I have 2 databases I'm working with that hadn't had stats run in a long time. I noticed this on one of them last week, and at the time the shared pool usage was >80%. After running stats, it dropped down to <30%.
I've noticed that the shared pool usage on the other database is currently >80%, and stats haven't been run in a long time. Before I run stats, I was wondering:
Why did shared pool usage drop so much after running stats? Is it because it made the cost based optimizer that much more efficient that that much less shared pool was used, or is there something else, like perhaps it flushed the shared pool?
There is NO direct relationship between table/index statistics
and the amount of SGA being used.
You jumped to an unwarrented conclusion that there is a cause and effect relationship between the two independent observations.