@odd I think it works this way: Google is a domain registrar so they get the data on new domains registered automatically. They then put those domains on Googlebots crawl list to check out.