Thanks for a great article, Lucas Holtz. The New Rural Project has been doing a lot of work since 2021 to provide information and GOTV in seven rural counties east of Charlotte [Union, Anson, Richmond, Scotland, Moore, Hoke and Robeson]. I'm hoping their efforts pay off for Harris-Walz.
Also, what is your impression of the effect of Hurricane Helene will have on western NC in the upcoming election?
Thank you so much, Jane! Big fan of the New Rural Project!
Not entirely sure what the effect will be on turnout just yet, but we're likely to see *some negative impact. The NC elections board is still figuring out the logistics of re-opening the elections offices in Buncombe County and the other impacted counties. The silver lining is no voting equipment or ballots were damaged in the storm, so the faster that resources are rushed in to help folks recover and that those offices are re-opened the better!
I don't want to see anyone deprived of their right to vote in this election. Part of the problem, as you likely know, is that the NC Supreme Court ruled that RFKjr could have his name removed from the ballot, requiring the state election board to reprint millions of ballots. Mail-in ballots were supposed to be sent out by September 6th, but they didn't go out until the 24th, and many were probably lost in the storm. I've heard that voting machines weren't damaged, but many of the polling places have been. Also, this year NC requires voter ID to cast a ballot. It's likely a lot of people won't have valid ID. We're going to need some emergency measures to make sure folks can vote. In-person, early voting starts in 2 weeks. Fingers crossed.
I loved this article, thank you Lucas! I’m with you and Mike on this. I think Harris had a real shot of winning North Carolina for all the reasons you mentioned. I think the combination of Trump and Mark Robinson madness negatively impacts their voter turnout in the same style James Comey’s email announcement helped to drove down voter turnout for Hillary in 2016. I hope everyone impacted by Hurricane Helene will be okay. It’s an awful tragedy. I hope people all across every impacted area still get a chance to vote.
One request: Maybe it's because I have night view on, but the map with the circles doesn't make any sense because the shades of blue or red are too similar, and the meaning of the color shift is not intuitive. Perhaps for the right shift and left shift circles, keep the same color but use diagonal lines within the circles (leaning right or left) instead of that slight change in color.
Thank you, Linda! Good suggestion! I was tinkering with the map a bit due to the dark mode. Here's a link to it in day mode as well https://datawrapper.dwcdn.net/pRpub/10/
Part of the problem is that I'm finding it hard to keep the concepts clear in my brain, lol. Your analysis is so original, which is what's wonderful about it, but my brain is practically hard-wired with the old ways of looking at things.
Looking at the day mode link, the colors on the map don't match the colors on the key, I think. Maybe because they are transparent? Is there a way to make them opaque?
Feel like I have to quibble with your analysis even though I'm just a rando with access only to public data. Yes, I also believe that Harris will win North Carolina but want to keep things simple in analysing.
Thanks for the post.
First the quibbles:
1. Not sure why you just mention Harris vote share vs Biden (or somebody elses vote share). Trump underperforming his margins in this election polls vs his 2020 results matters as well. For example, the white college result for 2020 is Biden 47 - Trump 52 (Trump +5) whereas the current margin taken as average for this category in polls is Harris 55 - Trump 44 (Harris +11). That's a 16 points improvement
2. According to NC SOS data from 5 Oct women were 49.47% RV. It is true that in 2020 election women were 49.6 of registered voters and men were 42.1 of registered voters but as the proportion of people who actually voted it was women 50.4 and men 41.4. In other words not only is it the gender split in NC higher than in other states amongst registered voters but in 2020 76.5% women RV voted and 73.8% men RV voted resulting in that 9 point gender difference.
3. I respect your using the angle of 'moderate' voters but it seems so vague and prone to changing
Here's what I see winning pathways for Harris:
1. Shares for over 65y in 2020 was Biden 40 - Trump 59 (Trump +19). Current margin from average polls is Trump +1 (Harris 48- Trump 49). In 2020 over 65y were about 25% RV but were about 28% people who actually voted (over 65 are people who consistently vote in elections and they vote at the highest percentage of their RV - around 85%).
If you then take then next age category, the 40-64y, this category represents about 41% RV but were 44% of the people who actually voted in 2020. This category Biden 49 - Trump 49.5 in 2020 and current margins from poll average is Harris 47.5 - Trump 49. This category votes at 82% of its RV sample and in 2020 represented 44% people who voted.
So far from this two age categories you've got 66% RV.
If you take current RV of 7704452 (as per NC SOS 5 Oct) and assume same turnout as 2020 (these are pretty reliable categories in terms of turnout) and then apply the current Harris -Trump margin for each category you get at Harris 1992891 voters and Trump 2047468 (so that's Trump +54577 votes).
