Skip to content

Estimation Table

Compare
Choose a tag to compare
@justinbmeyer justinbmeyer released this 16 Oct 19:46
· 283 commits to staging since this release

Estimation Table

image

This adds the start of an Estimation Table report to help understand your project's estimates and how percent complete is calculated. It's very rough right now, but look for enhancements shortly.

For each issue type it will show:

  • Estimated Days - The provided estimate converted to days. If no estimate was provided, the cell will be blank. If the estimate has changed, it will show Previous => Next values. You can also click the Estimated Days value and see a breakdown of how the value was calculated (more on this below).

  • Timed Days - How many business days are between the start and end of the issue. If the estimate doesn't have dates, the cell will be blank. If the timing has changed, it will show Previous => Next values.

  • Rolled Up Days - How many days of work are used to represent the issue, comparing children estimates, and the issue's estimates and the issues timing. More below on this calculation.

Estimated Days Tooltip

When you click an Estimated Days value a tooltip will appear with the following information:

image

This breaks down how the current and last (previous) Estimated Days values are calculated.

There are 3 equations:

Estimated Days = Adjusted Estimate / Points per day per work track

The estimate is the "Adjusted Estimate" divided by the amount of Points of each parallel work track the team can do.

Adjusted Estimate = Median Estimate * LOGNORMINV( Confidence )

The adjusted estimate is the "Median Estimate" inflated with a log-normal distribution using the Confidence.

Points per day per work track = Estimate Point Per Sprint (Velocity) / Parallel Work Tracks / Days Per Sprint

Points per day per work track is how many points of work the team can do on one "parallel work track".

This is the number of points a team gets done per sprint divided by the number of tracks they typically run in parallel divided by the number of days in a sprint.

Rolled Up Days

Rolled Up Days is calculated as follows:

  1. Set the issue's self timing data to Timed Days if it exists, then Estimated Days if it exists, then the average time for the issue type.
  2. Sum all the issue's children's roll up values as children's sum
  3. Use whichever is bigger issue's self timing or children's sum

The psuedo code might look like:

Rolled Up Days = MAX( CHILDRENS_SUM, TIMED_DAYS | ESTIMATED_DAYS | AVERAGE_FOR_TYPE )

What's Changed

Full Changelog: v0.5.0...v0.6.0