Sabato's Crystal Ball

2006: The Big Picture

Twelve Days and Counting

Larry J. Sabato, Director, U.Va. Center for Politics October 26th, 2006

facebooktwittergoogle_plusredditpinterestmail

Dear Friends of the Crystal Ball,

We’ve been through a lot together this election cycle, and now we’re headed for the last round-up on November 7th. Your Crystal Ball staff will be tweaking our predictions right up until election eve, taking into account last-minute shifts, scandals, and surprises. But the outline of this election couldn’t be clearer: 2006 is a Democratic year. The analysts and observers are just debating how big a Democratic year it will be.

Republicans are more despondent than they have been for decades, and party leaders have found it especially difficult to wear a brave face into battle throughout October. The good news for the party, though, is that elections are held in November. The ticking clock is becoming more audible in the background, but even in the precious twelve days left before Election Day, a narrowing window remains for the “Glum Old Party” to seize new momentum and recoup at least some of its grandeur.

To be sure, the way most races have progressed this month, it would seem that just about the only scare-free Halloween fact for the GOP is that their nightmarish October is drawing to a close. Oddly enough, though, many Democrats we correspond with are just as spooked as their counterparts. Indeed, Democrats have grown so accustomed to snatching defeat from the jaws of victory this century that a plethora are haunted by the expectation a Bush-Rove-Mehlman “trick” will thwart a Democratic “treat” in the days leading up to Election Day.

But Democrats need not dwell on the fears of the conspiracy theorists in their ranks; the Crystal Ball believes their concerns are misplaced. The “trick” up the GOP’s sleeve in the last six years has largely been the coupling of a consistent, hard-driven national security message and a superior get-out-the-vote ground game. We know that at least the latter half will be in action this year. But as every party turnout strategist knows, at its most effective, a good field operation is worth only one or two points at the ballot box. In as lopsided a year as 2006, Republicans must cross their fingers for the onset of helpful forces larger than those they control in order to cut their impending losses.

A Republican gift of yet-to-be-determined value arrived yesterday in the form of a significant judicial state ruling the Crystal Ball had quietly been anticipating from a distance. Sure, the names Albin, LaVecchia, Long, Poritz, Rivera-Soto, Wallace and Zazzali don’t ring a bell to most political junkies outside of the Garden State (and dare we say, many inside the Garden State as well), but New Jersey’s seven Supreme Court justices have unwittingly come to the luckless GOP’s aid; they may have just influenced several key midterm races in a way no fundraising visit from Bush, Cheney, Giuliani, Hastert, McCain, Mehlman, or Rove could have. The state high court’s decision to mandate the legislature to pass full legal rights to New Jersey’s same-sex couples could not have come at a worse time for Democrats all across the country.

In 2004, anti-gay marriage ballot initiatives in 11 states greatly aided Republican prospects in key races by boosting turnout of base conservatives, in particular, easy-to-organize evangelical churchgoers. As a social wedge, the issue was fresh in voters’ minds following a court ruling to similar effect in Massachusetts that social conservatives decried as unwarranted judicial activism. The GOP’s rallying cry against “activist judges” helped Bush carry critical Ohio among other states and assisted the party’s efforts to retain congressional control. But in the absence of more rulings and with the passage of time since 2004, there is little question this wedge had lost some of its edge in the run-up to the 2006 midterm elections.

All of that may have changed. So long as the GOP is able to both keep the story alive and dedicate some resources to capitalize on it–a big if, might we add–New Jersey’s ruling holds the potential to revive dormant conservative hostility towards judicial liberals at a time when many conservatives, disheartened by the Foley scandal and other Washington improprieties, may have considered sitting this midterm out. In the face of renewed calls to arms against gay marriage and civil unions, anti-gay marriage ballot initiatives in eight states (Arizona, Colorado, Idaho, South Carolina, South Dakota, Tennessee, Virginia, Wisconsin) would now be favored to prevail with more sizeable margins than initially anticipated. But more importantly for our purposes, added evangelical turnout could prove a decisive boon to GOP turnout in key races for control of Congress all over the nation–perhaps in Tennessee and Virginia’s absolutely crucial Senate races.

The Crystal Ball reads the implications of this political hot potato with caution. It’s possible that the story could fade from media memory in a day and ultimately be forgotten by voters; it’s also possible that the political impact of the Garden State’s ruling could be conveyed by religious conservative organizations under the radar, only surfacing to haunt the Democrats on November 7th. Time will tell. Nonetheless, Democrats are now scratching their heads and asking, “Why couldn’t New Jersey’s Supreme Court have announced this decision November 8th?” If Democrats’ gains are not as robust as they would have liked when the votes are counted twelve days from now, Republicans may always take glee in this order of events.

So as the Crystal Ball unveils its latest estimates, potential intervening events like yesterday’s serve as good reminders that we must constantly recalibrate our predictions as races are impacted by forces both large and small, and both national and local, all the way through to the one day clearance sale we have come to know as Election Day. Still, the fact remains, Democrats will be picking up seats in the House, the Senate, and the statehouses. The question is, of course, how many?

