many he predicted last time.  My guess is less than 10, which would seem about right--- add 20 to his prediction and you've got a good idea of what the Dems will gain this time.

Nothing has improved for the Republicans this time, aside from the fact that the lost many of the close seats last time.

But there were another dozen or so that were still very close, and they don't have the monetary advantage this time that they had last time.

I see the Dems picking up at least 15, probably closer to 25.

check to continue the U.S. occupation of Iraq.  

It will be about 8 years of Bush not one year with Dem Congress...
Any troop KIA or WIA is one too many. Any Iraqi death or imprisonment is one too many ....for that pretend GOVT.
Folks are paying their heating bill with credit cards...We ain't seen nothin on the economy.
I want Santa to bring me for Xmas 2008?
Edwards Prez..Obama VP
Dennis as Speaker
Dodd Majority Leader in Senate

...I promise to be very..very good.

Cook believes far too strongly in the PVI and last cycle he listed seats with entrenched Democratic incumbents as in play for the longest time.  North Dakota At Large springs to mind where Earl Pomeroy has not had much of a challenge in years but Cook listed it last cycle.

Second, he consistently believed rosy Republican assertions when there was nothing to back it up. In NY-24, NRCC Chair Tom Reynolds said, "I know New York and there is no way that we lose this seat."  Well, the polls said it and their state senator went down semi-easily.  Anybody could see it coming.  I sure saw it coming.  Cook did not.

Another case was South Dakota At Large where the GOP had a pathetic candidate.  He also bought Alan Mollohan's seat in WV and John Spratt's seat in SC as pretty much gone.

Beyond that, Cook got famous in an era when very few seats turned over.  He underestimated turnover.  Very gradually increasing his estimates pretty much each time for about a year but still underplaying it by about 14 seats (if tossups are split evenly and leans go with the favored party, Cook predicted Democrats would win 219 seats in 2006).  To his credit, Cook did discuss at length the possibility of a wave election for months before it happened.

My best guess right now is a gain of 15 to 20 seats but retirements (and unretirements like Jim Ramstad's) plus the Presidential race could effect that.  Republicans won a majority of the close elections in 2006 and Democrats have raised a lot of money for pretty much all of their freshmen.  Both good signs.    

That have been attacking Pelosi and Reid since days one because they couldn't achieve world peace and end poverty instantly upon taking power.

I think we will have to settle for small gains with the far left of the party peeing inside the tent instead of out.

Personally I expect little change in the House this year, at best a Democratic pickup of no more than 6 seats.  I think it's a lot harder to gain seats in the House than in the Senate.  I also think the Democrats missed some opportunities in 2006.  Consider NC8.  The unknown, underfunded challenger got no help from the national party yet lost by only about 400 votes.  The conventional wisdom is that this year with more money he has a great chance.  However, this is a district with about 125,000 votes in off year elections and about 225,000 votes in Presidential elections.  The incumbant has had several close calls in off year elections but has generally won handily in Presidential year elections.  Instead of needing another 500 votes to win like last year, the challenger will probably need to pick up 15,000-20,000 votes next year.

Now, admittedly, some close calls last year were in Democratic-leaning districts where the extra voters will help.  But the majority were in Republican-leaning districts where the higher turnout will make things harder.  Of course, maybe this will be another "wave" election, but I think it is more likely that the Democrats will merely solidify their advantage rather than make any significant gains.

I'm much more optimistic in the Senate, where +3 appears to be an absolute minimum, +4-+6 likely, and +7-+8 not out of the realm of possibility.

What will be the impetus to vote Democratic?

In 2006, we had a combo of corruption and Mark Foley.  These are now over.

What's left is the Iraq war.  How will we play this?

I would take everything with a grain of salt now.   Cook's predictions evolve with events.    In the year leading up to November '06, there was a steady progression of changed assessments, from Safe Republican to Republican Favored, from Republican Favored to Republican Lean, from Republican Lean to Toss-Up.  And remarkably few previously Republican seats to even Democratic Lean.

Cook's methodology puts a heavy weight on both incumbency and PVI, which is pretty rational actually, except that it doesn't work when the national mood is very pronounced one direction or another.

Actually, the Republicans DID have a larger majority once.  Going into the 1996 House elections, they had 236 seats, to the Democrats' 198 (+ Bernie Sanders, so really 199).  The 1994 elections resulted in 230 Republican seats to the Democrats' 204 (+ Sanders), but all the party switchers to the GOP over the following 2 years resulted in the larger majority by fall 1996.

Source:  http://en.wikipedia.org/wiki/United_Stat es_House_of_Representatives_elections,_1 996

moondancer put it well:

They will lose some of the pickups from 06.  They have done a horrible job of sustaining message.  I think the republicans have done a good job in tarring congress

First of all, we're going to lose some seats that we picked up last time.  I don't see how candidates like Boyda or Lampson win in the districts they represent, especially with a presidential race activating those Republicans that stay at home in off years.

