Those turnout numbers might also throw some water on concerns that coronavirus would keep people away from the polls, Nate.
With the state almost fully reported, turnout in Virginia is around 1.3 million voters, much bigger than 2016, when it was 785,000. It seems like a problem for Sanders’s narrative about turnout that the higher the turnout in a state, the worse he seems to be doing.
Over at NBC News, they’ve called Tennessee for Biden. His lead there continues to grow has more of the Election Day vote has been counted; he’s now at 31 percent there, Sanders is at 24 percent, Bloomberg is at 19 percent and Warren is at 9 percent.
I don’t think it’s ever that simple. Not all Bloomberg supporters would automatically go to Biden. And, as has been mentioned, Bloomberg holding above 15 percent means fewer delegates for Sanders when it’s all said and done.
With 20 percent of precincts reporting in Texas, Sanders is at 29 percent, Biden at 22 percent and Bloomberg at 18 percent. What do you make of the argument that if Biden loses Texas, Bloomberg is to blame?
Going back to the chart Laura just posted (I put it here too) and our conversation about how much Texas might matter, Biden is really hurt by so many Texans having decided and voted well before the last few days. The race is shaping up to be so close — it’s pretty safe to say that had everyone in Texas voted today, Biden would have won.
Maine has just 5 percent reporting, and it’s a smattering of smaller towns, so I wouldn’t take Biden’s 8 percentage point lead to the bank just yet. Great Pond’s one precinct has reported, and four of its five total votes went to Biden. Since Maine used a caucus in 2016, it’s hard to benchmark how Sanders is doing, but it was certainly a state he’s expected to do well in.
Since exit polls have updated, it’s worth checking back in with when voters have made up their minds, according to preliminary exit polls. That ranges from 53 percent in the last few days in Minnesota — which makes sense in a state in which the home candidate dropped out just yesterday — to just 18 percent in California, where a large chunk of voters vote early by mail. (The exit polls do also poll absentee voters in California in addition to physical election-day voters.)
With talk tonight of the “party deciding” on Biden ahead of Super Tuesday, it’s worth looking back at the long history of the parties using money and influence to decide which candidate wins the nomination.
With 99 percent of precincts reporting in Virginia, turnout so far is 67 percent greater than it was in the 2016 primary (or about 525,000 more votes). We know that there has been only a 7.4 percent increase in voter registrations (or 392,000 registrations) since 2016, so turnout as percent of registered voters does seem to be higher.
Is it too early in the evening to critique the primary process? I feel like there’s gotta be something to say about a process that goes on for a year, yet seems to be advantageous to the best-known candidates.
The early returns show Biden with a slight lead in Tennessee, but don’t bet too much on those early reports, as we may not see returns from Davidson County (Nashville) for a bit because of the tornado strikes this morning. But maybe a potentially good sign for Biden is that, according to preliminary exit polls, 48 percent of Tennessee voters said they wanted to return to Obama’s policies, a group that Biden has done well with in other states.
Meanwhile, there are still lots of reports of huge lines to vote in Los Angeles County. At UCLA, Sanders supporters are handing out water and pizza to folks in line, which makes sense, given that when I stopped by the campus last week, it was hard to find a student who wasn’t supporting Sanders.
But Nathaniel, I think you’re thinking of things in too much of a binary way. If Biden was favored by 1 point in Texas and Sanders wins by 1 there … that’s probably not going to matter as much as Biden winning a couple of huge upsets or winning states like Virginia by 30 points. Especially because Texas might not get called until late in the night — or at some point late this week — in that eventuality.
But Nate, our model thought Biden was favored in Texas (albeit narrowly). So if Sanders wins that, that will be an overperformance for him there, which should probably offset some random Biden overperformance in Massachusetts? (Now, if Biden overperforms in Massachusetts and Minnesota and Maine, that’s different.)
For what it’s worth, Laura, Latinos in Texas are typically more moderate than Latinos in Nevada or California — and they also tend to be a little older. So it would make sense if Sanders’s win among Latinos in Texas isn’t as decisive.
I’m in the ABC News studio right now, Micah, and I think you might be wrong about that. The overall upward trajectory for Biden, plus winning by huge margins in the South, plus an upset win or two would lead to a pretty good narrative for him, I think. And it would also probably put him in a good position in our model, if Sanders only narrowly won Texas.
According to preliminary exit polls, Sanders is doing quite well among Latinos in Texas — though not quite as well as he did among Latinos in Nevada, where he garnered 50 percent in a more crowded field. But the share of the Democratic electorate that is Latinx didn’t increase since 2016 — maybe because Texas has open primaries and the Republican primary isn’t particularly interesting this year, so Republicans may be casting ballots in the Democratic race — so we don’t yet know whether Latino turnout will be high enough to help Sanders win the state. And like in other Southern states, Biden is doing very well among black voters, who make up 20 percent of the Texas Democratic electorate. (In fact, Biden’s lead among black Texan voters is larger than Sanders’s lead among Latinos.)
