Whats interesting here is that Deffeyes defines peak oil as the point when exactly 50% of what Hubbert linearization for the world predicts total Qt to be. He says we crossed that 50% point on December 16, 2005. But others define 'peak Oil' as passing the maximum protracted daily production level - these two points could be vastly different in time - if technology is borrowing from the second half of the distribution, we could pass 50% Qt and keep increasing production for several years - it just means steeper decline rates on the back end. This is what I think is happening in any case. (And if Simmons is right about Ghawar, clearly this will be the case)- In fact we might have passed 50% of what mankind really pulls out many years ago...
I am not of the "Deffeyes" school here. As you said, I would define the peak as the maximum daily production rate (of all liquids). There will be a very long tail. And if 2006 passes and I have not seen an increase over the maximum seen in Spring of 2005, as noted by Stuart in various posts, then I believe that new supply will never offset depletion ever again--though we will probably be in CERA's undulating plateau for a few years with no spare capacity whatsoever. The difference between CERA and me here being about 16 years if 2006 turns out the way I think it will. Already HO has reported early year OPEC declines that may be partially offset by Haradh coming onstream soon. But, it's a waiting game now. How will Russia do this year? Mexico? Will something happen with Iran? Will Nigeria finally fall apart? Will Iraq get even worse? How much will Canada squeeze out of those tar sands? How moody is Hugo?
According to the latest IES update, it appears the updated world supply for December may have eclipsed May 2005 by about 100,000 (about 0.012%). A very small margin, and I would not be surprised if EIA has it the other way around. In any event, preliminary estimates from IEA are that Jan fell significantly from December and will again be below May 2005. Might be the "undulating plateau."

http://omrpublic.iea.org/

Ooops - misplaced a percentage point - make that 0.12%. Same difference.