We have most of the states/districts we'll cover in this series planned out. But let me know if there are any 2026 races you'd like to see us write about.
I would love to see WA-3. Washington uses a top two primary which has had some interesting results in ‘22 and ‘24. The district skews R, but the Dem keeps winning. I can supply more info if needed.
Michigan seems like a fascinating state next year:
Dem Gary Peters is stepping down, so you have an open US Senate race in a purplish state.
Also, Gretchen Whitmer is term limited, so the governor spot is open, too.
Which means pretty much every Dem & Rep player of note is going to be running for 1 of those 2 spots. Lots of decisions on which race candidates join.
Add in Detroit Mayor, Mike Duggan, who has overseen Detroit's turnaround (not saying he should get all the credit, but, he certainly deserves some) is in the Governor's race as an Independent, which should make that even more interesting.
One thing I am surprised did not get mentioned is the Democrats probable advantage in a runoff. Runoffs skew more college educated, so there is a real argument that to beat Ossoff, Rs have to win in the first round. That said, I think people overstate a tad Ossoff's strength. I'm not convinced Split Ticket uses Ossoff's first round performance (when he underperformed Biden by about a 1.5 percent) into account, which is odd, as there may not always be a runoff, but the first round will always occur. I also think both the Georgia Senators get graded on a bit of a curve, given that every one of their wins were mostly due to Republicans shooting themselves in the foot. (Undermining faith in the voting machines, running Herschel Walker, etc) This race is probably Lean D, but that has more to do with the runoff and the probable national environment. Ossoff is a solid incumbent, but if the Rs run anyone halfway decent, I highly doubt you will see an overperformance much beyond anything attributable to the national environment.
Hmm, interesting points. A "spoiler" 3rd party in the general (like Hazel in 2020), presumably a right-libertarian or paleolibertarian whether or not running on the Libertarian Party label (or maybe a paleocon? Ballotpedia.org suggests no minor parties have ballot status in GA as of 2024).
Ossoff do rather worse than Biden (two party margin for both) in the 2020 1st round, but I also wonder how much of his underperformance relative to Biden in that race might have been low name recognition, something that he has partially overcome now.
Ossoff actually lost the first round by a little under 2 points, and due to how high profile his house race in 2017 was, he did not have low name recognition. (Thanks in part to the ads we got slammed with" I do agree though, that many of his weaknesses have largely been nullified by his tenure as an incumbent. He is not a weak candidate by any means, just one who is probably reliant on incumbency and national environment, which will probably be enough next year.
Agreed that his name recognition would have been rather higher than your average losing Congressional candidate (to your point about the profile of the GA-6 special), particularly in the Atlanta metro, but for marginal propensity voters 2020 versus the different distribution marginal voters of 2026?
The LP had ballot status in GA for years, but screwed up the last couple years (in many ways!).
Usually, the LP would run a candidate for a minor statewide office with no Democratic opposition (usually Sec of Ag), and get enough votes to qualify for ballot access the next time. Maybe the purpling of GA has pushed the LP out as Dems are seeing more seats as viable. That could hurt the Dems as the LP tends to steal more GOP votes; of course, that's moot if the Dems can win outright without a runoff and without the LP.
The vote to stop sending certain weapons to Israel is interesting. I don’t think it hurts him, does the Black middle class and Black churchgoers who are part the swing constituency in Georgia care about sending weapons to foreign countries? This is where moderates/centrists tell on themselves. It is true that Black voters are more socially conservatives, but they are not “moderate” in the way elites want them to be. If Ossoff can sell that vote not as concern for Palestinian casualties but as America First not wanting to send American taxpayers to a rich country accused of human rights violations, it shouldn’t be a problem. But I did Jewish orgs condemning him for that vote. How large is that? One of the most interesting dynamics I would love to see more detailed polling on, especially in a state like Georgia
The problem with this sort of analysis is the point you barely touched on and then ignored: “If they can’t win places like the Peach State, they probably don’t have a majority coalition, period.”
Focusing on Georgia as Georgia might win Georgia. But without paying more attention to the big picture, all that will do is make Georgia stop being the tipping point.
