Two big balloons popped over the weekend. The Bears were routed at home by Miami, and those Super Bowl reservations already made by their fans now seem a bit premature.

And that Democratic wave in the polls that Stuart Rothenberg and Charles Cook have been surfing too? It might be a ripple.

New national polls by ABC/Washington Post, Pew, and Gallup, now show Democratic advantages in the generic Congressional vote of 6%, 4% and 7%, respectively . Now, that might not sound all that good if you are a GOP partisan, but compared to deficits  5—10% higher just a few weeks back, it is an important sign of  late movement.

So too, the Rasmussen daily tracking of Bush's approval rating has shown sharp improvement in the last week. Today, the approval/disapproval numbers are 45—52, for a 7% deficit  . The deficit was 17—18% a week ago. Rahm Emanuel, the über aggressive head of the DCCC, says the late poll movement is making him nervous. Having already lost 15 pounds while running around the country this year, the National Enquirer and People may soon be asking questions as to whether Rahm is an anorexic, if tomorrow's results prove very disappointing to his Party.  What do you call this,' not eating yourself sick'?

I would caution my friends on the right, however, from getting too excited yet.

The late GOP surge, mostly coming from Republicans coming home, and from some movement among independents, is probably a result of a combination of factors, including John Kerry's stupid remark and delayed apology, the Saddam verdict, the good unemployment number Friday.

Perhaps most importantly, some voters who are upset with Republicans may be reconsidering whether the alternative is really better (particularly on Iraq). The election is coming down on the margins, where the close races will be decided.

The Pelosi/Reid Democrats, seem to offer a blank slate of options of their own. Choosing the Democrats because it is the only way to express disappointment or anger with the conduct of the war in Iraq must be weighed against the unknown Democrat policies in prosecuting the war. I think the GOP is doing better in the last week among the disappointed, and still very poorly among the angry.

Turnout

It is worth noting, as Michael Barone pointed out over the weekend, that in 1998, the generic Congressional polls just before Election Day showed an 8 point edge for the Democrats. But the GOP held  the House with 225 seats, and that was with  a playing field of districts less favorable than today, due to redistricting in several states after the 2000 census, and in 2003 in Texas  that aided the GOP.  The GOP had a superior turnout operation in both 2002 and 2004 than did the Democrats (the Democrats were better in 2000), and the Democrats may have improved some since 2004. But on the margins, in close races, the GOP probably has a 1—2 point edge from its turnout operation.

Given how close some Senate races are, and with very close races in 15—20 House districts, the superior turnout operation could mean the difference between victory and defeat in a few Senate races, and 5—10 House races. If an incumbent GOP Senator is down 6 or 12 (DeWine, Santorum), the turnout operation will likely not be enough to protect their seat. But in the Senate races in Missouri, Virginia, Montana, Maryland, and perhaps even Rhode Island, it might. Call it the 'great escape' if it turns out that way.

It is very difficult to make race—by—race forecasts on House races, and easier to do so for the Senate. That said it is time to stop hedging, and make a forecast.

Prediction

I predict the Republicans will hold the Senate, and lose the House. Neither result is a certainty. However, if Republicans spring an upset and hold the House, they will almost certainly hold the Senate too.

The Senate

At this point, I think the GOP is safely ahead in 7 Senate races and also ahead by 5% or more in Tennessee and Arizona, which are closer races.  I think the GOP holds both Tennessee (Corker) and Arizona (Kyl).  With 40 seats held by Republicans not up this year, that gives them 49. With Dick Cheney as a tie—breaker in the Senate, it means the GOP needs only to win one more race to get to 50—50 and retain control of the Senate.

I think Lieberman will win easily and break the hearts of the Kossacks and assorted wing—nuts on the left. It might be that their hearts will be broken more broadly by Wednesday morning in other races too.  Counting Lieberman as a Democrat, and soon to—be—elected Vermont Socialist Bernie Sanders as one too, the Democrats have 27 seats that are not up this year, 16 Democrat—held sets in which the Democrat is safely ahead this cycle, and one more, New Jersey, where the Democrat (Menendez) seems to be ahead by about 5%. I think Menendez will win.

There are also two GOP—held seats that appear to be likely Democratic pick—ups: Pennsylvania and Ohio. Ohio is a bit closer (DeWine) than Pennsylvania (Santorum). With these two, the Democrats are at 46. 

That leaves five races undecided. For the Democrats to win control of the Senate, they need to sweep all five. I consider this pretty unlikely at this point, though not impossible.  Two weeks ago, Montana (Burns), Rhode Island (Chaffee) and Maryland (open—Democratic held seat) all appeared to be leaning or firmly in the Democratic column. No more. All are tossups. 

