Sabato's Crystal Ball

Ratings Changes: Obama’s Debate Disaster Moves the Needle

Larry J. Sabato, Kyle Kondik and Geoffrey Skelley, U.Va. Center for Politics October 11th, 2012

Two months ago, we said that “barring a major blunder by either candidate,” the presidential debates were unlikely to be all that decisive.

After one debate, it’s fair to say that while President Obama didn’t make an obvious verbal gaffe during his first debate with Mitt Romney, Obama’s entire, listless debate performance can be characterized as a “major blunder.” And it’s costing him significantly in the race.

This is what a historically bad debate performance looks like: On Tuesday afternoon, Romney took the lead in the RealClearPolitics average of national horserace polls for the first time this calendar year by grabbing a tiny, less than one percentage point lead (Obama was up 3.1 points the day of the debate).

As of this writing, we do not have reams of credible, new information about the swing states, but in the days to come there will be many more surveys of the top states. Based on what we know now, however, we’re going to make a few changes to our maps.

We’ve long thought that in a close presidential race, Florida would likely end up in Romney’s column. Given that it was the president’s third weakest win in 2008, it naturally — along with more Republican Indiana and North Carolina — would be one of the states to fall out of Obama’s orbit in a much tighter race. So we’re moving Florida to Leans Republican, even though the polling there still indicates it is a toss-up. We’re also moving Virginia back from leans Democratic to toss-up. We know that the Obama campaign has long fretted about Virginia, understanding that the 2008 Obama vote was no predictor of 2012 success in the Old Dominion. There’s no longer any compelling justification that Obama has the advantage here. We could see it going to either candidate by a point or two or three.

Given that some reliable political science models have long predicted that this was going to be a close, competitive race, perhaps the debate just restored the race to its natural, tight pre-convention state.

David Paleologos, the veteran Suffolk University pollster, went so far as to say earlier this week that he would no longer poll Florida, North Carolina and Virginia because they were obviously moving to Romney. He might be right about North Carolina — we’ve had it rated leans Republican for more than a year — but Florida and Virginia, to us, still remain highly competitive, even if we’ve moved Florida to Romney’s column. We of course reserve the right to put states back in the toss-up column as the conditions on the ground merit, as we have done with Virginia.

Chart 1: Crystal Ball Electoral College rating changes

Map 1: Updated Crystal Ball Electoral College ratings



Notice that our map still has President Obama at 277 electoral votes, or seven more than he needs to win the election. Romney, now at 235 after our rating change in Florida, needs to pick off the remaining toss-ups — Colorado, New Hampshire and Virginia — to get to 261 electoral votes (of those, the Granite State is probably the hardest lift). Then Romney would need either Ohio or Wisconsin, or both Iowa and Nevada. Compared with the earlier map, Romney doesn’t need a giant transformation — and this should cheer the GOP.

At the same time, Obama maintains more — and more plausible — routes to 270 in the Electoral College. The states Romney needs to flip are ones we still think favor Obama currently. Our sources in Nevada and Ohio still believe their states will stay in the president’s column even if only barely, despite the new, tighter polls we have seen, and Iowa and Wisconsin’s traditional Democratic lean — the Badger State has gone Blue in the past six presidential elections, and the Hawkeye State has voted Democratic in five of the last six — cannot be discounted.

It is an understatement to say that Obama must significantly improve his performance in Tuesday night’s debate (how could he do worse?), and it is equally imperative that Romney keep the pressure on the president. Given that the debate will be a town hall, the format may be somewhat better for the president than the last debate. And keep this in mind: The press had an incentive before the first debate to report Romney as a winner in order to keep the race competitive. This time, though, much of the media may be inclined to give Obama the benefit of the doubt: After all, what’s better than a comeback story? Plus, after Obama’s abysmal showing, the expectations for him are far lower for this debate than last time, and the expectations for Romney much higher.

While this race has been topsy-turvy enough that it’s foolish to discount anything, we do not expect a big shift in the race out of tonight’s vice presidential debate. Maybe Paul Ryan can keep the ball rolling for the Republicans with a solid debate, and perhaps Joe Biden can lift Democratic spirits after Obama’s dud in Denver — or, possibly, one candidate will create a controversy that takes his party off-track at a critical juncture in the campaign’s final month.

Yet history shows that despite some memorable moments — such as the widely-viewed Sarah Palin-Joe Biden spectacle in 2008, John Edwards’ clumsy questioning of Dick Cheney over Cheney’s lesbian daughter in 2004, and Bob Dole’s “Democrat wars” gaffe in 1976 — VP debates don’t seem to change the campaign dynamic very much, at least according to polls taken by Gallup before and after the VP debates. Note that there were no VP debates in 1960 and 1980, even though those years featured some of the more memorable presidential face-offs.

Chart 2: Gallup presidential polling before and after vice presidential debates, 1976-2008

Sources: Nielsen (1976-2004, 2008), Gallup’s U.S. Presidential Election Center

Democrats’ Senate position remains strong

Although Obama’s lead in the polls has evaporated, this has not necessarily affected Senate races down-ballot. Take Virginia, for example. Despite the fact that Romney and Obama are essentially tied in the Old Dominion, two polls taken after the presidential debate last Wednesday both found Tim Kaine (D) leading George Allen (R) by seven points. While we don’t think Kaine’s lead is that big, the polling and what we’re hearing solidifies our view that the Virginia Senate race still leans Democratic, even if the state’s presidential race no longer does. (At the same time, if Romney gets a head of steam and ends up winning Virginia by at least several points, Allen can be pulled in by means of the coattail effect.)

