Thursday, November 6, 2008

The Georgia Senate Runoff

There hasn't been a whole lot of talk around here during this cycle devoted to races other than the presidential race. [I don't think there are too many people that are complaining about this.] However, with the Senate race here in the Peach state heading for a December 2 runoff between incumbent Saxby Chambliss and Jim Martin, I thought it appropriate to shift the focus to the one remaining high profile race anyone's campaigning for.

[It certainly isn't the only undetermined race at this point. There will be a recount in Minnesota's senate race and the Ted Stevens' situation in Alaska makes that one worth watching if only for more speculation about who would potentially fill his shoes if he were to win and be forced out of office. Sarah Palin, I'm looking in your direction. The former VP choice on the GOP side won't have direct appointment powers on a replacement because the two conflicting laws Alaska has on the books call for a special election within 90 days. However, what is not known is if the governor has the power to appoint someone on an interim basis for that period of time. We'll have to hold off on that speculation for now, but part one -- Stevens winning -- looks likely.]

But back to Georgia...

So what do we know about this race? I could tell you, but I better show you with a map first.

[Click Map to Enlarge]

Sure, that doesn't tell you anything more than you already knew. Chambliss spent the evening of November 4 watching his percentage in the vote returns creep closer and closer to the 50% plus one vote mark that the candidates had to avoid in order to prevent a runoff. And the incumbent Republican missed it by .2%. That aside, though, Georgia likely won't become the center of the political universe for the next month since the Democrats won't get to 60 seats in the Senate, and unless the Minnesota recount overturns the apparent result -- and I can't think if a case where a recount led to a anyone other than the original projected winner winning -- then the best the Democrats can hope for is 59. And Ted Stevens will have something to say about that.

Like Al Franken in that recount in Minnesota, Jim Martin will have a difficult time getting over the hump in Georgia. But let's talk a little about where the former state senator will have to do well between now and December 2. The first thing we can do is look at where the race was close but favored Chambliss on Tuesday. But let's filter that through where Lt. Governor Mark Taylor did well in 2002. Why Mark Taylor and not Max Cleland? To start, Taylor won in 2002 when Cleland did not. But Taylor was also the last Democrat to win a statewide office this high. Where the former lt. governor did well six years ago -- in a Republican-leaning election -- would add quite a few more counties to the map, but when you factor in how well Martin did on Tuesday in some of those south Georgia counties, you only end up with a handful of additional areas to potentially target. If you look at the map below, those are the counties in white. And all of them share a border with a county that went for Martin (except Turner County which borders another potential target county, Ben Hill County).

[Click Map to Enlarge]

Well, let's not leave that 2002 Cleland-Chambliss senate race out of the equation altogether. We can add one other layer to this by asking where Chambliss won on Tuesday that he did not six years ago. Again, we can add a few more counties to the list (the ones in gray above), but none of them, other than Seminole County in the far southwest corner of the state were within 13 points on Tuesday night. In other words, they just aren't viable targets for Martin.

The flip side of the coin on this is that there are also areas where Martin outperformed Cleland and could be vulnerable in the runoff. Oddly enough, there are six counties in this category (those in light blue) to counterbalance the six counties where Chambliss exceeded his own numbers from 2002. That is somewhat problematic for Martin and throws it back to those white counties. The problem there is that while there were seven close counties that favored Chambliss a couple of days ago, there were 18 close counties where Martin edged out the incumbent. And as we saw in the presidential race, if any momentum develops toward the end of the race, the potential that all the close areas break for the momentum-possessing candidate increases. By that measure, Martin clearly had some level of momentum on Tuesday; if only Obama's coattails.

But Martin won't have those coattails on December 2, but he will have to face the challenges described above as well as to overcome what is likely to be a rather significant drop in the turnout rate. The last time that there was a Senate runoff in Georgia under this 50% plus one vote law was the Wyche Fowler-Paul Coverdale race in 1992. Fowler, an incumbent Democrat, won the first round, but lost to Coverdale in the runoff when turnout decreased by 44.31%. Not only will Martin have to gain ground on an incumbent in some of the counties above (and likely more), but he'll have to get out the vote in a more efficient way than Saxby Chambliss. Paul Coverdale did come from behind, but for Martin, the state isn't trending toward the Democrats in 2008 the way it was moving toward the Republicans in 1992.