Two weeks ago, the best Republican chances of holding the Senate appeared to be to win one of the two hotly contested Southern/border state tossup races: Virginia (Allen) and Missouri (Talent), and hold Tennessee. Now they only need to win one of five tossups, with Tennessee looking safe. 

Looking only at pollsters I consider reliable, Missouri, and Virginia appear to be ties, with perhaps a very slight late edge to the Republican incumbents. Montana, Maryland and Rhode Island appear to have a slight edge to the Democrats, though the momentum in all three races is with the Republicans.

The movement in national approval ratings for Bush and for the GOP in the generic Congressional races, is just not there (yet) in the Senate races, where except for Rhode Island, the movement has been very small (Montana, Maryland) or glacial.   Forced to pick, I think the GOP turnout operation, and the 'southern—ness' of the states, carries Allen and Talent to razor thin victories and Steele pulls a big upset in Maryland. I think Burns and Chafee lose, narrowly in both cases.

The voters hate Republicans in New England this year (more so now that Peyton Manning beat the Patriots last night), and Burns in Montana never won by much even in good GOP years. For Steele to win, he needs 30% or more of the African American vote. I think he gets it. If he does not lose many GOP white voters due to his race (an unknown that no—one is talking about), that should be enough.

Of the five, Talent and Steele have run very solid campaigns, and deserve to win. Both have faced a barrage of the despicably misleading stem cell ads with Michael J. Fox http://www.americanthinker.com/articles.php?article_id=5977 .

For the record, I could be wrong on all five calls. I am not confident on any of them. The GOP could win 3 of 5 and I could still be wrong on 4 of my 5 calls, with Chafee and Burns among the winners.  That is how close the races are. So my Senate pick is GOP 52, Democrats 48, but the GOP could end up anywhere from 49 to 54. Second most likely outcome: GOP at 51. 

The House

As for the House, I feel a bit better than before about the GOP chances of holding control, but still think it is a long—shot.  I give the GOP a 20% chance of hanging on (losing fewer than 15 net seats). Last week, I thought the GOP would lose 20—25, this week maybe 15—20. Again, there are not a lot of new House polls that are out to see if the national generic trend (which by the way, Time, Newsweek and CNN deny has occurred) is showing up in local races. So this is guesswork.

As Jay Cost has pointed out, the wave theorists who think the GOP will lose 35—40 seats have been good at pointing to large numbers of House races the GOP is defending, but not at satisfactorily explaining how they reach their predicted level of seats lost with that list (unless you assume that the Democrats are actually favored in most of the tossup and lean Republican races).  

Both Stuart Rothenberg and Charles Cook identify 60—70 GOP held seats that are being targeted by the Democrats. But Rahm Emanuel said that list is only 48 seats on Meet the Press yesterday. If that is the case, and you do the Cost analysis on a much smaller list of vulnerable seats, the likely turnover of seats is even further from what the wave theorists predict than Cost suggests.

But I think the GOP is in big trouble in a few states with House seats:

Florida: The Foley seat (Fla. 16) seems back in play, though Fl. 22(Shaw) seems to be in greater danger of turning over. Fla. 13 (open seat of Katherine Harris) might also fall.

Connecticut: I think Rob Simmons (Ct. 2) will survive. I am not so sure about Chris Shays (Ct. 4) or Nancy Johnson (Ct. 5).

Pennsylvania could be the real Waterloo Tuesday for the GOP House chances, starting with weakness at the top of the ticket in Governor and Senate races. . Mike Fitzpatrick will probably survive in Pa, 8, but Curt  Weldon (Pa. 7), Jim Gerlach (Pa. 6) and Don Sherwood (Pa. 10) are all in trouble.

Ohio is another free—fall state for the GOP. DeWine seems to have stabilized his position in the Senate race a bit, which might help a few embattled GOP incumbents, but not much.  I think Ohio 18 (the Ney seat) is gone, and Deborah Pryce is behind in Ohio 15. Steve Chabot (Ohio 1) and Jean Schmidt (Ohio 2) are also not out of the woods.
 
Indiana: This is looking a bit better than a month ago.  I think Mike Sodrel will hold on in Indiana 9. Chris Chocola is closing the gap in Indiana 2, and John Hostettler will probably go down in Indiana 8.

