Private thesis pressure test
Pressure test the thesis before the meeting.
Before the next PM, risk, IC, or strategy meeting, bring one live thesis where the evidence has become harder to interpret. Crene maps what must be true, shows where the logic is fragile, and records what changes each week.
One live thesis
A view tied to a real decision
Daily reads, weekly freezes
Four model reads each day, five frozen weekly records, and a final decision record
Private engagement
Private, founder led, and scoped around one live thesis
Signs a thesis needs review
The evidence moved, but the thesis did not.
The team agrees on the conclusion but disagrees on why.
Several supporting assumptions depend on the same hidden premise.
The memo is stale, but the position or strategic view remains live.
The PM asks what changed, and the answer is scattered across notes, models, and calls.
Nobody can say which evidence would change the decision.
What the instrument reveals
A live thesis instrument, delivered through a private founder led review.
Thesis pressure map
Which assumptions are carrying the view, resisting it, or creating hidden concentration.
Hidden concentration
Where several supporting assumptions depend on the same premise.
Model disagreement
Where the headline consensus hides different causal explanations.
Weekly frozen records
What moved, what remained stable, and what requires human review.
Review conditions
Evidence that would strengthen, weaken, or reverse the view.
Final decision record
What the team believed, what changed, and why the decision moved or held.
Live example first read
What the live Warsh instrument surfaced.
Submitted thesis:
Will the Federal Reserve raise the target range for the federal funds rate at any point during 2026?
What Crene surfaced
The 15 point headline spread understates disagreement inside the assumption map. Thirty one decision weighted conditions have model spreads of at least 35 points.
Hidden dependency
Much of the supporting pressure is concentrated in the direct policy path rather than independent evidence across mechanisms.
Model disagreement
The headline consensus moved sharply, but the underlying model ensemble has not converged on why.
Unresolved question
Does the repricing reflect new primary evidence or correlated model interpretation across related conditions?
Review condition
Review the moving policy path conditions as one cluster and identify which constituent condition would most change the anchor if confirmed or reversed.
Decision relevance
The headline probability moved, but conviction should depend on whether the underlying evidence is independent and durable.