President-elect Obama could still be the wildcard here. But will the incentive be there to intervene without a 60th, filibuster-proof seat on the line?

Recent Posts:
Obama is the Unofficial Winner of North Carolina

More on North Carolina: UPDATE

What's the Matter with North Carolina?


Jack said...

Thanks for the analysis of GA-Senate. I'm following these races too closely - still spending half my time staring at numbers as they slowly change for some of the House races still too close to call - so it's always great to get more info. I know it'll be tough for Democrats to win, but I do think there will be plenty of incentive for Obama to help out - a senate seat is a senate seat, 60 or not.

And to address two other points you made:

As for Alaska, I'm pretty sure Palin could make an appointment. After all, Frank Murkowski, upon becoming governor of Alaska, resigned his senate seat and appointed his replacement. That was Lisa Murkowski. It's very unfortunate he did that, because that greatly damaged Frank's reputation, leading to the rise of Sarah Palin. I don't know, however, if laws have been changed since then.

Also, recounts have led to a change in winner in at least one statewide election - though it certainly was not a high-profile one.

Jack said...

Update: laws in AK have clearly changed, as Lisa Murkowski did not have to stand for election for two years, until the regularly scheduled AK general election, in which she beat Tony Knowles (who also lost to Palin in 2006).

Anonymous said...

I'll have to find the link I had on the Alaska law. Palin can appoint someone, but that could be challenged in court because of the conflicting laws. But one of the laws would be struck down and the appointment would likely stand.

Unknown said...


I definitely agree about "a Senate seat is a Senate seat". Furthermore, if this is truly a change election, time spent campaigning there could be a sort of investment in the future. It would be very advantageous to the Democrats if they were able to make inroads into the South, and this would be a good way to do it.

Besides, even with preparations to be president, he's probably got more time now than when he was campaigning, so he could afford the time to do it.

Robert said...

Obama needs to be careful not to be too partisan. Showing up at a few well-chosen locations in the state, thanking all his supporters for their votes, and saying that he would like to have Jim Martin on his team to help change Washington. At the same time, key operatives from places like NC and FL will be aggressively organizing the GOTV. I'm expect taht Scott will weigh in saying that Obama will have just the right touch.

Jack said...


I don't think there's any danger of being too partisan. It's normal practice for presidents to campaign for members of their own party — Bush campaigned for Republicans in midterm elections during his presidency — and certainly it's no different for presidents-elect. Of course, he shouldn't spend all his time in Georgia, but one or two appearances should be fine.

I have faith that Martin can win for three reasons: Bill Foster, Travis Childers and the late (politically speaking) Don Cazayoux.

Shark Girl said...

Does Martin accept PAC money from Robins Air Force Base contractors? I'm wondering what his stand is on corruption at the base.

I already know where Chambliss stands. His staff refused my complaint and said they depend on funding from a certain person, who is a key defendant in my cases.

Another member of his staff also publicly mocked me. I traced the IP number and got the MAC address from one of Chambliss' computers located in the Atlanta office.

I'd like to know if ANY politician has the guts to stand up to the corruption at Robins.

Unknown said...

Interesting read, but I must say lets not lose sight of the fact that candidates are not trying to win counties. They are trying to win votes. The population and the margins of victory of the counties are very important and mostly ignored here.

In presidential elections we look at different states, because of the electoral college, but within a state looking at how counties vote only allows us to examine geographic trends, but in this analysis it offers very little in analyzing the runoff.

Anonymous said...

That's right, Clay. There is no electoral college in Georgia. Other than Henry County, the highlighted counties aren't going to offer many votes to really swing the runoff one way or the other.

The exercise was more to identify those areas that have fluctuated between the parties in recent elections and to show how areas that, should there be turnout that favors one side over the other, could tip the election toward one of the candidates across many of these closer counties. I think we can agree that those closer counties, in a wave scenario, are more likely to break toward one candidate versus some of the less competitive counties. And while this isn't an electoral college system, the cumulative effect of winning all of those counties, say, could make a or the difference on December 2.