Arizona: The GOP will lose the open seat (Arizona 8),and is in trouble in Arizona 5 (Hayworth)

Colorado: Another bad state for the GOP this year. Republicans will lose the open seat (Colorado 7) and could lose Colorado 4 (Marilyn Musgrave).

Kentucky: Two seats at risk: Geoff Davis (Ky 4) and Anne Northrup (Ky 3).

New York: This is the real potential wave state if it is a very bad night for the GOP. The open seat (NY 24) is probably lost, and John Sweeney is in big trouble (NY 20).  Tom Reynolds will probably hold on in NY 26, but GOP incumbents are not out of the woods in several other races:  NY 19 (Kelly), NY 25 (Walsh), NY 29 (Kuhl). With Elliot Spitzer winning by 45% and Hillary Clinton by 30%, it will take a lot of ballot switching for the GOP to hold on in some of these races.

Then there are some solos: Texas 22 (De Lay seat) is almost certainly lost. So is Iowa 1 (open seat).  (Charles Taylor has been behind for months in NC 11 against former quarterback Heath Shuler.  This race has  the most unusual ads by any independent group this cycle.

Heather Wilson is a bit behind in New Mexico 1, Dave Reichert is slightly ahead in Washington 8, and the Democrats seem to have momentum in New Hampshire 2 (Bass).

The GOP also has to defend very competitive open seats in Wisconsin 8, Minnesota 6 and Illinois 6. I think the GOP is positioned to win at least 2 of the 3, and has moved ahead in Minnesota 6.  Arnold Schwarzenegger's big win in California will help a few embattled GOP incumbents: John Doolittle (Ca. 4, the incumbent with the worst name this year)) and Richard Pombo (Ca. 11). Finally, the GOP had made a late effort to defend an open seat in Nebraska 3, Kansas 2 (Ryun),  Idaho 1(Otter) and the at—large seat in Wyoming (Cubin).

It is a measure of the GOP's problems this year that these seats were still in play on the final weekend. It is not hard to see the GOP losing 20 or more seats from this collection.

A trend to the GOP?

If there is a real 3—5% shift toward the GOP in the close House races reflecting the movement in the national polls, it could save or solidify the leads of Wilson in New Mexico, Musgrave in Colorado, both Kentucky seats, most of the New York incumbents, Gerlach, Chocola, Hayworth, Shaw, 1 or 2 of the Connecticut incumbents, a few seats in Ohio, and some of the Midwest open seats. These are all very close races, in many of which the GOP candidate is slightly behind in the latest poll.  The Pew poll says there has been more movement to the GOP in the Northeast (where it was weakest) than in the Midwest.

My forecast now is 15—20 lost seats, down from 20—25 in my previous assessment. This reflects the 5 at—risk GOP seats that probably have already moved due to the national shift. I just do not know which ones they are.  Since there are 10—12 GOP held seats that will almost surely turn over, it is not difficult to see the Democrats winning  3—5 of the tossups or leaning—Republican seats to gain control.

The X—factor: GOP pickups

There is one x—factor for the GOP. They could win a few Democratic held seats. Georgia 12 (Barrow ) is a tossup, and Illinois 8 (Bean) and Georgia 12 (Marshall) are at best slight leans for the Democrats. Georgia will be a good state at the top of the ticket for the GOP on Tuesday , and that will help the two GOP challengers (Burns and Collins). Illinois will be a competitive state due to the gross incompetence and corruption of the Democratic Governor Rod Blagojevich, who is only a few points ahead in his re—election bid . That might help the GOP in the close suburban districts: Illinois 6 and Illinois 8.  Indiana 7 (Carson) is a long shot for a GOP pickup.

If the late momentum helps the GOP cut its losses, and the Party wins a few Democratic seats, that could be the formula for narrowly holding the House tomorrow. But again, I think it is a long—shot.

In conclusion, things are looking up a bit for the GOP, but it will still not be a banner year for the Party by any means. Michael Barone has said the Iraq war probably cost Bush 2—4% in his popular vote margin in 2004, which means he might have won several more states without that overhang: New Hampshire, Pennsylvania, Wisconsin, Minnesota, Michigan, maybe Oregon. With 350 Electoral College votes, and a 6 or 7 million popular vote margin, Bush would have won as decisively as Clinton in 1992 and 1996, and the 'Ohio was stolen' conspiracy theorists — RFK Jr, and Mark Crispin Miller — would have had to invent something else.

This year, the Iraq overhang is larger—maybe 5—7 % I think.

I will be back in a few days to tell you why I was wrong about everything!!

Richard Baehr is the chief political correspondent of American Thinker.