Chart 3: Senate rating changes

We do have two ratings changes to make this week. The most notable shift comes in Massachusetts, where incumbent Sen. Scott Brown (R) and big-name challenger Elizabeth Warren (D) are locked in one of the most intense fights in the nation. Although we have had this race as a toss-up for months, we are now moving the race to leans Democratic. Our reasoning? Not only has Warren moved slightly ahead in the poll average (by 1.7 points according to RealClearPolitics as of Wednesday afternoon), but Massachusetts is positioned to vote so heavily Democratic for president that coattail could take down Brown on Election Day. A recent CBS News article estimated that Brown may need 300,000 voters to split their tickets on Nov. 6 for him to eke out a narrow victory — and that may be just a bit too much for Brown to hope for. If this were a midterm cycle and not a presidential year, it would be easier to imagine Brown holding on against Warren. But this isn’t a midterm — or a special election like the one Brown won in early 2010.

Our other change comes in Arizona, where we’re moving the race between Rep. Jeff Flake (R) and former Surgeon General Richard Carmona (D) from leans Republican to toss-up. While Flake still has a narrow lead in the poll average, Carmona has made the contest a real battle. The fact that both national party committees are now fully involved is a tell-tale sign. The National Republican Senatorial Committee is investing in the race to combat the $1.5 million its Democratic counterpart has already tossed into the fray. In the end, Flake probably has a clearer path to victory given the relatively Red hue of Arizona’s politics. But Carmona is certainly making things interesting, and a large Hispanic vote could upend conventional wisdom in this Senate matchup.

Nationally, this puts the Senate races solidly or leaning this way: 50 Democrats to 44 Republicans, assuming that all the favorites win and all the senators not up for reelection return next year. There are six remaining toss-ups: AZ, CT, IN, ND, NV and WI. At this point, Democrats might have the tiniest of edges in Connecticut and Wisconsin, while Republicans might have similarly tiny edges in the other four. The bottom line is that Republicans would need to win all the toss-ups and pick off one of the leaning Democratic seats in order to win the 51 seats they would need for a guaranteed majority. As of right now, our very tentative projection is a 52-48 Democratic majority in the Senate, or a net loss of one seat for the Democrats. As our readers know, we call every contest and keep updating and revising our projections until shortly before Election Day.

Map 2: Updated Crystal Ball Senate ratings map



In the House, watch for state waves

While no signs of a national House wave have emerged — Democrats hold tiny leads in national generic ballot averages — we are starting to see signs of what can best be described as state waves for Democrats in places where President Obama will roll to victory: California, Illinois and New York.

Chart 4: Crystal Ball House rating changes

Notice that we’ve moved two California races toward the Democrats: The open CA-26 to the Leans Democratic column, and CA-10 — held by Republican Rep. Jeff Denham — to the toss-up column. We believe we’re seeing a broad shift in the nation’s largest state — it has 12% of the country’s House seats — toward the Democrats, bolstered by the state’s new redistricting map, which made many seats more competitive. The important thing to note about the California seats is that of the nine listed on our competitive race charts — one leans Republican, two are toss-ups and the other six at least lean to the Democrats — President Obama would have won all nine seats in 2008. At this point, our ratings indicate a minimum Democratic pickup of two seats in California, with a maximum gain of five. That maximum gain is not out of reach, although Republicans have an outside shot to defeat Democratic Reps. Lois Capps (CA-24) and Jerry McNerney (CA-9) and thus limit their losses.

Democrats also have a chance to net five total seats out of Democratic Illinois. We continue to favor them to pick up at least four seats, as we have done for months, and there are now two toss-ups, Democrat-held IL-12 and Republican-held IL-13. Again, Democrats need to sweep all six to gain a net of five seats.

New York also appears to be a ripening possibility for the Democrats: We currently favor the Democrats to beat at least one Republican there — freshman Rep. Ann Marie Buerkle (NY-24) has been another consistent underdog in our ratings — and potentially net as many as four seats, although we do not think scandal-plagued Rep. Michael Grimm (R, NY-11) is actually all that vulnerable. Beyond Buerkle, the Democrats’ best opportunities remain freshmen Reps. Nan Hayworth (R, NY-18), who might be trending to the leans Democratic column before too long given the partisan makeup of her district, and Chris Gibson (R, NY-19), who we perhaps too generously moved to the leans Republican rating recently. Key to Democrats’ ability to put up crooked numbers in the Empire State is holding on to Rep. Kathy Hochul’s (D, NY-27) seat; she is locked in a total toss-up with ex-Erie County Executive Chris Collins in the most Republican district in the state. Republicans could still play to a draw in New York.

Maxing out in these three Blue states would probably push the Democrats into a double-digit seat gain nationally, even counting their likely losses in Redder territory. Our current projection is a Democratic gain of seven seats, up from six seats last week and four seats three weeks ago. Still — and we’ve been a broken record on this — Republicans are heavy favorites to keep control of the House.

Other ratings changes involve Rep. Dan Benishek (D, MI-1), who is now an underdog in his hotly contested Upper Peninsula seat, as well as a number of Democrats and Republicans who are now in stronger positions as their races move further away from the competitive toss-up and leans categories. One notable change here is in Washington’s open First District, where polls still show the race as competitive but where national Republicans have shown very little interest in competing. We believe ex-Microsoft executive Suzan DelBene (D) is a bigger favorite than she appears there.

Finally, we’re moving Rep. John Tierney (D, MA-6), perhaps belatedly, to toss-up. We had long resisted the move because even though Tierney is damaged goods — his wife and brother-in-law were involved in an illegal gambling ring, prompting questions about what he knew about his family’s legal problems — Massachusetts is so deeply Democratic. But the district isn’t quite as Democratic as the state as a whole, and former state Sen. Richard Tisei (R) is a good opponent.

To see our rating tables for all 435 House seats, click here.