Skip to content

Commit

Permalink
Add forecast explainer
Browse files Browse the repository at this point in the history
  • Loading branch information
cmheisel committed Nov 23, 2016
1 parent 6ab96c6 commit 18e6f91
Showing 1 changed file with 16 additions and 0 deletions.
16 changes: 16 additions & 0 deletions dashboard/templates/forecast.html
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,22 @@ <h3>What if...</h3>
</form>
{% endfor %}
{% endif %}

<div class="col-md-12" style="padding-left: 0px">
<h3>How do these forecasts work?</h3>
<p>This is a <a href="http://focusedobjective.com/monte-carlo-forecasting-exercises/">Monte Carlo forecast</a> based on the team's historical throughput.</p>
<p>If you want a great hands-on exercise try <a href="http://focusedobjective.com/monte-carlo-forecasting-exercises/">Troy Magennis' diceware exercise</a></p>
<p>Here's what happens:</p>
<ol>
<li>We start with the number of tickets remaining.</li>
<li>We simulate a hypothetical day by randomly picking one of the <strong>Completed that day</strong> values from the table on the right and subtract it from the number of tickets remaining.</li>
<li>We do step #2 over and over again, until the number of tickets remaining is <strong>0</strong>.</li>
<li>We record the <strong>number of simulated days it took to get to 0</strong>, that's our first simulation.</li>
<li>We repeat steps 1 through 4 <strong>10,000</strong> times, assuming the scope doesn't change.</li>
<li>Looking at those dates we pick the ones that represent when the project will end <strong>50%, 75% and 95%</strong> of the 10,000 times.</li>
<li>Then we round those 3 dates up to the <strong>Friday nearest them</strong> because this forecast is only accurate down to the week.</li>
</ol>
</div>
</div>

<div class="col-md-6" style="padding-right: 0px">
Expand Down

0 comments on commit 18e6f91

Please sign in to comment.