Probabilistic Forecasting

Distinctions

Is

Expressing forecasts as ranges with probabilities.
Quantifying uncertainty explicitly.
Using historical data to simulate outcomes.
Providing multiple confidence levels.
Enabling risk-aware decision-making.
Treating forecasts as distributions rather than points.
Communicating likelihood rather than certainty.
Updating forecasts as new data arrives.
Separating commitment from prediction.
Making variability part of the model.

Is Not

Single-point estimation.
Deterministic forecasting.
Commitment-based planning that treats estimates as guarantees.
Averaging without variability.
Best-case and worst-case guessing that is not grounded in data.
False Precision.
Ignoring historical data.
Binary thinking.
Static forecasts that are not updated with new info.
Hiding uncertainty.
Treating velocity as fixed capacity.
Date-driven certainty.

Boundary

Probabilistic forecasting expresses likely outcomes under uncertainty using historical data.
It does not eliminate uncertainty or turn a forecast into a promise.

Systems

Often uses Monte Carlo simulation and throughput-based forecasting.
Can be used to express likely Sprint outcomes at multiple confidence levels based on historic throughput.

Relationships

Relationship	Concept	Rationale
addresses	False Precision	It makes uncertainty explicit instead of presenting uncertain outcomes as exact values.
can reinforce	Capacity vs. Dedicated Capacity	Capacity forecasts create false confidence when shared capacity is interpreted as dedicated capacity.
can reinforce	Forecast vs. Commitment	Forecasts are often misread as promises when organizations fail to separate likelihood from commitment.

Perspectives

Stance	Who (Point)	What They See (View)	Optimize For	Insight	Blind Spots

Works Consulted