Those projections are actually an improvement from only a month or so ago. If the GOP can hold the Dems' net gain to under +5, I think that would be a victory of sorts.
For a comparable projection, Larry Sabato (Center for Politics) says the most likely outcome is a +4 (net) gain for the Dems.
I keep seeing these kinds of predictions and I wonder why we measure success in this way?
This isn't the House or Senate we're talking about here. The gross number doesn't mean all that much.... it's WHICH states we win/lose that matters.
PA will hurt. TX would hurt ALOT (though I doubt it's in too much danger) since it hurts Bush and the party (and keeps us from picking up 3-4 seats in TX in 2004 from redistricting)
On the other hand, If we win CA I won't mind a 5 seat net loss because that's nationally a big win. If we take CA AND MD without losing TX or NY it's actually a good year for republicans.