Special Election Results

[Brian Langstraat]

The special election results for Georgia’s sixth congressional district are interesting from an election science perspective.

The voting system is a top-two runoff where: 

All candidates ran on one ballot, with a runoff election scheduled for June 20, 2017, for the first- and second-place finishers, if no candidate received 50% of the vote.

(https://en.wikipedia.org/wiki/Georgia's_6th_congressional_district_special_election,_2017)

The special election results can be seen here:

http://www.nytimes.com/elections/results/georgia-house-special-election-district-6

Jon Ossoff (D) with 48.1% and Karen Handel (R) with 19.8% will advance to the runoff election.  Ossoff has a 28.3% lead over Handel, but she may have a good chance, since the vote was 48.73% Democratic, 51.18% Republican, and 0.09% Independent.

How would have different voting systems affected this election?

What could the results of the runoff election be?

[Brian Langstraat]

My thoughts on my questions:

How would have different voting systems affected this election?

Instant-Runoff Voting (IRV) with only second or third choice allowed would have resulted in a victory for Ossoff, since Handel would not have been able to get enough transfer votes to win.

IRV with full choice allowed would have resulted in close finish between Ossoff and the top Republican, since the top Republican could have been able to get enough transfer votes to win.

Approval Voting (AV) would have resulted in close finish between Ossoff and the top Republican, since the top Republican could have been able to get enough approval votes to win.

Predicting the results for Range Voting (RV) would be difficult, since I am not sure how people felt about each candidate.  Perhaps, the enthusiastic scores for Ossoff would have outweighed the low scores from Republicans.

A non-partisan election using any of these voting systems would have resulted in a win for Ossoff, since the Republican resistance would not have been easily coordinated.

What could the results of the runoff election be?

Since special and runoff elections tend to lean toward the candidate with enthusiastic supporters, I predict that Ossoff will narrowly win.

However, Georgia's Republican establishment may be able to rally enough support around Handel to overcome the current 28% gap.

[Brian Langstraat]

On June 20th, Karen Handel (R) with 51.9% defeated Jon Ossoff (D) with 48.1%.

This result had very similar partisan results to the initial election:

Ossoff has a 28.3% lead over Handel, but she may have a good chance, since the vote was 48.73% Democratic, 51.18% Republican, and 0.09% Independent.

The runoff election had about a third more voters than the initial election.

From this result, I think that the following results would have occurred with a single election using various voting systems:

Ossoff win using IRV (with 2nd or 3rd choice) and any election system with non-partisan ballots.

Handel win using IRV (full choice), AV, and probably RV.

[parker friedland]

Great analysis of Georgia special election results:
https://www.youtube.com/watch?v=6qTF_gbtkdo

Further criticism of Ossoff:
https://www.youtube.com/watch?v=LJuzOqKjLVg

[parker friedland]

Both Rob Quist and James Thompson, the two democrats discussed in the video I linked above were form Bernie Sander's Our Revolution orginization:
https://www.youtube.com/watch?v=6qTF_gbtkdo

However that video did mention a third election where another candidate from Our Revolution won 58% of the vote in a New York State assembly election in a district that Trump won by 60% of the vote. That is a 42% swing! What an upset! No wonder the only election the establishment media was willing to cover was the Georgia special election. I wonder if CNN, MSNBC, or FOX, even dedicated a sliver of coverage to these other special elections.
http://www.newsday.com/long-island/politics/in-upset-democrat-pellegrino-wins-9th-district-assembly-seat-1.13656477

Well by now, it is 200% settled between Bernie's strategy and Hillary's strategy, that Bernie's strategy is by far more successful. The Clinton pundits who still argue that the Democratic Party still needs to be more "centrist" (which is a code word for taking more money from corporate overloads) are like the 3% of scientists that still argue that man made climate change isn't real.

