It seems like GOP Senate candidates are gaining ground in key states, even as Trump's poll numbers flag. This is a really interesting development.
It's only October 12th. I expect to see the effects of Clinton's upcoming landslide in the polls for downballot races only a week and a half or so from election day.
What, do coattails work on a time-delay now?
Come on dude, you're smart. Even in 2006 Webb inched ahead of Allen the last few days of the campaign, not after the Macaca gaffe or the Foley scandal.
You're talking about something different here, and not something that really seems generally applicable. I would need a lot more examples than just George Allen to start making predictions based on a theory of time-delayed results, in part because it lies in the face of conventional logic that summer gaffes don't really matter since they have time to be forgotten. Also Allen's case doesn't involve coattails; the candidate himself made the fatal gaffe that sunk his campaign.
Donnelly and Heitkamp in 2012 pulled ahead the last few days before the election.
Also Udall (CO) and Merkley in 2008. That's from the top of my head.
Are these in response to scandals or just late movement? (And by scandals I don't count Mourdock being an awful candidate who should never have beaten Dick Lugar in the primaries. 
)
Obviously it was late movement. These people had the wind on their backs and that's why they prevailed in those close contests. It's the same reason why Tom Tillis won despite being an awful candidate and trailing consistently during the entire campaign.
Like they say, "“Better to be lucky than good".