One thing I don’t see mentioned much is how weak the Republican senate candidates were in 2020, when Ossoff and Warnock first won. Perdue and Loeffler were no Herschel Walker, but both were…subpar to say the least.
This cannot be overstated. The counter is that Loeffler was not a long-time incumbent (appointed to replace the resigning Johnny Isakson*), but Perdue had served a full term.
*Fun fact: AFAIK, Isakson was the last Georgia Republican to run for a statewide office on a pro-choice platform, in 1996.
One thing to keep in mind is that there is significant daylight between the GAGOP and MAGA. Many of the statewide Republicans (Kemp, Carr, Raffensperger, etc.) have non-MAGA bona fides. That is why Kemp would be a game changer for the Senate race; former SoS Raffensperger would be the same, but less so, due to SoS being a lower office and probably some remaining MAGA grudges from '20. Carr has already announced for replacing term-limited Kemp for governor, so he's out for Senate. Raffensperger could also run for guv.
As an aside, the governor's race isn't looking too promising for the Dems; I haven't seen any numbers, but their bench looks weak--albeit somewhat but notably stronger than it has been in years. The leader is Keisha Lance-Bottoms, the embattled former Atlanta mayor. I bring this up because how it goes will impact GAGOP spending across the ticket.
Where the MAGA factor would come into play is with some of the lesser names, like some of the US reps like Carter that have been toadying up to Trump. MAGA giveth with GOTV of marginal GOP voters and MAGA taketh away from moderates.
What is surprising is that Gary Black, former Ag sec and the Team Kemp pick against Herschel Walker, hasn't thrown his hat in the ring. If the loony candidates like Walker stay out, then Black would have a fighting chance. He had good numbers and cross-party appeal, but the GA GOP electorate said GO DAWGS! instead.
Kemp is smart waiting it out. He could be the GOP's savior for POTUS if MAGA goes off the rails and Vance becomes a toxic heir apparent to Trump. He also gets a second chance at the Senate in '28 against Warnock--which could pay dividends if Trump 47 ends on a high note.
Besides incumbency, Ossoff's big advantages are out-of-state money and being a great retail politician--one of the best players among active pols. Of course, Ossoff is going to go where the Dem brand goes, for good or ill. He hasn't put much daylight between him and the DNC or the Omnicause, so that will hurt him with some moderates if the GOP stays sane.
No, obviously assigning new numbers of representatives affects the electoral college. I just didn't know the number of electors decision was called redistricting.
"State legislators and members of Congress that Robinson has spoken to simply don’t find the Senate seat appealing enough. Even though running for statewide office would be the next logical step in their careers, they 'just don't want to do it, because Congress is so broke that… many of them are like, why would I want to go do that? Nothing happens.'"
This worries me more than anything else here. Our democracy cannot continue to function when Congress is paralyzed. One thing that would help a lot is ambitious new Senators and Representatives from purple states and districts, impatient with gridlock and eager to make deals and get things done. But if things have gotten so bad that those people are opting out of even making a run... oof.
Rather than doing a simple average of Ossoff vs. the various GOP candidates, a weighted average might be more informative: weighing either by their support in primary polls or probabilities given by betting markets.
Where am I being dense? How could redistricting in 2032 possible affect anyone's Electoral College chances? Doesn't the Electoral College go by statewide results. with only a couple of non-meaningful exceptions?
Redistricting is the act of drawing new lines for all the districts in the state for the number of districts they get. So the first step is just "how many districts do I get?" which is based off the state population relative to the country.
Gerrymandering is the part of redistricting where you draw district lines that benefit your party over the other (for state reps). Obviously like you thought this does not affect the electoral college.
Florida is absolutely fucked in about five different (self inflicted) catastrophic ways, and it's already showing some incredibly negative trends. So I suspect that by 2032 it's gonna be experiencing a population crash. Texas, on the other hand, might get it's shit together. Or might not. But at least pockets of that state are trying for civilization still.
It's not clear to me whether this column is discussing Georgia as an indicator state or as a battleground state for the 2026 or 2028 elections. An indicator state is whose outcome would be helpful to know for predicting the national outcome. A battleground state is one in which influencing the outcome could significantly influence the national outcome. It's the difference between correlation and causation.