Finally you get to the 18-39 category which represents about 38% RV but in general elections only about 62% of these voters actually vote. The good news is that in 2020 the shares for this age category were about Biden 57- Trump 40 and current margins are about Harris 49 - Trump 43 (and this is the conservative take on it). So if we assume the same turnout for this age category as in 2020 (which I'd say is unlikely given the youth enthusiasm for Harris) and apply that conservative 49-43 margin you get at Harris 889432 and Trump 780522 (Harris +108910) for this last age category.
So in total you have Harris 2882323 - Trump 2827990 (Harris +54333)
I trust older voter to have made their mind up (so poll margins to be correct) and to turn out to vote as reliably as they done in the past. And I trust that margin for younger voters to actually be better for Harris than polls suggest but anyway she wins if it's correct
I can see an analysis based on white voters with a college degree (where Harris is significantly overperforming Biden as mentioned above) but you're then left guessing about the white voters without a college degree (where Harris vs Trump margins are as Biden vs Trump) and here I think it's a fair assumption to say that if there's a category where Trump over-performs or finds new voters, it will be in this category. That's why the analysis based on W-college & W-non college is problematic for me.
ALTERNATIVELY....the other angle to look at splits per urbanicity.
NC RV split into urban 12.6%, suburban 40% and rural 47.4%.
From the polling averages Harris is significantly over-performing Biden in the suburban category. 2020 was Biden 39 - Trump 60 (Trump +21) and current margin is Harris 50 - Trump 47 (Harris +3). That's a 24 point move towards Harris.
The current margins for urban and rural are similar to 2020 but very possible that Harris will over-perform in urban and Trump will under-perform in rural (sadly Helene might contribute to this).
With regards to the 317439 current Latino RV in NC (4% of RV), even if Harris wins them all not sure it will be enough given the 11% under-performace with Blacks (compared with Biden 2020).
Anyway, sorry for the long post, NC is an obsession currently.
Thanks for a great article, Lucas Holtz. The New Rural Project has been doing a lot of work since 2021 to provide information and GOTV in seven rural counties east of Charlotte [Union, Anson, Richmond, Scotland, Moore, Hoke and Robeson]. I'm hoping their efforts pay off for Harris-Walz.
Also, what is your impression of the effect of Hurricane Helene will have on western NC in the upcoming election?
Thank you so much, Jane! Big fan of the New Rural Project!
Not entirely sure what the effect will be on turnout just yet, but we're likely to see *some negative impact. The NC elections board is still figuring out the logistics of re-opening the elections offices in Buncombe County and the other impacted counties. The silver lining is no voting equipment or ballots were damaged in the storm, so the faster that resources are rushed in to help folks recover and that those offices are re-opened the better!
I don't want to see anyone deprived of their right to vote in this election. Part of the problem, as you likely know, is that the NC Supreme Court ruled that RFKjr could have his name removed from the ballot, requiring the state election board to reprint millions of ballots. Mail-in ballots were supposed to be sent out by September 6th, but they didn't go out until the 24th, and many were probably lost in the storm. I've heard that voting machines weren't damaged, but many of the polling places have been. Also, this year NC requires voter ID to cast a ballot. It's likely a lot of people won't have valid ID. We're going to need some emergency measures to make sure folks can vote. In-person, early voting starts in 2 weeks. Fingers crossed.
I loved this article, thank you Lucas! I’m with you and Mike on this. I think Harris had a real shot of winning North Carolina for all the reasons you mentioned. I think the combination of Trump and Mark Robinson madness negatively impacts their voter turnout in the same style James Comey’s email announcement helped to drove down voter turnout for Hillary in 2016. I hope everyone impacted by Hurricane Helene will be okay. It’s an awful tragedy. I hope people all across every impacted area still get a chance to vote.
Send more Lucas Holtz!
Yes, more Lucas Holtz, please!
One request: Maybe it's because I have night view on, but the map with the circles doesn't make any sense because the shades of blue or red are too similar, and the meaning of the color shift is not intuitive. Perhaps for the right shift and left shift circles, keep the same color but use diagonal lines within the circles (leaning right or left) instead of that slight change in color.
Thank you, Linda! Good suggestion! I was tinkering with the map a bit due to the dark mode. Here's a link to it in day mode as well https://datawrapper.dwcdn.net/pRpub/10/
Part of the problem is that I'm finding it hard to keep the concepts clear in my brain, lol. Your analysis is so original, which is what's wonderful about it, but my brain is practically hard-wired with the old ways of looking at things.
