Probabilistic Conviction
Make decisions with conviction while preserving uncertainty and humility
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 94%
Probabilistic conviction resolves the tension between acting decisively and admitting uncertainty. The decision-maker identifies a range of plausible outcomes, assigns approximate probability weights, and chooses an action whose expected payoff and downside remain acceptable across scenarios. Conviction comes from the quality of the analysis and willingness to act, not from pretending one forecast is guaranteed. Humility remains operational: the investor states what is unknown, identifies evidence that would alter the weights, and updates as conditions change. Results are then evaluated in two dimensions—whether the process was sound and whether the realized outcome was favorable—because a high-probability decision can still lose. The mechanism produces decisions that are decisive but less vulnerable to narrative certainty, surprise, and hindsight bias.
Origin
Extracted from Masters in Business through Ted Seides's discussion of investors who use absolutes in a world of probabilities.
Core principles
- 01Treat investing as a probabilistic game
- 02Express confidence without claiming certainty
- 03Assign probability weights to plausible outcomes
- 04Judge decisions by process as well as results
- 05Update beliefs when evidence changes
How to run it
- 1
Frame the uncertain decision
Define the decision, time horizon, and outcomes that materially affect success or failure.
Pro tip Include an outcome that contradicts the prevailing consensus.
Watch out Do not collapse a range of outcomes into a binary prediction unless the decision truly is binary.
- 2
Weight the possibilities
Assign approximate probabilities to the plausible outcomes using available evidence and base rates.
Pro tip Use ranges when a precise percentage would imply false accuracy.
Watch out Confidence in a story is not evidence that its probability is high.
- 3
Identify update signals
Specify observations that would increase or decrease each probability before committing capital.
Pro tip Write these signals down to reduce motivated reasoning later.
Watch out Do not move the criteria after events unfold.
- 4
Choose a robust action
Select a position, structure, or decision that offers acceptable expected value without catastrophic exposure to a plausible adverse outcome.
Pro tip Adjust position size when uncertainty is unusually high.
Watch out A favorable expected outcome does not justify unlimited risk.
- 5
Review process and outcome separately
After the result, assess whether the probabilities and action were reasonable based on information available at the time.
Pro tip Record both lucky wins and unlucky losses.
Watch out Do not label every winner a good decision or every loser a bad one.
In the wild
An allocator sees persuasive cases for recession, soft landing, and renewed inflation. Instead of declaring one inevitable, the team weights all three, identifies data that would change the weights, and builds a portfolio with liquidity and diversified exposures that can tolerate each scenario.
→ The portfolio remains actionable without depending on one confident forecast.
Common mistakes
Confusing conviction with certainty
Capital-raising narratives often reward absolute language, but certainty conceals the distribution of possible outcomes.
Ignoring disconfirming evidence
A probabilistic estimate must change when relevant evidence changes; otherwise it is merely a fixed narrative with numbers attached.
Judging only by the result
A low-quality decision can win through luck, while a well-structured high-probability decision can lose.
Is it for you?
Best for
It is best for investors, allocators, and leaders making consequential decisions under uncertainty.
Not ideal for
It is not ideal for situations governed by deterministic rules or fully known facts.
From the transcript
“I've always viewed investing, as I think everyone properly should is as a probabilistic game.”
“The best ones can blend that conviction with humility.”
“Now you could put probability weights to it.”
From the episode
Ted Seides on the World’s Elite Money Managers
Ted Seides