Probabilistic Pre-Mortem Investing
Stress-test decisions across multiple scenarios before committing capital
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 6
- Confidence
- 97%
Probabilistic Pre-Mortem Investing replaces a single confident forecast with several plausible futures. Before implementing a trade or changing a portfolio, the team defines scenarios, examines how the proposed position would behave in each one, and actively debates assumptions and blind spots. The pre-mortem occurs before losses or emotional pressure distort judgment. Constructive conflict functions as a check against the biases embedded in command-and-control decision-making. The output is not a perfect prediction; it is a position whose risks, sensitivities, and failure conditions are understood in advance. Because unexpected events will still occur, the method aims to reduce surprise and improve preparedness rather than eliminate uncertainty.
Origin
Greg Peters describes using pre-mortems, debate, and probabilistic scenario analysis at PGIM Fixed Income before adjusting portfolios or making major investment decisions.
Core principles
- 01Conflict reveals risks that consensus can conceal
- 02Multiple scenarios are more useful than a single forecast
- 03Evaluate portfolio consequences before conditions change
- 04Disagreement is constructive when participants share a common purpose
How to run it
- 1
Frame the decision
State the proposed trade, allocation, or portfolio change clearly enough that its expected result and key assumptions can be challenged.
Pro tip Separate the investment thesis from the preferred implementation.
Watch out A vague proposal cannot be stress-tested meaningfully.
- 2
Build multiple scenarios
Describe several distinct futures, including the base case, adverse cases, and outcomes that contradict the prevailing market narrative.
Pro tip Use ranges and probabilities rather than one point estimate.
Watch out Do not limit scenarios to events already familiar from the recent past.
- 3
Map portfolio effects
Determine how the trade and the wider portfolio would behave under each scenario, including carry, liquidity, correlation, and downside effects.
Pro tip Look beyond the asset itself for cross-market consequences.
Watch out A trade that works alone may amplify risks elsewhere in the portfolio.
- 4
Create constructive conflict
Ask colleagues with complementary perspectives to attack the assumptions, identify blind spots, and offer alternative interpretations.
Pro tip Explicitly treat disagreement as assistance rather than disloyalty.
Watch out Personal or malicious conflict destroys the trust required for useful debate.
- 5
Pre-commit the response
Define what evidence would confirm or invalidate the thesis and what action should follow in each case.
Pro tip Record triggers while the team remains unemotional.
Watch out Do not improvise every response after markets become disorderly.
- 6
Revise and size
Change the position, hedge, or sizing to reflect the scenario analysis, then proceed only if the risk remains acceptable.
Pro tip Retain the scenario map for subsequent reviews.
Watch out Scenario analysis is wasted if it cannot alter the decision.
In the wild
A bond team considers extending duration after yields rise. It models persistent inflation, a soft landing, and a recession, then evaluates carry and price sensitivity in each case. Debate reveals that the recession hedge is valuable but the inflation scenario creates excessive concentration, so the team extends duration only partially and adds an inflation-sensitive offset.
→ The portfolio gains defensive duration without relying entirely on one rate forecast.
A market consensus assumes imminent central-bank cuts. The team compares that case with continued growth, sticky services inflation, and delayed easing. Micro-level company evidence contradicts the recession narrative, prompting the team to reduce its dependence on early cuts.
→ The decision becomes resilient to a wider range of policy outcomes.
Common mistakes
Treating the base case as certainty
A point estimate conceals the range of plausible outcomes and encourages positions that fail when reality deviates slightly.
Suppressing disagreement
Demanding consensus removes the checks and balances that expose individual bias and portfolio blind spots.
Running the analysis after the loss
A post-crisis review may teach lessons, but it cannot provide the unemotional preparation and response rules of a genuine pre-mortem.
Is it for you?
Best for
It is best for investors and leadership teams making consequential decisions under uncertainty.
Not ideal for
It is not ideal for trivial, easily reversible decisions where detailed scenario analysis would cost more than a mistake.
From the transcript
“I'm very much into the idea of pre mortem.”
“I'm a very big fan of looking at a probabilistic scenario based approach”
“I think looking at it through a multiple scenario lens is incredibly powerful.”
From the episode
US Treasuries to Bond Market Portfolio With PGIM