As we take a snapshot of the electoral landscape from the perspective of twelve days out, our best guess is that Democrats are on track to net 21 to 26 seats in the House, 4 to 6 seats in the Senate, and 5 to 7 governorships. We place slightly more confidence in our ranges for the Senate and governor categories, as so many House races are fluid.

Changes could push these projections in either direction, but here’s a catch: inside of two weeks left, underdog candidates who seek to make their move must do so now or forever hold their peace. The second-to-last week of a campaign is typically the last week voter persuasion is conducted before full-blown get-out-the-vote efforts take over, and for good reason: new arguments and attacks rarely resonate during the final week, as they are often viewed suspiciously by the media and quickly get labeled “too little too late.” In the last few days of campaigns, it’s mostly the gaffes and stumbles of candidates that truly swing momentum in a new direction, and we can count on our fair share of them this year! Of course, the Crystal Ball will be keeping a watchful eye for these and will provide you our latest information on the races we see as “late breakers” next week. By then, we just might have a better idea of whether Democrats will run up the score, see their gains fade, or simply run out the clock on the GOP as things currently are.

Our Republican friends have been emailing us by the hundred, insisting that we (and the other nonpartisan election rating services) are all wet. If we are wet, I suspect it is from the Democratic wave that is passing over us. Yet I will say two things: (1) You could be right. Would it be the first time that all the experts have been off-base, whether in politics or sports or meteorology or any other field? (2) You could be wrong. Maybe it is because you have contact mainly with people who agree with you, in your family, your choice of blogs, your church, and your community. I will always remember an exchange between my dear mother and father on election night 1964. Seeing a certain controversial candidate handily winning yet another term of office, my mother exclaimed, “How can that be? We never vote for him and no one we know votes for him, yet he’s winning!” All of us are products of our immediate environment.

If it is any comfort to our GOP readers, we were deluged with complaints from Democratic writers in 2002 and 2004, when the Crystal Ball predicted overall Republican victories. In 2004, I’m proud to say the Crystal Ball’s predictions were accurate in 475 of the 479 Governor, Senate and House races on the ballot (99.2%) on election night. Of course, looked at more critically, we missed one House, one Senate and two Governor contests. I guarantee you we will miss some again, though I pray not too many.

What does this tell you? First, I hope it says that we are fairly reliable as election predictors. Second, we are also fallible and will never have a perfect record. Third, and most important, we simply want to get the calls right, just like all our nonpartisan colleagues.

As you know, we love historical analysis at the Crystal Ball, and though it can sometimes be misleading, we have long believed that the Sixth Year Itch would need to be scratched again in 2006 (see previous emails here). If it happened to Teddy Roosevelt, Woodrow Wilson, Franklin Roosevelt, Harry Truman, Dwight Eisenhower, Lyndon Johnson, Gerald Ford, and Ronald Reagan, the itch was highly likely to strike George W. Bush, too. That great exception, Bill Clinton, precociously had his sixth year itch in the second year, but the one exception does not disprove the rule.

Table 1. How High the Midterm Wave?

Year President HOUSE

Gain or Loss
SENATE

Gain or Loss
Size of Wave For or Against President
1946 Truman (D) -55 -12 High Against
1950* Truman (D) -29 -6 Medium Against
1954 Eisenhower (R) -18 -1 Low Against
1958* Eisenhower (R) -48 -13 High Against
1962 Kennedy (D) -4 +3 Low Neutral/For
1966* Johnson (D) -47 -4 High Against
1970 Nixon (R) -12 +2 Low Neutral/Against
1974* Ford (R) -48 -5 High Against
1978 Carter (D) -15 -3 Low Against
1982 Reagan (R) -26 +1 Medium Against
1986* Reagan (R) -5 -8 Medium Against
1990 Bush (R) -9 -1 Low Neutral/Against
1994 Clinton (D) -52 -9 High Against
1998* Clinton (D) +5 0 Low For
2002 Bush (R) +6 +2 Low For
2006* Bush (R) -21 to -26 -4 to -6 Medium or High Against

* Indicates a “Sixth-Year Election.” 2006 gains/losses based on current Crystal Ball projections.


Like every other part of life, there is a cycle to politics in a basically two-party system. Parties gain power, enact some or most of their platform, ascend further in popularity, then begin to exhaust their agenda, and fall back, until the other party is given a chance to follow the same trajectory. The GOP control of Congress has traversed this well-trod path, and George W. Bush’s thin presidential victory in 2000 was followed by two very good Republican election years. The season is changing anew, as we will see in less than two weeks.

Now to our specific tallies in the big races of ’06, whose results may prove the accuracy of the historical big picture we’ve outlined.


