Eleven Democratic candidates are running in the three most Republican congressional districts in the state — the 5th, the 6th, the 9th. Make that 12 if Del. Sam Rasoul, D-Roanoke, carries through on his exploratory bid. However, his campaign is predicated on new districts being drawn. The other 11 are running in the districts as they are now.
Do those candidates know something the rest of us don’t know? Or is this a classic case of hope winning out over experience?
Let’s see what math tells us — recognizing that math leaves no room for hope whatsoever. Math is grounded in experience: Two plus two is always four. If someday two plus two equals five, everything changes. The point being: Math can’t predict unpredictable realignments, which sometimes happen in politics.
There is already one big question looming over the midterms in Virginia that math can’t help us with: Will our districts even look the same? Virginia Democrats want to pass a constitutional amendment in an unusual spring referendum that would allow the General Assembly (where Democrats hold a majority) to redraw the current maps to make them more favorable to their party — what we typically call gerrymandering. The courts might stop this and voters might stop this. Or maybe not. If they don’t, the analysis that follows is all for naught, but for now we must work with what we have.
All three of those districts — now represented by John McGuire, Ben Cline and Morgan Griffith — voted for Republican Winsome Earle-Sears over Democrat Abigail Spanberger in this year’s gubernatorial election, but let’s dig into each of them to see what it would take for a Democrat to win these districts as they’re currently drawn.
5th District: Now held by John McGuire, R-Goochland County
This district is the least Republican of the three. Here’s how the governor’s race came out, according to calculations by the Virginia Public Access Project:
Earle-Sears: 183,191; 53.7%
Spanberger: 157,129; 46.1%
What we want to pay attention to are the raw numbers, not the percentages. Now let’s look at the most recent congressional results in the 5th District, the 2024 election that coincided with the presidential election.
John McGuire (R): 249,564; 57.3%
Gloria Witt (D): 184,229; 42.3%
What should we notice out of this? The first thing is how many more votes there are in a presidential year, which is why that’s not a good measure to go by. We really need to go back to the previous congressional midterms, in 2022, to get a fair comparison. Here’s where we encounter our first complication: McGuire wasn’t in office in 2022, Republican Bob Good was. Maybe Good attracted votes McGuire won’t get, maybe he lost votes that McGuire would retain. All that’s unknowable, so we’ll just have to assume that Good’s vote in 2022 is close to what McGuire would get in November 2026 (assuming he’s the nominee again and doesn’t get ousted in a primary by Good). Lots of unknowns. In any case, in 2022:
Bob Good (R): 177,191; 57.6%
Joshua Throneburg (D): 129,996; 42.2%
However, 2022 may not be a good comparison, either. Those midterms came when Joe Biden was president, and the party in the White House usually suffers. To get the best comparison, we really ought to go back to 2018 — the previous midterms under Donald Trump — except the districts were different then. So these are the numbers we have to work with, even if they are not exactly the ones we want.
The next question to ask is what turnout will look like next year. This year, 59.4% of registered voters cast ballots. Last year, 70.48% did. In the 2022 midterms, only 49.28% voted, but in those 2018 midterms, 59.5% did — the highest turnout in a congressional midterm in Virginia since the Motor Voter Law expanded registrations in the 1990s.
That’s almost exactly the percentage we saw this year, so I feel confident using the gubernatorial numbers as a guide. This was obviously a Democratic year in Virginia and history would suggest next year will be, as well, but we can adjust for that if we need to.
If we use this year’s governor’s race as our standard, that’s not an encouraging measure for potential Democratic candidates in the 5th. Spanberger was an exceptionally strong candidate and Earle-Sears an unusually weak one, yet that weak Republican still carried the 5th.
Will McGuire run as poorly as Earle-Sears did? No Democrat should count on that. For the sake of argument, let’s assume McGuire gets every vote Earle-Sears did and no more — that’s 183,191 votes. For a Democratic candidate to top that, he or she would have to get the same vote that Witt did in a presidential year — 184,229. Is it reasonable to think the Democratic candidate next year will get presidential-level turnout while the Republican one does not? I will leave that to your imagination but, mathematically, that’s what would be required.
