Guide to the Great Election Spreadsheet
For the last few months, I've been working on a spreadsheet-- first in LibreOffice, then on google docs --that shows what I consider to be the top elections happening next week in the US. It includes(or at least intends to include) details about all the elections for the US House of Representatives, the US Senate, and all state Governors.
The sheet currently has four sub-sheets: House Swing View, Senate, Governor, and Polls Close. I'll explain these below.
House Swing View: First some background. Skip to the TLDR paragraph if you don't feel like reading all this. The Partisan Voting Index, compiled in July by the Cook Political Report, is widely regarded as a reliable source for the political leaning of any given Congressional district. My district for example, Colorado's 1st, has a rating of D+29, meaning that it is believed that an election in my district would most likely end with the Democratic candidate winning by 29 percentage points over the Republican candidate. Most districts are not nearly that one-sided though. As this sheet details, over 100 are +/- 10 points. After every national census, every Congressional district is evaluated, and often redrawn, based(ideally) on population change since the previous redrawing. This process was completed in 2022, following the 2020 census. When it was completed, or at least when it looked much like it does today, the Cook Political Report predicted the Republican Party would retake the House of Representatives simply by virtue of the redrawing that had taken place. The PVI was released in July, substantiating the claim that 220 districts lean toward the GOP, while only 208 lean toward the Democratic party, with the remaining 7 districts in a statistical tie.
However, something happened between redistricting and the release of the 2022 PVI. Something big. Something seismic. The Supreme Court overturned the Roe v. Wade, eliminating the federal protection of abortion rights in the US. The PVI did not reflect this at all in the July report-- or possibly just assumed it would have no effect on the landscape. I think that was a critical oversight, and it is my strong belief that the Dobbs decision(the case that overturned the aforementioned precedent) changed everything in the political landscape, and that the numbers reported in the PVI, and composed before the Dobbs ruling(and before the leaked draft a month earlier) should be about 7-8 points bluer, on average. For the sake of conservatism though, I am holding to an aggregate move of 5 points.
TLDR, for the skippers: The first sheet shows every district's PVI rating, with D being positive and R being negative(for purposes of being able to give the cells a color scale from red to blue), 0 indicating it doesn't lean D or R. Each column from there shows the districts' standing in the event of a swing by the number at the column header, from 1 to 5, then 10 and 15, for some crazy wishcasting. In the frozen rows at the top, totals are computed for how the House outlook is affected by these swings. In addition, there is a small ranking near the top-right that shows the number of districts each party has by margin.
Senate and Governor: These are fairly straightforward, with related links for applicable campaign and analysis information. There are also some artifacts in them from when I built the next sheet.
Polls Close: This sheet shows, in chronological order, the individual elections by the time the polls close and when results can start to be counted and reported. This, along with the House Swing View, is where the lion's share of the effort of building this spreadsheet went. I've labeled, where possible, which candidates are incumbents and party-incumbents. Pursuant to my aforementioned prediction, I have highlighted the candidates in all the races I think the Democrats will win-- which is to say I have highlighted all the races with a PVI of "-5"(R+5) or better. I put totals of all districts with PVI ratings of greater than 15 points for either party in the frozen rows, to give some deference to the more-or-less established districts. The remaining 249 districts are less predictable. As of the moment of writing this blog post, no major news websites have links accessible yet to state-based results, except NBC News, so that link is in there now. More will be added on the day of the election, as link structures become available.
I will be tracking results from ballotpedia, the Washington Post, the New York Times, the AP, CNN, NBC, CBS, ABC, and any other sources that might have something distinctive and reliable, and I am open to reasonable suggestions. Just to be clear, and upfront, I will not be checking any far-right sources, except for novelty purposes, as they have worked very hard to give the false impression of what Americans really care about this cycle by spamming the information space with sham polls.
I will add a column to the Polls Close sheet to show actual results in the same format as the PVI listing, and I'll find some way to indicate whether a particular race has been called and confirmed. I will be making these updates, god-willing,m in realtime on election night, and probably for the duration of the counting, which is expected to last for several days.
