Archive for October 20th, 2010

Handicapping the Midterms

Although I am blogging far less about the 2010 midterms than I did in the exciting 2008 Presidential campaign, this year really has some intrigue and uncertainty.   The scenarios I painted just over two weeks ago remain the same, and the evidence still points to a 45-50 seat Republican gain in the House, but recent polls have slightly good news for Democrats (emphasis on slight).

Republicans expect this election to follow the pattern of a “wave” election — late voters switch to the winning side, the GOP, and close races get swept into the “R” column.  This means that every poll which has a Republican leading will yield a Republican victory, and most Democrats still nursing a small lead will likely go down to defeat.   Given the state of current polls, that could mean a pick up of 60 to 80 Republican seats in the house, an historic victory.   Evidence supporting this is the Gallup poll of likely voters, showing that in a low turn out election the GOP has a 17% point lead in the generic ballot.  Even the high turn out scenario shows them with an 11% “enthusiasm” lead.

Yet it could be that this election is more like 2008 — Obama had a big lead in mid-October, and it appeared the country was embracing change.  But as election day neared, the Republicans rebounded and made that election a bit closer.  Simply, Obama’s “wave” was early.    It is possible that the Republican wave peaked in September, and there isn’t a late surge coming.   After all, Democratic disillusionment was extremely high in August into September, and the headlines negative for Obama and the Democrats.    If there is no “second wave” for the GOP, the 45-50 range of pick ups is likely.

Recent polls, however, suggest the possibility of a Democratic late mini-surge.

1.  Obama’s job approval.   Rasmussen has been unsteady — the tracking polls always are, but recently Obama has been close to 50%.   This could indicate an increase in Democratic likely voters, as the poll focuses on likely voters, unlike Gallup.  But Gallup today shows Obama with a 48-44 approval edge, higher than recent polls.   It’s impossible to put much weight on the tracking polls, just a few days ago Rasmussen had Obama down almost ten.  Still, there are signs that his popularity is at least stable, and perhaps improving.

2.  The Senate.   Recent polls suggest that the Democrats are faring marginally better than they used to be in Kentucky (Paul, R +5), Washington State (Murray, D +2) , Colorado (Buck, R +1,+3), Alaska (Miller, R +5), Wisconsin (Johnson, R +2), West Virginia (Raese, R +3), and Pennsylvania (Sestak, D +1).  These are toss ups.   In Nevada Harry Reid faces the fight of his life; unless the Democrats do close the enthusiasm gasp, he is very likely to lose.

If trends do not change, the Republicans will likely gain seven seats with the Senate split 52-48.   To gain a majority they’d need a wave to wipe out Boxer in California (unlikely) and then pull off upsets in Connecticut (even more unlikely) and Washington (the most likely of the three).   The Democrats could, however, really trim their loses if they get the mini-wave.  If the close elections above all went Democratic, they’d lose only two (which would include picking up Alaska, which is bizarre due to a Murkowski’s write in candidacy).  Missouri (Blunt, R +6) is still within reach if the Democrats have a late surge.   The chances of the Democrats losing only one seat is as likely as the Republicans gaining a majority, yet each remains possible.   To break even in the Senate the Democrats would have to capture North Carolina (Burr, R +8).    It looks like it’ll be 52D-48R, but it’s not out of the question that it becomes 49D-51R or 59D-41R (assuming independents caucus as they have been).

3.  The House.   Nearly 100 seats are “in play,” and polls show the races close.   Recent polls have benefited Democrats, but only slightly and in polls of varied methodology.    Examples are NY 19 (leans GOP) where Hall (D, +1) has a slight lead.  MA 10 is a toss up, with Keating ahead, (D, +3).  AZ 3 is another GOP leaner, with Dan Quayle’s son now slightly behind Halburd (D +2).  In Virginia Hurt (R, +6) once had leads over 20% against incumbent Tom Perriello.  It’s still a Democratic longshot, but people sometimes come home to the incumbent.  Another toss up is NY 25, with Maffei (D, + 12) having a larger lead than earlier polling which showed him up 3.   The movement over the past week is distinctly towards Democrats, but in small doses and a scattering of races.   Who knows if this is indicative of anything other than a few local idiosyncrasies.   This gives some evidence against a renewed Republican wave, but is not strong enough to suggest things break Democratic at the end.  The uncertainty level is great.

To show how large the GOP advantage is, if the toss ups were split evenly, the GOP would gain 57 seats for a majority of 235R – 200D.   The toss ups are mostly Democrats in Democratic seats, and if the current situation continues it would probably be a bit less severe of a loss, about 47 seats, or 225R – 210D.  That’s the range most experts predict (45-50 seat gain for the GOP).  If the GOP had a wave and the toss ups almost all went GOP, it would a gain of about 70 (248R – 187D).  If the Democrats took all the tossups they’d only barely hold on to the House, losing 35 (213R – 222D).  However, a late Democratic resurgence — feasible since these are incumbents in often Democratic districts — could flip some currently “leans GOP” seats, trimming Democratic loses to perhaps under 30.    Clearly, the Republicans have the advantage, only their worst case scenarios show them not gaining the House.   (The realm of possibilities include 283R – 152D, R +103 to 173R-262D, D +7 — that spectrum clearly favors the Republicans).

4.  The polls themselves.   There is some evidence that due to methodology Rasmussen has a structural “house effect” that advantages Republicans.   This means that a Rasmussen poll that shows a Republican up three may really indicate an even race.   I find that plausible, but if the enthusiasm gap is as large as many think, Rasmussen’s assumptions could be accurate.  Others claim that the lack of cell phone polling and other factors under counts Democrats.   That seems more like wishful thinking from the left since most polls still have good track records — you can usually trust polls to give you what they claim, a snapshot that is within the margin of error about 90% of the time.  That also means outliers are always present — and one tends to trust outliers that benefit one’s own party and dismiss those which benefit the other.

What to make of this?  It’s really an interesting off year election.  It is also fascinating how fickle the American electorate is.  Strong rejection of the GOP and an embrace of change and President Obama to apparent readiness to give an historic victory to the Republicans.   In 2012, it could shift back the other way.   Lacking the ideological fervor of the party activists, it appears the public just wants to figure out who can actually solve the problems.   It may be that neither can, at least not alone.