To see the difference, an old saying in American politics was, "As goes Maine, so goes the nation." From 1820 to 1932, when the Whig or later Republican party did better than average in Maine's September statewide elections, that party won the White House. When Whig or Republican party did worse than average, a Democratic President was elected in November. This pattern was noticed far back as 1852, if not earlier, and the saying was recorded in 1880, and the pattern persisted.
The major political parties treated this a causal effect and put disproportionate effort into Maine's September contests, even though Maine voted in November for President like all other states. Then in 1936, with Republicans winning Maine in a landslide but Republican Presidential candidate Alf Landon taking only 8 of 531 electoral votes, the saying changed to, "As goes Maine, so goes Vermont."
An indicator state is one close to the national tipping point. A battleground state is one close to a toss-up. The post starts out looking at Georgia's distance from the national tipping point, suggesting it's looking at Georgia as an indicator, but then delves into the idiosyncrasies of the specific races, candidates and local demographic and voting patterns, all of which reduce its value as an indicator. The best indicator states respond mostly to national forces that affect other states as well. Furthermore, the column seems to suggest that Georgia's indicator status makes it a strong battleground candidate.
For a simple statistical test--far too simple on too little data to be very meaningful--is to use a simple OLS regression on Georgia's distance from tipping point over time to test the assertion that the distance is shrinking. The coefficient is negative, as the post asserts, but both quantitatively and statistically nonsignificant. Georgia got 0.001% closer to tipping point per year, with a t-statistic of 0.05 (2 is a conventional level for statistical significance).
On the other hand, if we estimate battleground value by the product of Republican advantage times 100 minus Republican advantage, we get a t-statistic of 2.63 and a coefficient of 8.22 per year, suggesting that Georgia is moving quantitatively and statistically significantly closer to being a battleground state.
We have most of the states/districts we'll cover in this series planned out. But let me know if there are any 2026 races you'd like to see us write about.
I would love to see WA-3. Washington uses a top two primary which has had some interesting results in ‘22 and ‘24. The district skews R, but the Dem keeps winning. I can supply more info if needed.
I'd be very interested in learning if there are any potentially viable challenges for Collins in Maine.
Michigan seems like a fascinating state next year:
Dem Gary Peters is stepping down, so you have an open US Senate race in a purplish state.
Also, Gretchen Whitmer is term limited, so the governor spot is open, too.
Which means pretty much every Dem & Rep player of note is going to be running for 1 of those 2 spots. Lots of decisions on which race candidates join.
Add in Detroit Mayor, Mike Duggan, who has overseen Detroit's turnaround (not saying he should get all the credit, but, he certainly deserves some) is in the Governor's race as an Independent, which should make that even more interesting.
One thing I am surprised did not get mentioned is the Democrats probable advantage in a runoff. Runoffs skew more college educated, so there is a real argument that to beat Ossoff, Rs have to win in the first round. That said, I think people overstate a tad Ossoff's strength. I'm not convinced Split Ticket uses Ossoff's first round performance (when he underperformed Biden by about a 1.5 percent) into account, which is odd, as there may not always be a runoff, but the first round will always occur. I also think both the Georgia Senators get graded on a bit of a curve, given that every one of their wins were mostly due to Republicans shooting themselves in the foot. (Undermining faith in the voting machines, running Herschel Walker, etc) This race is probably Lean D, but that has more to do with the runoff and the probable national environment. Ossoff is a solid incumbent, but if the Rs run anyone halfway decent, I highly doubt you will see an overperformance much beyond anything attributable to the national environment.
Good point on the runoff! I probably should have covered that
Loved the Houston County mention, as someone who grew up there!
Hmm, interesting points. A "spoiler" 3rd party in the general (like Hazel in 2020), presumably a right-libertarian or paleolibertarian whether or not running on the Libertarian Party label (or maybe a paleocon? Ballotpedia.org suggests no minor parties have ballot status in GA as of 2024).
Ossoff do rather worse than Biden (two party margin for both) in the 2020 1st round, but I also wonder how much of his underperformance relative to Biden in that race might have been low name recognition, something that he has partially overcome now.