I guess, Nate and Julia, there’s a difference here between what narrative we think is right and what narrative the media at large settles on. It’s the latter that will influence the race, and I have a hard time seeing most outlets play Super Tuesday as a Biden win overall if Sanders wins California and Texas — the two biggest states.
Butting into Nate and Micah’s conversation here, just a reminder that there are always lots of available narratives and the actual results tend to be only one factor in determining what narrative becomes accepted.
Election nights are long affairs, Micah, and resolutions are made to be broken. The thing for is, if Biden wins places like Massachusetts or Maine or Minnesota, plus all the states he’s supposed to win, that starts to look like a pretty bad night for Sanders, even if Sanders holds on by 2 points in Texas or something because early voting saved him.
Nate, I’m not settling on any narrative until we know the winner in Texas — I made that resolution to myself before the night started.
I think the Biden momentum narrative is overrated until we get more results from Texas and California. At the same time … there are a lot of theoretically Sanders-friendly states that have looked closer than you’d think based on early results, like Massachusetts and Maine. A Texas win coupled with his big margins in the rest of the South could also be a good scenario for Biden. So there’s just a liiiiiiitttle bit of a feeling of Election Night 2016, where we’re sort of waiting for the other shoe to drop. Maybe, anyway.
It’s still early in my home state of Massachusetts, but there’s already some pro-Biden evidence around Boston. The suburbs in the inner ring around Route 128 are mostly wealthy college-educated voters, and Biden is leading early in the count in Needham, Weston and Lexington, which should probably be Warren-friendly territory. The outer ring, Route 495, is more conservative and rural: Scott Brown ran up large margins here in his Senate victory. Biden is winning there too, in towns like Chelmsford, Littleton and Bolton. Sanders is doing well in early returns in Western Massachusetts, but if Biden can keep up his leads in both suburban rings around Boston, that will bode well for his chances.
Sure enough, we’re up to 25 percent of Election Day precincts reporting in Tennessee, and Biden has taken a narrow lead. He’s at 27 percent, Sanders has 24 percent, Bloomberg has 22 percent and Warren has 10 percent.
What The Delegate Race Looks Like At The District Level In Colorado
Voting centers in Colorado close at 9 p.m. Eastern, but most of the state votes by mail, so it could be a while before we have full results. As of this morning when we froze our forecast, Sanders had a 7 in 10 (69 percent) chance of carrying Colorado, while Biden is a dark horse at 1 in 5 (20 percent). On average, our forecast anticipates the final results to be Sanders 29 percent, Biden 23 percent, Bloomberg 19 percent and Warren 18 percent. But because Colorado allocates delegates on both the state and congressional-district level, you’ll want to pay attention to the results by district as well. Here are our forecasts for that:
Sanders is favored across Colorado
Average forecasted vote share for the top four Democratic presidential candidates in Colorado congressional districts, according to the FiveThirtyEight model as of 9:30 a.m. on March 3
| District | Sanders | Biden | Bloomberg | Warren |
|---|---|---|---|---|
| CO-01 | 28% | 23% | 20% | 17% |
| CO-02 | 29 | 21 | 18 | 20 |
| CO-03 | 29 | 24 | 18 | 17 |
| CO-04 | 29 | 23 | 17 | 18 |
| CO-05 | 28 | 24 | 19 | 17 |
| CO-06 | 28 | 24 | 18 | 18 |
| CO-07 | 29 | 21 | 20 | 18 |
| State | 29 | 23 | 19 | 18 |
There’s a big difference in the share of the electorate that’s college-educated across the states that are voting tonight; preliminary exit polls show that it ranges from 60 percent in Massachusetts to 35 percent in Oklahoma. But so far, in the states that have been called, Biden and Sanders have a small advantage among non-college-educated voters in most states, while Warren is doing better among college-educated voters than those who don’t have a college degree.
What The Delegate Race Looks Like At The District Level In Minnesota
Sanders leads in every Minnesota district
Average forecasted vote share for the top four Democratic presidential candidates in Minnesota congressional districts, according to the FiveThirtyEight model as of 9:30 a.m. on March 3
| District | Sanders | Biden | Warren | Bloomberg |
|---|---|---|---|---|
| MN-01 | 28% | 24% | 16% | 15% |
| MN-02 | 29 | 22 | 17 | 15 |
| MN-03 | 28 | 22 | 17 | 15 |
| MN-04 | 27 | 23 | 16 | 16 |
| MN-05 | 27 | 24 | 15 | 17 |
| MN-06 | 29 | 23 | 16 | 15 |
| MN-07 | 29 | 24 | 15 | 15 |
| MN-08 | 28 | 23 | 15 | 15 |
| State | 28 | 23 | 16 | 16 |
Yeah, Sarah, we have so few results and polls aren’t even completely closed in the two biggest states yet! It’s still too early to say that any narrative is set.
