Princeton Election Consortium

Innovations in democracy since 2004

Outcome: Biden 306 EV (D+1.2% from toss-up), Senate 50 D (D+1.0%)
Nov 3 polls: Biden 342 EV (D+5.3%), Senate 50-55 D (D+3.9%), House control D+4.6%
Moneyball states: President AZ NE-2 NV, Senate MT ME AK, Legislatures KS TX NC

The road leads back to you, sweet Georgia

October 26th, 2020, 9:20pm by Sam Wang

As I wrote the other day, federalism itself is on the ballot this year. Two notable events today with the Supreme Court demonstrate that point – and emphasize how important your activism is in shaping which way the nation goes.

Firstly, Amy Coney Barrett was sworn in as the ninth justice, 8 days before Election Day. There is no doubt that she will join the aggressive stance of the Court to curtail voting rights – in time to help out the President who appointed her.

The second event is an example of such an approach. In a 5-3 vote, the Court overruled a lower court, which would have allowed votes mailed on Election Day to be counted if received by a certain later date. In a footnote, Justice Kavanaugh took the radical step of citing Bush v. Gore, which until now has been considered a bit of a third rail, even to justices on the right wing of the Court. (Indeed, Chief Justice Roberts took the trouble of distancing himself from the footnote.) Basically, Kavanaugh said (speaking for himself, not the entire majority) that the Supreme Court can override state courts in the interpretation of state election laws.

This is a massive overstepping of black letter law. If adopted by a majority, it would reshape a fundamental principle of the American system of government, that of dual sovereignty between state and federal law. That’s a big deal.

However, there is something that could change the situation in 2021, and which you can affect.

Today I focus on Georgia.

Georgia has two Senate seats up for election: the regular race, Ossoff (D) vs. Perdue (R); and a multi-candidate special election featuring Warnock (D), Collins (R), Loeffler (R), and others. The races are quite close. The special election is certain to go to a January runoff. If neither Ossoff nor Perdue can get to 50% (could happen, there’s a third-party candidate), that will also go to a runoff. In our Senate power rankings, Georgia voters score a total of 24 for both races, making them tied for fifth place with nearby South Carolina.

What does this have to do with the Supreme Court? Court reform.

Next year will be a difficult one for passing any legislation at all under current rules, which include the legislative filibuster. The legislative filibuster requires 60 votes out of 100 to end debate. Over the last 20 years the filibuster has been used more and more as a routine maneuver. In 2021, health care, coronavirus relief, voting rights, and judicial reform will all be dead on arrival in the Senate – if the filibuster is retained.

Some Democratic Senators currently don’t want to weaken the filibuster. Therefore their party will need considerably more than 50 votes to change the rules. For that reason, every seat in the Senate matters. This is why in the PEC power rankings, we assign power to Senate races not based on chamber control, but based on the winnability of individual races. (Note that our power rankings are symmetric, in the sense that they apply equally if you are a Republican or a Democratic voter. The leverage per vote is the same.)

Based on current polls, six Senate races are within 3 points or less: Georgia’s two races, Iowa, South Carolina, Montana, and Alaska. The range of outcomes would be anywhere from 50 Democratic votes to 56 Democratic votes. That is a very wide range of futures for the Senate, ranging from the Vice-President having to hang around a lot to break ties, to a pad of six votes to do various things – including change the filibuster rule.

Therefore the votes of Georgians are powerful. Even if there’s a runoff election, you are highly advised to vote in this election, whether by mail, early or on November 3. The reason is that only people who vote in the first election will be allowed to vote in the January runoff.

Also, if you have problems with your mail-in ballot, there is a “cure” process to repair your ballot. Make sure to use mechanisms like this one to repair any problems with your ballot. Finally, if all else fails…vote in person on November 3!

Everyone else, get to work in your states. Money donations are probably useful if you give today or tomorrow – they will be in time for last-minute efforts. We have curated the highest-leverage opportunities in the left sidebar in the links to ActBlue (for Democrats) and WinRed (for Republicans).

Tags: Redistricting

16 Comments so far ↓

  • 538_Refugee

    I just read a fairly lengthy article on Politico about the early voter turnout in Georgia. Another article hinted turnout will be at percentage levels not seen since only white land owners were allowed to vote. I think some people have woken up to what can happen if you take a cycle off after being motivated to vote in 2008/12. Civil rights lost the Democrats the south for decades. Now, can it bring it back?

  • Some Guy Named Gabbo

    My apologies Sam, this is off-topic. I figured the most recent post would be the easiest way to catch your attention.

    Are you going to make a checklist of bellwethers for election night as you did in 2016? I found that very helpful.

    Also, TIL it’s “bellwether,” not “bellweather.”

    • Sam Wang

      That’s a good idea. Thanks for suggesting it. In 2016, that checklist, as well as a graphing tool I constructed, let me know fairly early in the evening that things were going Trump’s way. It was a lesson.

      Note that because of the delay in counting states such as Wisconsin, Pennsylvania, and Michigan, there are new challenges to address. I have to think how to do that…suggestions welcome.

    • Some Guy Named Gabbo

      Hmm, I found your 2016 election guide ( and it’s not how I remember it. I recall that it was a list of counties and local races to watch, but this guide is mostly about recognizing polling bias based on the actual results. Either my memory is off, or I remember something from a different site.

      Regarding reporting delays this year, you could just make a guide similar to what you did last time with the caveat that it applies over the coming days/weeks. No instant gratification, but I’d certainly be happy with that. Here are a few suggestions that could be helpful on election night:

      – To provide a little perspective, a (very) rough estimate of when we should start getting a realistic picture of each swing state’s results. I found this website that could be helpful: I’m not confident in my ability to infer much from it though.