Ossoff actually lost the first round by a little under 2 points, and due to how high profile his house race in 2017 was, he did not have low name recognition. (Thanks in part to the ads we got slammed with" I do agree though, that many of his weaknesses have largely been nullified by his tenure as an incumbent. He is not a weak candidate by any means, just one who is probably reliant on incumbency and national environment, which will probably be enough next year.
You're right, I had a dyslexic moment :(
Agreed that his name recognition would have been rather higher than your average losing Congressional candidate (to your point about the profile of the GA-6 special), particularly in the Atlanta metro, but for marginal propensity voters 2020 versus the different distribution marginal voters of 2026?
The LP had ballot status in GA for years, but screwed up the last couple years (in many ways!).
Usually, the LP would run a candidate for a minor statewide office with no Democratic opposition (usually Sec of Ag), and get enough votes to qualify for ballot access the next time. Maybe the purpling of GA has pushed the LP out as Dems are seeing more seats as viable. That could hurt the Dems as the LP tends to steal more GOP votes; of course, that's moot if the Dems can win outright without a runoff and without the LP.
The vote to stop sending certain weapons to Israel is interesting. I don’t think it hurts him, does the Black middle class and Black churchgoers who are part the swing constituency in Georgia care about sending weapons to foreign countries? This is where moderates/centrists tell on themselves. It is true that Black voters are more socially conservatives, but they are not “moderate” in the way elites want them to be. If Ossoff can sell that vote not as concern for Palestinian casualties but as America First not wanting to send American taxpayers to a rich country accused of human rights violations, it shouldn’t be a problem. But I did Jewish orgs condemning him for that vote. How large is that? One of the most interesting dynamics I would love to see more detailed polling on, especially in a state like Georgia
The problem with this sort of analysis is the point you barely touched on and then ignored: “If they can’t win places like the Peach State, they probably don’t have a majority coalition, period.”
Focusing on Georgia as Georgia might win Georgia. But without paying more attention to the big picture, all that will do is make Georgia stop being the tipping point.
One thing I don’t see mentioned much is how weak the Republican senate candidates were in 2020, when Ossoff and Warnock first won. Perdue and Loeffler were no Herschel Walker, but both were…subpar to say the least.
This cannot be overstated. The counter is that Loeffler was not a long-time incumbent (appointed to replace the resigning Johnny Isakson*), but Perdue had served a full term.
*Fun fact: AFAIK, Isakson was the last Georgia Republican to run for a statewide office on a pro-choice platform, in 1996.
One thing to keep in mind is that there is significant daylight between the GAGOP and MAGA. Many of the statewide Republicans (Kemp, Carr, Raffensperger, etc.) have non-MAGA bona fides. That is why Kemp would be a game changer for the Senate race; former SoS Raffensperger would be the same, but less so, due to SoS being a lower office and probably some remaining MAGA grudges from '20. Carr has already announced for replacing term-limited Kemp for governor, so he's out for Senate. Raffensperger could also run for guv.
As an aside, the governor's race isn't looking too promising for the Dems; I haven't seen any numbers, but their bench looks weak--albeit somewhat but notably stronger than it has been in years. The leader is Keisha Lance-Bottoms, the embattled former Atlanta mayor. I bring this up because how it goes will impact GAGOP spending across the ticket.
Where the MAGA factor would come into play is with some of the lesser names, like some of the US reps like Carter that have been toadying up to Trump. MAGA giveth with GOTV of marginal GOP voters and MAGA taketh away from moderates.
What is surprising is that Gary Black, former Ag sec and the Team Kemp pick against Herschel Walker, hasn't thrown his hat in the ring. If the loony candidates like Walker stay out, then Black would have a fighting chance. He had good numbers and cross-party appeal, but the GA GOP electorate said GO DAWGS! instead.
Kemp is smart waiting it out. He could be the GOP's savior for POTUS if MAGA goes off the rails and Vance becomes a toxic heir apparent to Trump. He also gets a second chance at the Senate in '28 against Warnock--which could pay dividends if Trump 47 ends on a high note.
Besides incumbency, Ossoff's big advantages are out-of-state money and being a great retail politician--one of the best players among active pols. Of course, Ossoff is going to go where the Dem brand goes, for good or ill. He hasn't put much daylight between him and the DNC or the Omnicause, so that will hurt him with some moderates if the GOP stays sane.
