The latest crystal ball prediction from Larry Sabato at the University of Virginia's Center for Politics contains some fear inducing predictions for Democrats. "...if the election were held today... Republicans would pick up nine governorships currently held by Democrats, but lose three they currently occupy. Therefore, the GOP would have a net gain of six governorships, plus opportunities to play offense in three more toss-up races in states where Democrats now reign." Sabato further predicts the GOP will gain 7 Senate seats and 27 House seats.
In Maryland, Sabato predicts that Robert Ehrlich would defeat Governor Martin O'Malley and he rates the race as a likely GOP win.
I have argued for weeks here, here, and here that Governor O'Malley faced a serious threat from Ehrlich - it's nice to see Sabato following my lead, and I'm willing to back up my analysis with data.
I feel obliged to mention as well that I've been discussing the potential for significant Republican gains in the House and Senate for several weeks. I think that Sabato is underestimating Republican gains in the House and Senate. Independent voters are becoming an ever more important factor on Election Day, and this introduces significant unpredictability into our elections. From one election to the next roughly 90% of Democrats and 90% of Republicans cast a vote loyal to their party, but fully 1/3 of Independents shift their support between the two parties. When Independents represented a smaller slice of the electorate this was not such a issue. Independents now comprise fully 1/3 of the American electorate - rivaling Republicans and Democrats in voter share. A shift of 1/3 of 1/3 of the electorate - or roughly 11% - can have a significant impact on election outcomes. Independents broke heavily for Democrats in 2006 and 2008 - delivering them the Congress and White House. But in VA, NJ, Massachusetts and nationally Independents have turned on the Democrats. Evidence of this shifting support can be seen in the Republicans current 3 point lead in the generic Congressional ballot - this is significant given that in the two most recent election cycles when Republicans gained seats in the House and Senate - 1994 and 2002 - the generic ballot in November of each election year had Republicans and Democrats either tied (at 46% in 1994) or Republicans trailing the Democrats (by 5% in 2002) - yet Republican performed better come election day. The simple reason being that Republican voters are more likely to turn-out. So Democrats need a significant lead to offset turn-out. If Republicans enter November with a lead in the Generic ballot then Democrats can say goodbye to their House majority and expect no better than a 1 seat majority in the Senate.
And 2010 matters because states will engage in redistricting after the 2010 census - gaining the governorships in states like Maryland, Illinois, Ohio, Iowa, Wisconsin, and Michigan would give Republicans an advantage in redrawing Congressional districts that could then protect a potential GOP majority for a decade.