      – A list of potential bellwether counties and congressional/state/local elections in states expected to have results on election night or the next day—Florida, North Carolina, Arizona, etc. We all know that if Biden wins Florida he’ll likely win the election, but which counties/elections give us an idea of where Florida may be heading?

      – In 2016 you noted to ignore early results from Virginia because it’s slow to count and would initially show a large Republican lead. Along these lines, a list of the order that each swing state releases their ballot results. e.g. “State ‘x’ releases early voting totals while in-person ballots are still being counted. Expect an early Democrat lead to tighten.” I’m not sure how easy this information is to come by though.
      I’ll chime in if I have any other ideas.

      On a different note, is Nebraska’s second district a potential bellwether for Biden in the Midwest? I’m oversimplifying, but if he wins NE-2 in an otherwise strongly conservative state it seems like it could be a good sign for other Midwest swing states, particularly Iowa. I ask about NE-2 in particular because, unlike the other Midwest states, Nebraska is already processing early ballots and starts counting them the day before the election ( Nebraska’s population is ~2 million (half in the Omaha metro area), so it seems possible we could have results there within a day or two, if not the night of, the election. I don’t know how quickly Nebraska traditionally counts ballots though, or how its urban/suburban area compares demographically with those in other Midwestern states. So my hypothesis is based on a lot of assumptions. Your thoughts?

    • Richard Wiener

      David Wasserman at Cook Political Report offers this bellwether for Florida

    • Some Gut Named Gabbo

      Thank you, Richard. Interesting article. I’ve heard about “The Villages” a few times this election, but not in this much detail.

    • CMeier

      Ohio closes early, so let me throw a couple of Ohio counties into mix as, um, bellwethers since Ohio will be close.

      1) Montgomery County, Dayton Area: Are the Ds getting the vote out?

      2012 137,139/51.66% for Obama and 124,841/47.03% Romney.

      2016 123,909/47.97% for Trump and 122,016/47.24% for Clinton.

      2) Delaware County, N burbs of Columbus, and one of the wealthiest counties in the U.S. What make this county interesting is that this county is giving money to Biden, measured by number of donors, at rate of 1.5-2 x the number of Trump donors. If the story that the burbs are deserting Trump is true, you will see it here.

      2012 37,292/37.84% for Obama, 60,194 /61.07% for Romney

      2016 57,568/55.43% for Trump, 40,872 /39.35% for Clinton.

    • Some Guy Named Gabbo

      Thank you for the insight CMeier. I’ll add them to my list for Tuesday.

  • ArcticStones

    Very encouraging that as of yesterday, 2.97 Georgians have voted early. That amounts to a stunning 73 % of the entire 2016 vote, and more than 40 % of all registered voters.

    Also fascinating is that over 55 % (!) of these early voters are women – and we know that women to a much greater degree are voting for Biden and down-ticket Democratic candidates.

    I am cautiously optimistic about Georgia, despite the tricks that Governor Kemp & Co have pulled and are pulling.

    • ArcticStones

      Stephen Fowler has made an excellent precinct-level (!) map of Georgia’s turnout levels so far. (Hope I am posting the link correctly.)

      Imagine if Democrats can weaponize their voters: i.e. recruiting most who have already voted for GOTV of friends, family and acquaintances who have yet to vote.

      That would result in increased Democratic margins in this final stretch – and on Election Day – eliminating the risk of Republicans “winning the day”.

    • ArcticStones

      * 2.97 million

  • ArcticStones

    Stephen Fowler has made an excellent precinct-level (!) map of Georgia’s turnout levels so far. (Tried unsuccessfully to post the link, but you can find it on his Twitter feed.)

    Imagine if Democrats can weaponize their voters: i.e. recruiting most who have already voted for GOTV of friends, family and acquaintances who have yet to vote.

    That would result in increased Democratic margins in this final stretch – and on Election Day – eliminating the risk of Republicans “winning the day”.

  • Joseph Bland

    Hmm. Are there any propositions in Georgia that are likely to draw out more folks if they got a bit more money? May be too late for this year, but worth thinking about for future elections….

  • jonW

    Sorry to append an off-topic comment here: has the Senate forecasting model changed drastically in the last 2 or 3 days? Or is it just that the older polls are dropping off more rapidly now we’re in the home stretch? I had been comforting myself every day by seeing that 56, even 57 seats for Dems in the histogram was still more likely than 49 Dems. But the histogram has undergone a huge change since the weekend even though the meta-margin hasn’t changed that much.

  • Joseph Bland

    My online research showed nothing useful for Georgia, but I did fine a very useful one for Alaska!

  • CMeier

    One point about the filibuster. Prior to 1970, the Senate could have only a single piece of business pending at one time. Thus a filibuster stopped everything that was ready to be brought to the floor until the filibuster was stopped by a cloture vote or the Senate agreed to move on. But after the filibusters of the 1960s Civil Rights legislation, a “reform” of the filibuster let the Senate have more than one piece of business pending on the floor at one time. The result was that a Senator need merely threaten a filibuster and the Senate would shrug and move on to whatever else was pending.

    The Senate could return to the old rule of one item at a time on the floor and keep the filibuster which would have the effect of returning the filibuster to its old role as a nuclear option to be used only when a piece of legislation is regarded as particularly heinous by a significant minority.

Leave a Reply to ArcticStones (Cancel)