When we look at specific localities, we see an even more daunting challenge for Democrats. The biggest locality in the district is Albemarle County, which is also strongly Democratic so constitutes the biggest source of Democratic votes in the 5th. This year, Spanberger ran almost at presidential-year levels in Albemarle. Witt took 42,689 votes there in a presidential year, Spanberger took 39,322 in a gubernatorial year — and still lost the district overall. She also ran appreciably better than the Democratic candidate for the U.S. House did in Albemarle in the previous Trump midterms of 2018: That year, Leslie Cockburn took 34,409 votes out of Albemarle. We’ve obviously had some population growth there since then but, even when we account for that, Spanberger does seem to have run very well in Albemarle. Cockburn and Witt both took about 64% of the vote in Albemarle; Spanberger cranked that up to 70%.
Maybe a Democratic candidate next year can duplicate that feat, but that candidate will still need to run stronger than Spanberger did in rural areas — and Spanberger already ran better in rural areas than Democrats have in several cycles. This brings us back to the same place I arrived at in a previous analysis earlier this year: For a Democrat to win in the 5th, that candidate will need to win rural votes at a rate Democrats haven’t done in a long time. Can it be done?
What the special elections tell us (if anything)
One big caveat to this and the other districts that follow: We’ve had five special elections for the U.S. House across the country this year. All saw big swings toward the Democrats, although they didn’t necessarily win all of those. Last week, there was a special election in Tennessee. A congressional district around Nashville that had gone to Trump by 22 percentage points was won by the Republican candidate by a margin of 9 percentage points — so that’s a 13-point swing. That was actually the smallest swing of the five special elections, according to data journalist G. Elliott Morris, who writes the incredibly useful “Strength in Numbers” newsletter. In Virginia’s special election in the 11th District, Democrats saw a 16-point swing. In one Florida district, the swing was 23 percentage points. If we leave that one out as an outlier, the average swing has been 15.5 percentage points. If we count that Florida race, the average swing toward Democrats has been 17 percentage points.
I’m quite content to say this indicates a Democratic trend — that’s pretty obvious — but I’m wary of applying these special election numbers to a general election. So is Morris: “Handicappers should be careful not to apply the full special-election swing when forecasting next year’s congressional midterms. Turnout in special elections tends to be much lower than in midterms, with lower-education voters and the party in control of the White House disproportionately staying home. … Swings in specials tend to exaggerate subsequent swings in midterms.”
Still, while we’re doing math, let’s do some more. If we applied the lowest of these swings — the 13 percentage points in Tennessee — to the 5th District next year, that would result in a Democratic victory. Even a swing half that much would lead to a Democratic victory, if we go by the gubernatorial results and not the 2022 midterms. The bottom line: A blue wave could put the 5th in play despite my skepticism about Democrats’ ability to win rural voters. After all, they may not need to win many extra voters; they mostly need dispirited Republicans to stay home.
6th District: Now held by Ben Cline, R-Botetourt County
We don’t need to do as much math for the other districts because we’ve already established that we can use this year’s gubernatorial numbers as a guide. This year in the 6th:
Earle-Sears (R): 178,635; 58.6%
Spanberger (D): 125,795; 41.2%
In the last congressional race, in 2024:
Ben Cline (R): 256,933; 63.3%
Ken Mitchell (D): 141,612; 34.6%
In the 2022 midterms, which favored Republicans:
Ben Cline (R): 173,352; 64.4%
Jennifer Lewis (D): 95,410; 35.4%
These numbers lay out the challenge for any Democratic candidate: Even if that candidate matched presidential year-level turnout, while Cline mustered only the low Republican turnout of this year, or the even lower turnout of the last midterms, Cline would still win.
That would be Cline 173,352 (his vote in 2022) versus 141,612 (the high-water mark that Mitchell hit for Democrats in a presidential year). That’s a margin of 31,740 votes.
