I'd suggest the following:

An open work item should never be planned for a completed iteration.
In general, you want the Planned For field to be as accurate as you can
make it. If you are not planning to fix it in the current iteration, it
should not be Planned for the current iteration. You could instead plan
it for the release (if you believe it will be fixed in the release), and
then plan it for a specific iteration once you believe you know what
iteration it will be fixed in. You should know well before the end of
an iteration whether a given work item will be completed in that
iteration (if not, that is a sign that the planning process needs to be
improved).

Cheers,
Geoff

On 12/5/2011 6:23 PM, miwalker wrote:

I had a question on best practices with the Planned For field in RTC
and how others are using it for defects.

When a defect is opened in an iteration what do you set for the
Planned For field? Do you set it to the current iteration? And what
if the defect is not going to be fixed in the current iteration? Do
you move it to the Defect Backlog until it's planned for a future
iteration?

Right now we open them initially with the current iteration as the
Planned For value. If they're fixed in the current iteration they're
closed with the Planned For = Current Iteration. If not fixed,
they're supposed to be moved to the Backlog but I'm finding they
aren't always getting moved and instead stay in that previous
iteration even after it's complete.