Campaigns don't matter?

[zorkpolitics]

Despite all our intensive poll watching and daily campaign dissections, angst about the GOP VP choice and eventually wondering who will win the debates, there is considerable data that campaigns don’t really matter. Presidential results are driven by partisanship, economic factors, job approval, and incumbency.
Good ole Larry Sabato has revised his prediction model and now has Obama winning if the 2Q GDP is positive. With two of the three predictors used in his revised model now set, a 1% 2Q GDP growth would equal a 50.2 to 49.8% Obama win. To win Romney needs the 2Q GDP growth to be negative.
So no need to read the news until Nov., be happy don't worry.

http://www.centerforpolitics.org/crystalball/articles/abramowitzpolarizationmodel/

[The Mikado]

I'll believe the "Campaigns don't matter" theory the cycle we get a candidate who wins while not leaving his front porch again.

Campaigns are, however, very susceptible to diminishing returns.  The campaign that spends $1 million does far better than the campaign that spends 0, but the campaign that spends $2 million sees a much more marginal improvement over the campaign that spends $1 million etc.