Search API "sampling": What kind of sampling?



Following the recent changes to the Search API documentation:

@edsu highlighted a key phrase in there: “Twitter Search API searches against a sampling of recent Tweets published in the past 7 days.”

It would be great if anyone from twitter could clarify or expand on this. Having a better sense of what’s indexed will definitely help avoid pitfalls & false assumptions when working with twitter data.

What exactly is meant by “sampling”? Does it use one of these methods: If so - Any details?

Or does sampling mean something else more complicated or something simpler? Something like "filter_level":"low" tweets less likely to be returned in Search API?

Any clarifications / corrections or relevant links would be great!