Skip to content

Expected outcome

The expected outcome is the specific result a forecast says should occur if it is right, written before the fact in terms precise enough to be checked. It is what separates a prediction from a direction of travel, and it is the field that makes scoring possible.

An expectation that cannot be shown to have failed cannot be shown to have held either.

Why it matters when the plan changes

Vague expectations are the most common reason judgements never get scored. Delivery will be challenging cannot resolve; the second milestone will slip by at least four weeks unless an owner is assigned by March can. Writing the second is harder, more exposed, and the only version that produces learning. Naming a milestone, a date and a condition fixes what would count as right before anyone knows the answer, the same discipline a proper scoring rule applies to a probabilistic claim.

The tension is that precision invites disagreement at the moment of writing, which is exactly when disagreement is useful and least welcome. A team that has to agree what would count as the forecast being wrong is having the argument early rather than at the review, where the value of the exercise actually sits. This is why forecasts pair the expected outcome with evidence, confidence and a named eval, each part built to stop the claim from softening once it is written down.

In practice

A forecast states that the integration will face people challenges. Nine months later everyone agrees it was right and nobody can say what it predicted. The same judgement written as a named milestone, a date and a condition would have resolved cleanly in either direction and taught the next forecast something.

Evidence

  • Forecasts are predictions that can later be compared with actual outcomes, which requires that the expected outcome was specified.

    Forecasting, Wikipedia (2026)

  • Proper scoring rules measure the accuracy of stated probabilistic predictions, which presupposes a resolvable claim.

    Brier score, Wikipedia (2026)

What it cannot tell you

An expected outcome tells you whether a forecast resolved as stated, not whether the reasoning behind it was sound or whether the underlying situation was well understood. A precisely written outcome can still be met by accident or missed for reasons unrelated to the original judgement, and the term is silent on which.

Questions

That two reasonable people would agree on whether it happened, judged against a named thing, a date and a threshold. Forecasting, Wikipedia (2026) notes that forecasts can later be compared with actual outcomes, which only holds if the expected outcome was written precisely enough to check.

Because vagueness is comfortable and precision is exposed. A vague expectation is never wrong, which protects the author and removes the entire value of having made the forecast. It also avoids the argument at the moment of writing, which is where the argument is cheapest.

Usually, because organisational forecasts are conditional by nature: this will slip unless an owner is assigned. Stating the condition makes the forecast actionable and makes resolution more complex, since the condition may have been met. Both effects are worth accepting.

That is recorded as what it is: the prediction held and the reasoning did not. The Brier score, Wikipedia (2026) shows that proper scoring rules measure the accuracy of stated probabilistic predictions, which only works when the prediction was resolvable in the first place.

Whoever makes the judgement, agreed with whoever will act on it. Agreement matters because the two parties frequently have different ideas of what would count as the forecast holding, and discovering that at the review is too late to be useful.