I was complaining about their methodology mostly. If you took the number of people running for office who had previous political experience (which means they had served in office) I would guess that there was a power log curve in there. If they have served 5 or more years in another public office before running for Congress I would expect they have a very solid chance to be elected.