Does anyone have documentation on Agilecraft rounding for WSJF?

[Jeffrey Multach]

In looking at the WSJF numbers, rounding at two decimal positions appears to be performed.

In reviewing numbers, sometimes the value is rounded in the standard manner but sometimes it is rounded down and sometimes it is rounded up instead.

With no apparent logic.

[Jeffrey Multach · Accepted solution]

Just noticed (this is the cloud version)

If you enter the numbers from the Feature Estimation page it works properly.

If you enter the number from the Feature details pane, it does not.

If you enter the numbers from the Feature Estimation page and then go to the Feature detail pane, the numbers on the detail pane will be recalculated on display.

etc. (I am not going to figure out all the nuances of this).

[Derek Huether]

Jeffrey, I was able to reproduce the 3rd decimal number rounding issue

In one instance it appears to round and the other it appears to truncate.

I'm happy to submit it as a bug on your behalf. 

Everything looks good for the 1st and 2nd decimal number calculations.

[Jeffrey Multach]

Please submit a bug on my behalf.

Thank you.

[Derek Huether]

Jeffrey, I just wanted to let you know that your rounding issue is officially on the internal team's backlog to fix. I don't have a commitment date but know that you did find something.

The workaround is to have additional conversations, when epics have a WSJF score that is the same (out to the 3rd decimal point). Being the inputs are following a Fibonacci series, the output is to support a relative comparison, to support you in your prioritization endeavors. 

Best,

Derek

[Derek Huether]

@Jeffrey Multach the fix will appear in v10.103.0

[Derek Huether]

Jeffrey,
Could you please post a picture of the values in question, to allow us to further investigate?
Just post a picture here that looks something like this:

Thanks!

[Jeffrey Multach]

Sample attached.

[Jeffrey Multach]

Other values which end in .005 also do not round in a standard manner. For example:

0 0 1 40

0 0 5 8

[Derek Huether]

Jeffrey, I agree with you.
That does not look right. I did the same calculation and I get 3.63.
 

This could be a bug that has been resolved.

I'm on Cloud 10.75.

What are you running?