Michael Burry: Market Moves and Tactics
Michael Burry’s investing is associated with fundamental research, value, and a willingness to disagree with prevailing expectations. His mortgage-credit short became famous during the financial crisis, but copying a headline prediction or a delayed holdings report is different from reproducing the research behind a position.
The useful questions are concrete: what is the asset worth under conservative assumptions, why might the market disagree, and what could make the investment fail? LuxAlgo’s native charts and Quant, our coding agent, can support the chart-analysis and testing parts of that process. They complement company filings and instrument research rather than replacing them.
The mortgage short: research, timing, and staying power
In his 2011 Vanderbilt lecture, Burry described studying mortgage-securitization filings and examining weakening lending standards, interest-only payments, and borrowers’ dependence on refinancing. He expected payment resets and declining access to credit to expose those weaknesses.
According to that account, the first mortgage-short trades were agreed with Deutsche Bank in May 2005 and executed in early June. He bought credit default swaps referencing subordinated subprime mortgage-backed securities and ultimately worked with several dealer counterparties. The positions he retained began paying off in 2007.
It was not a frictionless wait. Burry described disputes over dealer valuations, investor withdrawal demands, and forced sales of other credit positions. The lesson is that a thesis needs financing and liquidity to survive the path to its outcome. Describing the episode only as a “2008 prediction” misses both the earlier research and the intervening pressures.
Understand the instrument behind the view
A credit default swap is a contract tied to specified credit risk. A protection buyer pays for exposure that can benefit from worsening credit conditions or contractually defined events. Reference obligations, settlement terms, premiums, and the counterparty all affect the economics.
Notional exposure is not the same as cash invested or profit. Likewise, a manager’s personal earnings, investors’ gains, and a fund’s lifetime return describe different things. Keep the period, capital base, fees, and calculation clear before comparing performance figures.
Video: Burry’s own account of the financial crisis
This lecture was recorded at Vanderbilt in April 2011. It provides historical context from Burry himself; its market and policy opinions belong to that time and should not be presented as a current crash forecast.
Value analysis begins with the business
Low valuation ratios can identify candidates for further work. They do not establish that a security is mispriced. Earnings may be temporarily inflated, book assets may be impaired, and cash flow may depend on underinvestment or working-capital movements that cannot continue.
| Research area | What to examine | What a simple ratio can miss |
|---|---|---|
| Valuation | Earnings, book value, cash flow, and the price paid | Cyclical peaks, one-time gains, or optimistic forecasts |
| Financial health | Debt maturities, interest expense, cash needs, and margins | Refinancing risk or cash that is restricted |
| Business quality | Customers, competitive position, costs, and industry change | Permanent deterioration mistaken for temporary weakness |
| Security structure | Share count, dilution, senior claims, and ownership rights | A surviving business whose old shareholders lose their claim |
Compare a company with peers, but also assess its value independently. A stock can be cheaper than an expensive sector and still be overpriced. Distinguish enterprise value from equity value and avoid subtracting cash twice when using an “ex-cash” multiple.
A margin-of-safety example
Suppose your assumptions produce an estimated equity value of $30 per share and the stock trades at $21. The price is 30% below that estimate: ($30 − $21) ÷ $30. The possible gain to the estimate is a different calculation, ($30 − $21) ÷ $21, or about 42.9%.
| Hypothetical valuation | Value per share | Difference from a $21 purchase |
|---|---|---|
| Adverse assumptions | $15 | −$6, or −28.6% |
| Base assumptions | $30 | +$9, or +42.9% |
| Favorable assumptions | $36 | +$15, or +71.4% |
These are illustrative estimates, not Burry’s positions or a forecast. A discount to an uncertain valuation is not guaranteed protection. Change the revenue, margin, reinvestment, financing, and share-count assumptions to see which ones drive the conclusion.
Hidden value and the risk of a value trap
Small and less-followed companies can reward detailed research, but limited coverage does not make them systematically cheap. Thin trading can widen spreads, make prices stale, and make it difficult to exit a position without affecting the market.
Historical examples such as Industrias Bachoco require dated filings and valuation assumptions. A later annual report cannot validate an earlier entry multiple or establish a manager’s realized return. For a poultry producer, investigate feed costs, commodity cycles, currency exposure, and cash requirements before interpreting a low earnings multiple.
Pillowtex is a particularly useful warning about calling a low-multiple thesis a success. Its July 30, 2003 announcement reported a Chapter 11 filing, an agreement to sell substantially all plants, equipment, and brands, and the closure of substantially all operations. Those facts do not reveal Burry’s exact exits or returns, but they show why an apparently cheap company requires continuing scrutiny.
A historical small-company index result is also not an individual manager’s track record. A return measured from 1972 to 2013 cannot describe Scion’s own performance over that entire period. Verify the actual strategy, sample, costs, and dates before using a statistic as evidence.
Read holdings disclosures carefully
Names such as Alibaba, JD.com, PDD, and American Coastal may appear in historical discussions of Scion. An undated share-count table should not be treated as its current portfolio or an instruction to buy the same securities.
The SEC’s Form 13F guidance explains the reporting framework and its limits. Filings generally arrive within 45 days after quarter-end. Short equity positions and written options are excluded, and the reported holdings do not reveal the complete portfolio, entry rationale, or subsequent trading.
For reported options, do not confuse the value of the underlying securities with the premium paid or maximum loss. Compare the same security and reporting period, account for amendments and corporate actions, and separate disclosed positions from speculation about the manager’s intent.
Risk control means planning for unfavorable paths
In the valuation example, buying 100 shares at $21 requires $2,100. A move to the assumed $15 adverse value loses $600 before costs. The estimate is not a loss ceiling: an unlevered common-stock investment can lose its full purchase value, and other instruments have different payoff and funding risks.
Review concentration across positions that share the same economic driver. Several companies in different sectors can still be exposed to the same interest rates, currency, funding conditions, or consumer slowdown. Conviction does not make those exposures independent.
Define which evidence would invalidate the thesis and how often to review it. Patience can be appropriate when the facts remain intact; holding indefinitely after they deteriorate is a different decision. A desired annual return is a target, not proof that the portfolio can deliver it.
For listed securities, stop orders have execution limitations. A stop-market order may execute beyond its trigger, while a stop-limit order may remain unfilled. Event-driven price jumps and poor liquidity can make a planned exit materially worse than expected.
Use LuxAlgo to make the technical work explicit
Create a native LuxAlgo chart workspace for the company and relevant comparisons. Review price structure, volume, timeframe, and session alongside your fundamental notes. Indicators can describe market behavior; they do not establish intrinsic value or predict a refinancing outcome.
For a testable technical idea, ask Quant to implement explicit entry, exit, sizing, and session rules. For example, define the exact completed-bar condition required before a counter-trend entry and the condition that closes the trade. Review the generated code rather than assuming it captures an investor’s discretionary process.
Inspect strategy settings and the trade log, include realistic costs, and evaluate a later period. Use only information that existed before each simulated decision. Selecting today’s surviving bargains and testing only their successful histories creates a misleading sample.
Native footprint-based tools require supported market data. LuxAlgo’s data guide distinguishes supported crypto and U.S. equity footprints from candle-only forex, commodity, and futures feeds. U.S. equity order-flow activity reflects Cboe EDGX, not consolidated market volume. None of these charts reconstructs the mortgage-credit contracts in Burry’s historical trade.
Use the Journal to record the thesis, decisions, and outcomes. Quant and chart research do not automatically execute broker trades or enforce protective orders.
A practical review before taking a position
Identify the source and date of every important fact. Write a valuation range, state what could make it wrong, and choose a position size that accounts for adverse outcomes and liquidity. Separate fundamental conviction from technical entry rules, and revisit both when new evidence arrives. Burry’s history is most useful when it encourages independent research rather than imitation of a famous prediction.
Read next