[parker friedland]

I mean a 38% swing, not a 42% swing

[parker friedland]

https://www.youtube.com/watch?v=cr8EIqyCfg0
https://www.youtube.com/watch?v=11cjjIaA3DQ

[Phil Uhrich]

They completely ignored KS-5 until after the fact and the margin was so shocking, which then made them focus a bit more on MT-AL so that the republican could punch that neocon #Russiagate reporter in the face. And yes there is always a similar swing away from the white house in midterms. Which is why I voted for Trump after the corporate media and billionaire dem donors worked overtime to sink Bernie. With Clinton in Office we would be looking at a GOP supermajority in the senate and enough state legislatures for the GOP to amend the constitution on a party line vote after 2018 and president Cruz in 2020. I'll take 4 years of bumbling incompetence over that any day. I will be pissed if Clinton and her media sock puppets get him impeached or to change course on one of his very few sane proposals... Lets not start world war 3 and get nuked by Russia just because Israel and Saudi Arabia (the two leading sponsors of terror, propaganda, and TONS of $$$$ to DC) would really like us to regime change Syria so that Al Qaeda can take over the country. None of which you will hear about in in the NeoCon Times or the WarCriminal Post.
https://consortiumnews.com/2017/06/25/intel-behind-trumps-syria-attack-questioned/

[Brian Langstraat]

Interestingly, the Democratic candidates could have had a truly stunning victory by being more strategic.

With pre-election polls showing that Jon Ossoff (D) had less than a majority but still over twice as many potential votes as the next closest Republican candidate, he could have encouraged his supporters to vote for another Democratic candidate such as Ragin Edwards (D).

If between 41% and 59% of Ossoff's supporters voted Edwards, then both Democratic candidates would have advanced to the runoff election.

For example at 50%, the result would have been Jon Ossoff (D) with 24.3%, Ragin Edwards (D) with 24.3%, and Karen Handel (R) with 19.8%.

If 59% of Jon Ossoff's supporters found this strategy distasteful, then he still would have advanced to the runoff election.

Even if the Republican voters reacted to this strategy, then they would have had to quickly coordinate to nearly double Bob Gray's 10.8% of the Republican vote.

Assuming that the Democratic party was aware of this strategy, they attempted the unlikely quick victory of an absolute majority without needing a runoff instead of attempting a likely and truly stunning runoff between only Democratic candidates that would have guaranteed a Democratic congressman for Georgia’s sixth congressional district.

[Phil Uhrich]

I thought about that too, but If you can tell me exactly how many people he needed to text the night before and how many of them would have listened to pull that off than you should be out gambling somewhere. It's hard enough to get voters you know favor you to show up and vote, much less vote strategically for the other guy with very little room for error.

[Brian Langstraat]

Beyond just texts, he could have used his ad blitz on the last day or 2 to tell his supporters to vote for Ragin Edwards (D) or another "clone".

Assuming the ad blitz would reach about half of his supporters, he could tell everyone who saw the ad to just vote for the "clone".

It would not have been much of a gamble mathematically, since the following would have occurred:

0% to 40% supporters to "clone" = Jon Ossoff (D) and Karen Handel (R) in runoff

41% to 59% supporters to "clone" = Jon Ossoff (D) and "clone" (D) in runoff

60% to 100% supporters to "clone" = "clone" (D) and Karen Handel (R) in runoff

It would have been a small gamble politically, since the supporters may find it strategically distasteful and the Republican party could react effectively.

Worst case is Ossoff losing over 58% of his support and a nearly doubling Bob Gray's 10.8% of the Republican vote to result in Jon Ossoff (D) under 19.7%, Bob Gray (R) with 19.8%, and Karen Handel (R) with 19.8%, so there would be a runoff between only Republican candidates.

On the other hand, I heard that this one of the most educated districts in the country, so they could have spun this strategy as "intelligent" instead of "distasteful".

[Phil Uhrich]

You trust CD polls a lot more than I do. Besides, the Clintonite DNC 'Democratic Strategists' made their millions win or lose. If they advised him to do that and it backfired they would have a hard time finding their next democrat in republican drag that they can advise to chase suburban republican voters while ignoring poor people. Because that is an easier sell to their billionaire donors than wanting to help the vast majority of americans who have seen decreasing median lifetime earnings with every single cohort since Reagan put us back on the economic model that created the gilded age. Mission accomplished.

[Brian Langstraat]

If we assume that the smart 'Democratic Strategists' were aware of the "clone" strategy, then there could be another "conspiracy theory".

Per their analysis, the 'Democratic Strategists' had high confidence that the "clone" strategy would guarantee a Democratic winner.

But what kind of Democratic winner would there be?

