reveals.

September 16, 1992. Quantum Fund's $1B-in-a-day short of the British pound that forced the UK to exit the ERM. The FX/short/derivative gap to 13F.

1997 to 2023. Munger's 26-year hold + board seat — the canonical concentrated long-duration value position, fully traceable through SEC Form 4 + 13F + DEF 14A filings.

2013-2016 long-side activism — $3.6B position, public open letter to Tim Cook, ~$2B realized gain. The cleanest SEC-filing-trail activist case in modern markets.