Because the polls for Obama are much stronger than they should be given the state of the economy, and he expects that to tell over the next month:

Nate Silver: Could 2012 Be Like 2008? - NYTimes.com: The “now-cast” estimates that Mr. Obama would have a 97.8 percent chance of winning an election held today… five and a half percentage points in the national popular vote. By contrast, the Nov. 6 forecast expects Mr. Obama to win by… 3.6 percentage points…. Two things account for this disparity:

First, there are still some effects from the convention bounce… adjustment [which] is phasing out of the model, but it hasn’t done so completely.

Second, the Nov. 6 forecast is still using economic data along with the polls…. [T]he economic index still accounts for about 30 percent of the forecast.

The forecast is thus 0.3(Economic Fundamentals) + 0.7(Polls). Polls is 5.6%. Forecast is 3.6%. That tells us that--if the convention bounce is over--Silver's economic fundamentals give Romney a 1% lead...