For the last few months, I've been working on a spreadsheet-- first in LibreOffice, then on google docs --that shows what I consider to be the top elections happening next week in the US. It includes(or at least intends to include) details about all the elections for the US House of Representatives, the US Senate, and all state Governors.
The sheet currently has four sub-sheets: House Swing View, Senate, Governor, and Polls Close. I'll explain these below.
House Swing View: First some background. Skip to the TLDR paragraph if you don't feel like reading all this. The Partisan Voting Index, compiled in July by the Cook Political Report, is widely regarded as a reliable source for the political leaning of any given Congressional district. My district for example, Colorado's 1st, has a rating of D+29, meaning that it is believed that an election in my district would most likely end with the Democratic candidate winning by 29 percentage points over the Republican candidate. Most districts are not nearly that one-sided though. As this sheet details, over 100 are +/- 10 points. After every national census, every Congressional district is evaluated, and often redrawn, based(ideally) on population change since the previous redrawing. This process was completed in 2022, following the 2020 census. When it was completed, or at least when it looked much like it does today, the Cook Political Report predicted the Republican Party would retake the House of Representatives simply by virtue of the redrawing that had taken place. The PVI was released in July, substantiating the claim that 220 districts lean toward the GOP, while only 208 lean toward the Democratic party, with the remaining 7 districts in a statistical tie.
However, something happened between redistricting and the release of the 2022 PVI. Something big. Something seismic. The Supreme Court overturned the Roe v. Wade, eliminating the federal protection of abortion rights in the US. The PVI did not reflect this at all in the July report-- or possibly just assumed it would have no effect on the landscape. I think that was a critical oversight, and it is my strong belief that the Dobbs decision(the case that overturned the aforementioned precedent) changed everything in the political landscape, and that the numbers reported in the PVI, and composed before the Dobbs ruling(and before the leaked draft a month earlier) should be about 7-8 points bluer, on average. For the sake of conservatism though, I am holding to an aggregate move of 5 points.
TLDR, for the skippers: The first sheet shows every district's PVI rating, with D being positive and R being negative(for purposes of being able to give the cells a color scale from red to blue), 0 indicating it doesn't lean D or R. Each column from there shows the districts' standing in the event of a swing by the number at the column header, from 1 to 5, then 10 and 15, for some crazy wishcasting. In the frozen rows at the top, totals are computed for how the House outlook is affected by these swings. In addition, there is a small ranking near the top-right that shows the number of districts each party has by margin.
Senate and Governor: These are fairly straightforward, with related links for applicable campaign and analysis information. There are also some artifacts in them from when I built the next sheet.
Polls Close: This sheet shows, in chronological order, the individual elections by the time the polls close and when results can start to be counted and reported. This, along with the House Swing View, is where the lion's share of the effort of building this spreadsheet went. I've labeled, where possible, which candidates are incumbents and party-incumbents. Pursuant to my aforementioned prediction, I have highlighted the candidates in all the races I think the Democrats will win-- which is to say I have highlighted all the races with a PVI of "-5"(R+5) or better. I put totals of all districts with PVI ratings of greater than 15 points for either party in the frozen rows, to give some deference to the more-or-less established districts. The remaining 249 districts are less predictable. As of the moment of writing this blog post, no major news websites have links accessible yet to state-based results, except NBC News, so that link is in there now. More will be added on the day of the election, as link structures become available.
I will be tracking results from ballotpedia, the Washington Post, the New York Times, the AP, CNN, NBC, CBS, ABC, and any other sources that might have something distinctive and reliable, and I am open to reasonable suggestions. Just to be clear, and upfront, I will not be checking any far-right sources, except for novelty purposes, as they have worked very hard to give the false impression of what Americans really care about this cycle by spamming the information space with sham polls.
I will add a column to the Polls Close sheet to show actual results in the same format as the PVI listing, and I'll find some way to indicate whether a particular race has been called and confirmed. I will be making these updates, god-willing,m in realtime on election night, and probably for the duration of the counting, which is expected to last for several days.
Only logged in users may enter comments.
| subscribe: | posts | comments |
| validate: | html | css |
| login: | not logged in |