In the runoff, very few Republican and right-leaning independent voters would care to show up, so the Democratic voters would essentially have a high-stakes Democratic primary.

John Ossoff or the "clone" candidate would pivot to much more Progressive positions and have a good chance of winning.

If the Progressive candidate won this election (and no other Democratic winners in major special elections), then it would prove that the only Progressive Democratic candidates could win.

In the upcoming Democratic primaries for the 2018 election, emboldened Progressive (Justice) Democratic candidates could win a majority of candidacies causing billionaire donors and the Clintonite DNC 'Democratic Strategists' to loss most of their power in the Democratic party.

Thus, the smart 'Democratic Strategists' would rather risk a Republican representative than risk a Progressive revolution!

[Jameson Quinn]

Seriously?

There were two possible ways to win in the first round: get over 50%, or get both of the top two slots. These were incompatible. They were close to achieving the first. Attempting the second, either from the start or as a last-minute thing, would have come with a set of risks — angering voters, provoking counterstrategy, or just getting it wrong. There is no need to posit anti-progressive conspiracies to explain the fact that they stuck with the first, even though it happened to fail.

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Brian Langstraat]

Phil,

You trust CD polls a lot more than I do.

I do not have 100% percent confidence in any polls unless they have a real confidence coefficient of 100% over their confidence interval.

I am not sure what each Congressional District poll uses for their confidence coefficient to find their confidence interval (margin of error), but I suspect it is around 95%.

Depending on their methodology, each CD poll could have decrease its real confidence coefficient to an estimated 85%.

However, the real confidence coefficient for multiple CD polls should increase as the number of CD polls with similar results increases to an estimated 90% with the 7 April CD polls (https://en.wikipedia.org/wiki/Georgia%27s_6th_congressional_district_special_election,_2017).

For the CD polls, my personal real confidence coefficient for is about 90% over a [34%,49.5%] confidence interval for Jon Ossoff's results.

I was fairly confident that Jon Ossoff would win with a non-majority plurality.

I was fairly confident that he had around 5% chances of not proceeding to the runoff and winning with a majority.

For the CD polls, what is your personal real confidence coefficient, confidence interval, and rationale?

Jameson,

There were two possible ways to win in the first round: get over 50%, or get both of the top two slots. These were incompatible. They were close to achieving the first.

They were not that close to getting over 50%.

The average of the polls in April was 42.6% with Ossoff's best poll (including the top margin of error) at 49.5% and worst poll (including the bottom margin of error) at 34%.

Actually, Ossoff impressively exceeded every poll's expectation with 48.1%.

Realistically, they should have expected narrowly missing 50% and hoped that Republican voters would not consolidate their support around Handel.

It should have been apparent from Trump's victory that Republican voters are good at consolidating support for pretty much any R candidate.

Attempting the second, either from the start or as a last-minute thing, would have come with a set of risks — angering voters, provoking counterstrategy, or just getting it wrong.

Those risks have been analyzed in earlier posts.

Is there political science research on the statistical risks of using a "clone" strategy?

From my personal perspective:

Angering Voters - If I was going to vote for Ossoff (and I wish I could have) and he told me to vote for a "clone" candidate to assure a Democratic runoff, then I would be happy that I supported a smart, unselfish candidate and vote for the "clone".  Perhaps, some indecisive voters would humorously vote for the "Clone Wars" candidate.

Provoking Counter-strategy - If I were a Republican and the Demorats tried an "An Attack of the Clones" ad blitz on the day before voting, then I would follow whatever instructions that were tweeted by Trump who was inspired by Fox News.

Just getting it wrong - What have Democratic strategists gotten right recently?

There is no need to posit anti-progressive conspiracies to explain the fact that they stuck with the first, even though it happened to fail.

I do not believe that there is some well-coordinated anti-progressive conspiracy, but there is unfortunate infighting between party factions.

For example, I voted for Bernie Sanders in the Democratic Iowa Caucus and think that the leaked DNC e-mails show behavior that is shameful but not conspiratorial.

I was using "conspiracy theory" as a framework to discuss a potential outcome of using the "clone" strategy.

Thus, the smart 'Democratic Strategists' would rather risk a Republican representative than risk a Progressive revolution!

becomes:

Thus, the Democratic strategists would rather play it safe than experiment with innovative strategies.

To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.