Yes, that's what the middle case looked like to me, too. The "class of 06" looks like an underachiever compared to the CERA/HO estimate of about 2/mbd in the high case. However, I'm not sure Stuart is "playing fair" by offsetting Haradh III against Ghawar declines in the low case. He's not doing that elsewhere as far as I can see e.g. Buzzard :)
Well, Buzzard is a new field. Ghawar is not :-) It's complex to know how to treat the various rework projects.
My general intention, when I get to it, is to do a major project analysis for at least 2002-2004. Then I can compare to actual production changes in 02-05 and get a feel for how the "everything else" changes when you take out all the major projects.