Recent 2006 House Crystal Ball Race Outlook Changes

Recent 2006 Governor Crystal Ball Race Outlook Changes

2006 Senate Crystal Ball HotRace Readings

Republican Held Seats up for Election in 2006: 15 (out of 55 held)

Solid R (7) Likely R (1) Leans R (1) Toss-up (3) Leans D (3) Likely D (0) Solid D (0)
IN (Lugar) AZ (Kyl) VA (Allen) MO (Talent) MT (Burns)
ME (Snowe) RI (Chafee) OH (DeWine)
MS (Lott) TN (OPEN) PA (Santorum)
NV (Ensign)
TX (Hutchison)
UT (Hatch)
WY (Thomas)
WV (Byrd)
WI (Kohl)
NY (Clinton)
NM (Bingaman)
ND (Conrad)
MA (Kennedy)
HI (Akaka)
FL (Nelson)
DE (Carper)
NE (Nelson) CT (Lieberman*)
WA (Cantwell) MN (OPEN) VT (OPEN)
NJ (Menendez) MD (OPEN) MI (Stabenow) CA (Feinstein)
Solid R (0) Likely R (0) Leans R (0) Toss-up (1) Leans D (2) Likely D (3) Solid D (12)

Democratic Held Seats up for Election: 18 (out of 45 held)

The Brutal B – October 26, 2006: +3 to +6 D

Click here for individual Senate race analysis.

2006 House Crystal Ball HotRace Readings

Republican Held Seats in Play: 63 (169 Safe/Solid R)

Likely R (17) Leans R (18) Toss-up (16) Leans D (12) Likely D (0)
CA-04 (Doolittle) AZ-01 (Renzi) CT-02 (Simmons) AZ-08 (OPEN)
CA-50 (Bilbray) AZ-05 (Hayworth) CT-04 (Shays) CO-07 (OPEN)
CO-05 (OPEN) CA-11 (Pombo) FL-13 (OPEN) FL-16 (OPEN)
FL-08 (Keller) CO-04 (Musgrave) FL-22 (Shaw) IN-02 (Chocola)
FL-09 (OPEN) CT-05 (Johnson) IL-06 (OPEN) IN-08 (Hostettler)
IL-10 (Kirk) ID-01 (OPEN) IN-09 (Sodrel) IA-01 (OPEN)
IN-03 (Souder) KY-03 (Northup) KY-04 (Davis) NY-24 (OPEN)
IA-02 (Leach) MN-01 (Gutknecht) MN-06 (OPEN) OH-18 (OPEN)
KY-02 (Lewis) NV-02 (OPEN) NM-01 (Wilson) PA-06 (Gerlach)
NE-01 (Fortenberry) NV-03 (Porter) NC-11 (Taylor) PA-07 (Weldon)
NH-02 (Bass) NJ-07 (Ferguson) NY-26 (Reynolds) PA-10 (Sherwood)
NY-03 (King) NY-20 (Sweeney) OH-02 (Schmidt) TX-22 (OPEN)
NY-19 (Kelly) NY-25 (Walsh) OH-15 (Pryce)
NC-08 (Hayes) NY-29 (Kuhl) VA-02 (Drake)
PA-04 (Hart) OH-01 (Chabot) WA-08 (Reichert)
VA-10 (Wolf) PA-08 (Fitzpatrick) WI-08 (OPEN)
WA-05 (McMorris) TX-23 (Bonilla)
WY-AL (Cubin)
WV-01 (Mollohan)
VT-AL (OPEN)
TX-17 (Edwards)
SC-05 (Spratt)
IA-03 (Boswell) OH-06 (OPEN)
IL-08 (Bean) LA-03 (Melancon)
GA-12 (Barrow) IL-17 (OPEN)
GA-08 (Marshall) CO-03 (Salazar)
Likely R (0) Leans R (0) Toss-up (0) Leans D (4) Likely D (8)

Democratic Held Seats in Play: 12 (191 Safe/Solid D)

The Brutal B – October 26, 2006: +21 to +26 D

Click here for individual House race analysis.

2006 Governor Crystal Ball HotRace Readings

Republican Held Seats up for Election in 2006: 22 (out of 28 held)

Solid R (6) Likely R (5) Leans R (4) Toss-up (2) Leans D (3) Likely D (1) Solid D (1)
CT (Rell) AL (Riley) AK (OPEN) MD (Ehrlich) AR (OPEN) OH (OPEN) NY (OPEN)
HI (Lingle) CA (Schwarzenegger) ID (OPEN) MN (Pawlenty) CO (OPEN)
NE (Heineman) FL (OPEN) NV (OPEN) MA (OPEN)
SD (Rounds) GA (Perdue) RI (Carcieri)
VT (Douglas) SC (Sanford)
TX (Perry)
WY (Freudenthal)
TN (Bredesen)
WI (Doyle) NM (Richardson)
MI (Granholm) NH (Lynch)
OR (Kulongoski) ME (Baldacci) PA (Rendell) OK (Henry)
IA (OPEN) IL (Blagojevich) KS (Sebelius) AZ (Napolitano)
Solid R (0) Likely R (0) Leans R (0) Toss-up (2) Leans D (4) Likely D (2) Solid D (6)

Democratic Held Seats up for Election: 14 (out of 22 held)

The Brutal B – October 26, 2006: +4 to +6 D

Click here for individual Governor race analysis.