For a Democrat to win, there are three scenarios:
- The first is for that Democrat to generate votes that didn’t come out even in a high-turnout environment. Is it realistic that there are 31,740 voters out there who didn’t vote in a presidential year and would come out in a midterm? Even if you assume an anti-Trump sentiment, are there 31,740 people who didn’t bother to vote when they could have voted against Trump himself who would feel motivated to vote against a Trump proxy, which is what Democrats would portray Cline as?
- The second scenario is for at least 31,740 of those voters to switch sides. That would be 18.3% of Cline’s support in 2022. Do we think nearly 1 in 5 Republican voters are prepared to abandon ship and take up with Democrats? Hold that thought. We’ll come back to it.
- The third is if the special election margins we’ve seen above carry over to the general election. As noted above, the odds are the trends do but the margins don’t. Just for the record, though, a Tennessee swing of 13 percentage points would take Cline’s 2022 margin down to 51.4%, his 2024 margin down to 50.3%, or this year’s gubernatorial margin down to … a Democratic upset.
9th District: Now held by Morgan Griffith, R-Salem
The numbers become even more daunting for Democrats here.
Earle-Sears (R): 194,951; 68.6%
Spanberger (D): 88,752; 31.2%
And in the most recent congressional race:
Morgan Griffith (R): 290,645; 72.5%
Karen Baker (D): 109,750; 27.3%
And in the 2022 midterms:
Morgan Griffith (R): 182,207; 73.2%
Taysha Lee Devaughan (D): 66,027; 26.5%
The 9th is the most Republican congressional district in the state and we see the magnitude of the challenge for a Democrat here. Even if Griffith managed only the vote he had in 2022, that’s still almost twice as much as Baker won last year — and she was the high point for a Democratic candidate in that district.
If we start with Spanberger’s 88,752 votes this year, a Democratic candidate would need to find 93,455 more votes just to match Griffith’s figure in the 2022 midterms. Is there a Democratic candidate in the 9th who can more than double Spanberger’s vote? Or, if we start with Baker’s presidential-year totals of 109,750, a Democrat would need to find at least 72,457 additional votes to match Griffith’s 2022 number. We obviously won’t know until one tries, but this is the math. Or, if we look at this another way, Democrats need a presidential-year turnout on their side — and then need 39.7% of Griffith’s 2022 supporters to switch sides.
Even if we saw a Tennessee swing of 13 percentage points in the 9th District, Griffith would still win by landslide margins if we go by the 2024 and 2022 elections. If we use this year’s gubernatorial numbers, a 13-percentage-point swing toward Democrats would still leave him victorious, with a vote share of 55.6%.
There may well be a depressed Republican vote next year if Republicans are unhappy about the economy, and it’s possible that some of those voters will cast Democratic ballots — but the numbers required seem unrealistic to me unless … Remember what I said earlier about how math can’t predict realignments? For Democrats to win in either of these three districts, that’s what they need — a fundamental shift in how people vote. Those kinds of shifts do happen; they just don’t happen very often.
In 1932, the year that saw Franklin Roosevelt elected during the Great Depression, Democrats picked up 90 seats in the House. In 1994, the Contract-for-America year that came during Bill Clinton’s first midterms, Republicans picked up 54 seats in the U.S. House. In the 1974 midterms, which came after the Watergate scandal, Democrats picked up 49 seats. The midterms of Trump’s first term saw a gain of 41 Democrats. Even if we see a convulsion of that scale in 2026, for Democrats in the 5th, 6th or 9th to win, they’d need to see that anti-Trump vote reach into counties that it didn’t in 2018. Trump 2.0 is not the same as Trump 1.0, so maybe that could happen. Maybe the math will change. For now, though, the math says those three Republicans are safe in the districts as currently drawn — unless their Democratic challengers can do something that hasn’t been done in a very long time: win over rural voters.
I write a weekly political newsletter, West of the Capital. In last week’s newsletter, I looked at how the redistricting spree set off by Texas Republicans might wind up being a wash nationally as Democratic-controlled states such as Virginia follow suit with their own new lines. If you missed it, email me and I’ll send you a copy. If you want to get signed up for this week’s edition, here’s how: