Fraud-Resilient Position Sizing
Assume some deception will evade diligence and cap the damage through sizing
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 93%
Fraud-resilient sizing begins with the recognition that diligence has an asymmetric disadvantage: an investor devotes only part of the research process to verifying authenticity, while a committed fraudster devotes continuous effort to sustaining the deception. Because some fraud cannot be reliably detected in advance, the portfolio must absorb its possibility structurally. The investor identifies material claims that cannot be independently verified, estimates the loss if they prove false, and caps position size accordingly. Diversification across independent exposures prevents one deception from threatening the entire portfolio. Greater concentration remains possible when evidence, transparency, and conviction improve, but the decision explicitly includes catastrophic and non-analytical risks. The output is not immunity from fraud; it is a portfolio in which an undetected fraud is painful rather than fatal.
Origin
Extracted from Masters in Business during a discussion of Michael Price's Sunbeam investment, Bernie Madoff, and FTX.
Core principles
- 01Some fraud will survive reasonable diligence
- 02A fraudster's attention advantage limits analytical protection
- 03Size positions for risks that cannot be fully verified
- 04Increase diversification when conviction depends on unverifiable claims
- 05Consider everything that could go wrong before concentrating
How to run it
- 1
Separate merits from authenticity
Assess both whether the investment is attractive if reported facts are true and whether those facts can be trusted.
Pro tip List the few claims whose falsity would destroy the thesis.
Watch out A respected title, institution, or reputation does not verify the underlying facts.
- 2
Map unverifiable risks
Identify financial statements, custody arrangements, counterparties, or operational claims that cannot be independently confirmed.
Pro tip Prioritize claims controlled entirely by one person or organization.
Watch out Do not assume more analysis can eliminate every deception risk.
- 3
Calculate maximum damage
Estimate the portfolio loss if the investment is a complete fraud rather than merely a disappointing business.
Pro tip Use a near-total-loss assumption where recovery would be uncertain.
Watch out Conventional volatility models may omit fraud entirely.
- 4
Size for survival
Cap the position so a total loss remains tolerable for the portfolio and its liabilities.
Pro tip Set the cap before enthusiasm or social proof increases conviction.
Watch out High expected return does not compensate for an existential position size.
- 5
Diversify independent exposures
Spread capital across investments whose authenticity and failure modes are not controlled by the same people or infrastructure.
Pro tip Check for hidden common custodians, auditors, or counterparties.
Watch out Many positions do not provide diversification if they share one point of failure.
In the wild
An investor likes a private fund's reported performance but cannot independently verify custody and valuation. Rather than treating reputation as proof, the investor limits the allocation to an amount the portfolio could lose completely and diversifies across unrelated managers.
→ An undiscovered fraud would cause a contained loss instead of jeopardizing the full pool.
Common mistakes
Trying to outwork the fraudster
The deceiver may spend all of their time maintaining the fraud, while the analyst must divide attention among many aspects of the investment.
Using reputation as verification
Prestige and prior authority can suppress skepticism without independently proving that assets or returns exist.
Concentrating beyond verifiable conviction
A concentrated position magnifies hidden authenticity risk along with the expected upside.
Is it for you?
Best for
It is best for investors exposed to opaque companies, managers, counterparties, or private investments.
Not ideal for
It is not a substitute for ordinary diligence, verification, or legal safeguards.
From the transcript
“for every you know, one or two percent of your analytical time that you're trying to figure out if what you see is real the…”
“Fraud is one of them, but you can diversify away from it.”
“So that comes out in position sizing and conviction and just making sure that you're thinking about all the things that could go wrong if…”
From the episode
Ted Seides on the World’s Elite Money Managers
Ted Seides