Plus, look at what we had to work with last year: The Iraq War, Foley's follies, congressional corruption and Bushco's ineptitude/arrogance.  The war is the only one of these issues that has a chance to be as influential next year as it was in '06.

Finally, the math just doesn't make sense.  There's simply not enough Republican districts that are +-3 PVI (where the bulk of pickups occur).  While I feel good about Democratic enthusiasm, recruiting/fundraising successes and Republican retirements, if the math doesn't make sense, it doesn't make sense (see OH-5).

I live in Missouri, and you'd have to lay me about 5/3 before I'd place a bet on Kay Barnes to unseat Sam Graves (MO-6, PVI R+5).  It's probably our best bet in the state, but that doesn't mean it's a good one.  There's just not a strong chance we can take districts like these.

While I'm a bit more bullish about our prospects in the Senate (where we can really demonstrate Republican obstructionism and the need for 60 Democratic seats), I think Cook is spot on about his predictions for the House.

11 months out, I'm thinking 3-5 Dem pickups in the house and 4-6 in the Senate.

Charlie Cook responds:

First, to answer the question raised about what The Cook Political Report predicted in 2006, our final predictions were for a 20-35 seat gain in the House (the final was 30), 4-6 in the Senate (final 6) and 4-6 in governors (final 6) (http://www.cookpolitical.com/column/2006 /110106.php).  In fact our predictions didn't change much over the last 60 days of the campaign.  We were the first major analysts/journalists to predict the House turning over, back in the first week of August of 2006, indeed before the consensus in the netroots that the House would turn.  Anyone going on our website can see that over the course of the year before the election, as the political climate worsened for Republicans and as the situations in individual races got tougher for the GOP, our projections of Republican losses increased as well, such projections are dynamic, not static, they change as the circumstances change.  Had the midterm election been held a year earlier, Republicans wouldn't have lost the House.

For those who haven't been following Congressional races for a long period of time, the Cook Political Report was founded in 1984, thus this is our 13th election cycle analyzing these races, though my own political experience preceeded that by 12 years.  Our analysis is based on having watched races in every single state and Congressional district for going on 24 years now, meeting with candidates from both parties as well as off-the-record conversations with campaign consultants, managers, party strategists from both sides as well as monitoring the local press coverage.  That's why the New York Times referred to the Cook Political Report as "a newsletter that both parties regard as authoritative."

Our House Editor David Wasserman's latest projection of 2-8 seats was based on a careful analysis of each district, the recruitment and development of campaigns in each district, analysis of money, the voting patterns in that district and, to a small extent this early, the national political climate (as the election draws closer, the climate becomes more important).

Many seem not to appreciate the difficulty of picking up large numbers of House seats in consecutive elections.  Only once since the end of World War II, a period covering 31 elections, has a party picking up 20 or more seats in one election and ten or more in the next (in 1950 Republicans picked up 28, then 22 in 1952).  In fact only two other times did a party pick up double digit House seats in consecutive elections, again Republicans in both 1978-80 and 1992-94. To put it differently, in eight elections starting in 1946, a party picked up 30 or more seats.  None managed a double digit seat gain in the next election.  In a big election, the lowest hanging fruit is picked, making it harder to pick as many easy seats in the next election.  

Only eight Republican-held House districts won by John Kerry in 2004, but there are 60 Democrats in districts carred by George W. Bush in that election. This isn't to suggest that Democrats will lose seats, only that there aren't that many Republicans sitting in districts in enemy territory.

This is not to say that Democrats cannot pick up ten ore more seats next year, but it is really hard to do that and at least today, the seats aren't there to predict it.

Today we have eight Republican House seats in the Toss Up column, all open seats, with another 14 Republican seats in the Lean Republican column, all incumbents.  Conversely, we have no Democratic seats in the Toss Up category, but have 14 Democratic seats in the Lean Democratic column.

Early on in election cycle we try to be very conservative about putting incumbents in the Toss Up column, even those who appear to be very vulnerable.  The simple reason is that with incumbent re-election rates of 95 percent, and even quite a few of the most endangered incumbents managing to survive, we would rather wait until into the election year to push incumbents into that highly endangered column.  We like to watch the challengers and their campaigns develop, see how they perform and how their money is going.  Over the next few months we expect that the ranks of Toss Ups will expand, we just don't want to do that prematurely.

One word of caution.  Many of the Democratic gains in 2006 came in districts in the South and Border South, in districts with strong small town and rural flavors, and most were won by Bush in both 2000 and 2004.  It is unlikely that any of the most probable Democratic presidential  nominees would carry many of these districts, and if they did, not likely by enough of a margin to be an asset.  In some of these districts the top of the ticket will be a liability.  I would not expect the hot house environment that Democrats enjoyed in 2006 to be replicated, notwithstanding the excellent polling done by the Democracy Corps that shows a favorable environment for Democrats today.  If the environment is good for Democrats, then enjoy it.  But I wouldn't count on it.  As suggested earlier, big years rarely come in pairs.