I'm probably stating the obvious, but it's clear that what is needed is a systematic and comprehensive study of the decline rate of existing production.  The question is whether the empirical data necessary for this is readily available.  Based on my limited exposure to Peak Oil-related material (compared to many), such empirical data is readily available for places like Mexico, US, North Sea, etc.  How available is such data for other places - particularly for the Middle East and the FSU?  Is that data so shrouded in secrecy that serious analysts such as Chris Skrebowski are left with no choice but to guess blindly?
Here we have an example of the Hiding Hand of Human Nature revealing its deceptive ways. We like to present "Good News" to our peers. We hate to deliver or advertise bad news.
So when a new well is discovered, the projected production numbers and reserves are always on the high side. When production starts, it often beats "expectations" much to the delight of our investors. But who is going to tell investors and loan officers that production is declining faster than expected? Who is going to hint to them that you may not be able to pay dividends or loan obligations this year? You'd have to be a fool to blast them with the bad news.
Contrary to the popular myth, bad news (of the financial sort) does not travel fast or far.