but has any major producing area (with at least 50 Gb or so in recoverable reserves) shown a production increase after hitting about 55% of Qt, based on P/Q versus Q?
 
Only if you do the linearization too early. For example
http://www.theoildrum.com/story/2005/9/30/21818/2120
You can't model UK production with a single logistic model.

Two cycles of discovery -> Two logistic models.