Considering that in real life, we have yet to find ANY planet that has life on it, it seems that any percentage chosen for a game is acceptable. It's impossible to be realistic with a number that pure fiction.
We could always use Star Trek as a reference point.
How many worlds did they visit that did NOT have native life? Maybe 10?
And those were generally humans with funny ears/noses/eyebrows who spoke English.
Anyway, the issue is how to reconcile NIL with the OTU, and in a way that does not burden the referee. Don McKinney has made that last point crystal clear with Marc, and makes sure to refresh Marc's memory regularly, in a manner that does NOT distract Marc from working on errata (ah life is so complicated).
My preference is to divide NIL worlds into categories.
1-2. NIL extinct for 10^2D years - reroll if 11 or 12
. If recent OR technic enough there may be ruins or artifacts. Example: Chamax.
3-4. NIL appear to be developing and may be fully "sophont" in 10^1D years. Example?
5. The Tionale Effect - NIL are there, but may be unknown or forgotten, and do not contribute to interstellar culture or economy and are not technic. There is no reason to visit them. Maybe they eat strangers. Who knows. They clearly don't care, and neither does the subsector. Example: Tionale, Malea, and the Gogs of Mirayn. 567-908 was here until the players let the cat out of the bag, so to speak.
6. Sophonts are known and there is some external interest or abuse. Examples: Craw and 567-908. Or, they are part of local interstellar society. Example: Amindii.
