# Audit which outcomes fall inside forecast ranges

Illustrative planning brief; no automatic product import.

Synthetic planning case: Four invented duration ranges have inclusive endpoints: A2–4 actual 3; B3–5 actual 6; C4–8 actual 5; D2–3 actual 3.

## Decision

Three of four outcomes fall inside the displayed ranges, or 75% observed sample coverage. Preserve widths and the missed case; four invented outcomes do not establish future coverage.

## Owned work

- Record the endpoint convention
  - Owner role: Evaluation owner
  - Acceptance evidence: Both ends are inclusive; D’s actual three is therefore inside.
- Retain misses and widths
  - Owner role: Reviewer
  - Acceptance evidence: B is outside and all four interval widths remain visible.
- Limit the confidence statement
  - Owner role: Planning lead
  - Acceptance evidence: The 75% figure is descriptive of this toy set and is not a calibrated next-task probability.

## Workflow

1. Record the endpoint convention. Check: Both ends are inclusive; D’s actual three is therefore inside.
2. Retain misses and widths. Check: B is outside and all four interval widths remain visible.
3. Limit the confidence statement. Check: The 75% figure is descriptive of this toy set and is not a calibrated next-task probability.

## Judgment

The ranges are supplied predictions without a verified probabilistic model. No confidence or prediction-interval calibration is established.


## Filled manual planning note

Three of four outcomes fall inside the displayed ranges, or 75% observed sample coverage. Preserve widths and the missed case; four invented outcomes do not establish future coverage. Inside count=3 of 4. Coverage=3/4=75%; mean width=(2+2+4+1)/4=2.25 days. Coverage alone ignores how wide the ranges are. The ranges are supplied predictions without a verified probabilistic model. No confidence or prediction-interval calibration is established.


## Workflow questions

### Could very wide ranges improve coverage?

They can contain more outcomes while becoming less useful; retain width and the decision purpose.

### Does 75% mean the next forecast is 75% reliable?

No. A tiny synthetic sample cannot establish that future probability.

## Product connection

Use the owned checks and downloaded brief to discuss this planning decision alongside your TeamBoostAI tasks. Confirm available fields, roles and account features separately. The example is manual; it does not calculate live analytics, create work or run an experiment in the product.

Confirm account availability before adopting this manual outline.

## Original worked case

Synthetic records, manual planning only. No account import or live analytics.

### Inspect the invented case records

Case | Inclusive range days | Actual days | Inside? | Width days
--- | --- | --- | --- | ---
A | 2–4 | 3 | Yes | 2
B | 3–5 | 6 | No | 2
C | 4–8 | 5 | Yes | 4
D | 2–3 | 3 | Yes, at endpoint | 1

### Reasoning

Inside count=3 of 4. Coverage=3/4=75%; mean width=(2+2+4+1)/4=2.25 days. Coverage alone ignores how wide the ranges are.

### Bounded result

Three of four outcomes fall inside the displayed ranges, or 75% observed sample coverage. Preserve widths and the missed case; four invented outcomes do not establish future coverage.

### Distinct decision

The guide checks interval membership and width rather than a percentile of completed-item history.

### Limits

The ranges are supplied predictions without a verified probabilistic model. No confidence or prediction-interval calibration is established.

### Definitions and method context

- Forecasting: Principles and Practice — accuracy — https://otexts.com/fpp3/accuracy.html — Genuine held-out forecast and point-error evaluation context. Binary scoring examples use their own explicit toy definitions; no real task model is validated.