Though you are correct.

I believe the talk you are referring to from Rand is about Google.

The main index uses data that was crawled in the last 120 days (we may expand that number).

But it includes all the domains/pages that we crawled and all domains/pages that were linked to by pages that we crawled. So a page/domain may exist in the index if there is a link to it on a page. But pages that don't exist tend not to have a lot of links to them as a result have a very low PA/DA.

Once crawl data gets to old it will not be used for further index so bad links will expire that way.

The main index uses data that was crawled in the last 120 days (we may expand that number).

But it includes all the domains/pages that we crawled and all domains/pages that were linked to by pages that we crawled. So a page/domain may exist in the index if there is a link to it on a page. But pages that don't exist tend not to have a lot of links to them as a result have a very low PA/DA.

Once crawl data gets to old it will not be used for further index so bad links will expire that way.

Though you are correct.

I believe the talk you are referring to from Rand is about Google.