Looking at the day mode link, the colors on the map don't match the colors on the key, I think. Maybe because they are transparent? Is there a way to make them opaque?
Any data geek approved by Mike is top shelf in my book. Thanks for this, Lucas. Hope to see more from you here in the MD community.
More GOTV staffers arriving in NC as we speak….
Feel like I have to quibble with your analysis even though I'm just a rando with access only to public data. Yes, I also believe that Harris will win North Carolina but want to keep things simple in analysing.
Thanks for the post.
First the quibbles:
1. Not sure why you just mention Harris vote share vs Biden (or somebody elses vote share). Trump underperforming his margins in this election polls vs his 2020 results matters as well. For example, the white college result for 2020 is Biden 47 - Trump 52 (Trump +5) whereas the current margin taken as average for this category in polls is Harris 55 - Trump 44 (Harris +11). That's a 16 points improvement
(white college margins taken from Cook Political Swingometer https://bit.ly/3BAmwei and the average in polls from the categories breakdown tracker I keep https://docs.google.com/spreadsheets/d/1ac2x234WHYxHt784ajQveNOkD07v8oFChcLHY5UZEEI/edit?usp=sharing)
2. According to NC SOS data from 5 Oct women were 49.47% RV. It is true that in 2020 election women were 49.6 of registered voters and men were 42.1 of registered voters but as the proportion of people who actually voted it was women 50.4 and men 41.4. In other words not only is it the gender split in NC higher than in other states amongst registered voters but in 2020 76.5% women RV voted and 73.8% men RV voted resulting in that 9 point gender difference.
(data taken from NC SOS turnout statistics for 2020 https://bit.ly/3Nfmf3a)
3. I respect your using the angle of 'moderate' voters but it seems so vague and prone to changing
Here's what I see winning pathways for Harris:
1. Shares for over 65y in 2020 was Biden 40 - Trump 59 (Trump +19). Current margin from average polls is Trump +1 (Harris 48- Trump 49). In 2020 over 65y were about 25% RV but were about 28% people who actually voted (over 65 are people who consistently vote in elections and they vote at the highest percentage of their RV - around 85%).
If you then take then next age category, the 40-64y, this category represents about 41% RV but were 44% of the people who actually voted in 2020. This category Biden 49 - Trump 49.5 in 2020 and current margins from poll average is Harris 47.5 - Trump 49. This category votes at 82% of its RV sample and in 2020 represented 44% people who voted.
So far from this two age categories you've got 66% RV.
If you take current RV of 7704452 (as per NC SOS 5 Oct) and assume same turnout as 2020 (these are pretty reliable categories in terms of turnout) and then apply the current Harris -Trump margin for each category you get at Harris 1992891 voters and Trump 2047468 (so that's Trump +54577 votes).
Finally you get to the 18-39 category which represents about 38% RV but in general elections only about 62% of these voters actually vote. The good news is that in 2020 the shares for this age category were about Biden 57- Trump 40 and current margins are about Harris 49 - Trump 43 (and this is the conservative take on it). So if we assume the same turnout for this age category as in 2020 (which I'd say is unlikely given the youth enthusiasm for Harris) and apply that conservative 49-43 margin you get at Harris 889432 and Trump 780522 (Harris +108910) for this last age category.
So in total you have Harris 2882323 - Trump 2827990 (Harris +54333)
I trust older voter to have made their mind up (so poll margins to be correct) and to turn out to vote as reliably as they done in the past. And I trust that margin for younger voters to actually be better for Harris than polls suggest but anyway she wins if it's correct
I can see an analysis based on white voters with a college degree (where Harris is significantly overperforming Biden as mentioned above) but you're then left guessing about the white voters without a college degree (where Harris vs Trump margins are as Biden vs Trump) and here I think it's a fair assumption to say that if there's a category where Trump over-performs or finds new voters, it will be in this category. That's why the analysis based on W-college & W-non college is problematic for me.
ALTERNATIVELY....the other angle to look at splits per urbanicity.
NC RV split into urban 12.6%, suburban 40% and rural 47.4%.
From the polling averages Harris is significantly over-performing Biden in the suburban category. 2020 was Biden 39 - Trump 60 (Trump +21) and current margin is Harris 50 - Trump 47 (Harris +3). That's a 24 point move towards Harris.
The current margins for urban and rural are similar to 2020 but very possible that Harris will over-perform in urban and Trump will under-perform in rural (sadly Helene might contribute to this).
With regards to the 317439 current Latino RV in NC (4% of RV), even if Harris wins them all not sure it will be enough given the 11% under-performace with Blacks (compared with Biden 2020).
Anyway, sorry for the long post, NC is an obsession currently.