No, obviously assigning new numbers of representatives affects the electoral college. I just didn't know the number of electors decision was called redistricting.
Apportionment
Thank you! The point of light I was looking for in the fog of my brain.
"State legislators and members of Congress that Robinson has spoken to simply don’t find the Senate seat appealing enough. Even though running for statewide office would be the next logical step in their careers, they 'just don't want to do it, because Congress is so broke that… many of them are like, why would I want to go do that? Nothing happens.'"
This worries me more than anything else here. Our democracy cannot continue to function when Congress is paralyzed. One thing that would help a lot is ambitious new Senators and Representatives from purple states and districts, impatient with gridlock and eager to make deals and get things done. But if things have gotten so bad that those people are opting out of even making a run... oof.
I'd love to see a compare-and-contrast of Georgia's Southern twin, North Carolina. While Cooper is a shoo-in, I'd love to see the rest of the math.
Rather than doing a simple average of Ossoff vs. the various GOP candidates, a weighted average might be more informative: weighing either by their support in primary polls or probabilities given by betting markets.
Where am I being dense? How could redistricting in 2032 possible affect anyone's Electoral College chances? Doesn't the Electoral College go by statewide results. with only a couple of non-meaningful exceptions?
Redistricting is going to change the number of electoral votes each state has
Thanks. I didn't realize that was called redistricting. I guess I was thinking of gerrymandering.
Redistricting is the act of drawing new lines for all the districts in the state for the number of districts they get. So the first step is just "how many districts do I get?" which is based off the state population relative to the country.
Gerrymandering is the part of redistricting where you draw district lines that benefit your party over the other (for state reps). Obviously like you thought this does not affect the electoral college.
Georgia is not projected to pick up any House seats. But Florida and Texas are.
Florida is absolutely fucked in about five different (self inflicted) catastrophic ways, and it's already showing some incredibly negative trends. So I suspect that by 2032 it's gonna be experiencing a population crash. Texas, on the other hand, might get it's shit together. Or might not. But at least pockets of that state are trying for civilization still.
It's not clear to me whether this column is discussing Georgia as an indicator state or as a battleground state for the 2026 or 2028 elections. An indicator state is whose outcome would be helpful to know for predicting the national outcome. A battleground state is one in which influencing the outcome could significantly influence the national outcome. It's the difference between correlation and causation.
To see the difference, an old saying in American politics was, "As goes Maine, so goes the nation." From 1820 to 1932, when the Whig or later Republican party did better than average in Maine's September statewide elections, that party won the White House. When Whig or Republican party did worse than average, a Democratic President was elected in November. This pattern was noticed far back as 1852, if not earlier, and the saying was recorded in 1880, and the pattern persisted.
The major political parties treated this a causal effect and put disproportionate effort into Maine's September contests, even though Maine voted in November for President like all other states. Then in 1936, with Republicans winning Maine in a landslide but Republican Presidential candidate Alf Landon taking only 8 of 531 electoral votes, the saying changed to, "As goes Maine, so goes Vermont."
An indicator state is one close to the national tipping point. A battleground state is one close to a toss-up. The post starts out looking at Georgia's distance from the national tipping point, suggesting it's looking at Georgia as an indicator, but then delves into the idiosyncrasies of the specific races, candidates and local demographic and voting patterns, all of which reduce its value as an indicator. The best indicator states respond mostly to national forces that affect other states as well. Furthermore, the column seems to suggest that Georgia's indicator status makes it a strong battleground candidate.
For a simple statistical test--far too simple on too little data to be very meaningful--is to use a simple OLS regression on Georgia's distance from tipping point over time to test the assertion that the distance is shrinking. The coefficient is negative, as the post asserts, but both quantitatively and statistically nonsignificant. Georgia got 0.001% closer to tipping point per year, with a t-statistic of 0.05 (2 is a conventional level for statistical significance).
On the other hand, if we estimate battleground value by the product of Republican advantage times 100 minus Republican advantage, we get a t-statistic of 2.63 and a coefficient of 8.22 per year, suggesting that Georgia is moving quantitatively and statistically significantly closer to being a battleground state.