Sarah, I agree with Julia that narratives fluctuate, and there at least a couple story lines competing already tonight. But I have to wonder how social media contributes to narratives solidifying more quickly. Or at least how quickly they consolidate within echo chambers (I know, I know, plenty of social science says echo chambers are a myth).
In Tennessee, with 14 percent of Election Day precincts reporting plus a good chunk of the early vote, it’s a notably close race. Sanders has 25 percent, Biden has 24 percent, Bloomberg has 23 percent and Warren has 10 percent. This was a state we expected Biden to win, so I suspect this is another case of the early vote being disproportionately good for Sanders, but this is worth watching.
As a person who has spent a lot of time studying election narratives, I’d say that they often remain in flux for a bit. Not to overly conflate general elections and primaries, but the 2018 election narrative went from “no narrative” to “Democratic wave” for example. Across the contests of this nomination season so far, there’s been a lot of change from “Biden is tanking” to “Biden is the anointed party favorite” based on the South Carolina result. Narratives seem very flexible to me, and I wouldn’t be surprised if things change as the night goes on.
Sarah, I think there’s still time for that narrative to change if Sanders starts to pull out some wins in states like Texas. The early returns in California are also probably not going to look good for Biden, which could be helpful to Sanders from a narrative-setting perspective (even though the results are likely change a lot as more votes are counted). But the pressure’s really on Sanders now as we move further west.
So far only four states have been called — three for Biden (Alabama, North Carolina and Virginia) and one for Sanders (Vermont). None of these have been that surprising, but it does feel as if a “Biden is surging” narrative has already been set. Is that premature? Or is there time for that to still change tonight?
One place I’ve got my eye on is southwest Virginia. Clinton narrowly won the rural, heavily white counties in the southwestern corner of the state in 2016, and Sanders is now winning about 31.4 percent of the vote there, which is better than his statewide performance of 23 percent at the moment. Those counties should give a hint of other Appalachian counties, especially in eastern Tennessee and West Virginia.
One thing to watch out for on election nights is the bias of early results, which can set the tone and the narrative even if later results run in different directions. I’m a big fan of Virginia as a former resident of that Commonwealth, but it’s worth keeping in mind that given the heavy presence of government workers in the northern part of the state, it might not be a bellwether in primaries.
The Upshot Needle projects that Warren will win only 1 delegate in Virginia, and that Bloomberg will win zero.
Today is also the first primary election for downballot races like those for the Senate and House, and the Associated Press is officially projecting that state Sen. Cal Cunningham will win the Democratic primary for U.S. Senate in North Carolina. He’ll face Sen. Thom Tillis in one of the most closely watched Senate races of the fall.
In California, preliminary exit polls seem to suggest that the electorate is quite a lot more liberal than it was in 2008 (there was no exit poll there in 2016). That might be an early good sign for Sanders, though of course we won’t know the results for a while.
To build on Nate’s point about early voting, at first glance, it looks like Biden is doing a lot worse in North Carolina than he was expected to be; he has 31 percent, while Sanders has 24 percent, Bloomberg has 18 percent and Warren has 12 percent. But most of what’s reporting so far is the early vote; I’d be willing to bet the Election Day vote will have much more Virginia-esque margins for Biden.
The one thing that will be interesting in Virginia will be the remaining precincts in the northern part of the state and the absentee votes from Fairfax County. I’m curious to see if Sanders or Warren do better among those voters than Biden, which could help push Biden’s margin below 30 percent in Virginia. It wouldn’t shock me if that happened. But as Nate pointed out, Virginia doesn’t have a lot of early or absentee voting compared to a state like Texas, for instance.
I keep on thinking back to my chats with Biden campaign staffers and surrogates this summer when they were in the midst of a whole lot of bad news cycles about Biden and black voters, and they just kept saying that his strengths would lie in the South on Super Tuesday. Man, say what you will about Biden as a candidate, but that campaign had a theory of the case from the very beginning and they stuck with it: Be the moderate and meet the majority (or plurality) of the party where they are. Other candidates probably misjudged slightly how liberal the party base was willing to be. But that was always the Biden strategy.
Meanwhile, President Trump is winning all of the Republican primaries and tweeting his appreciation for each individual win.
If the North Carolina exit polls are any indication, Biden still has a problem with young voters. According to preliminary data, Sanders crushed him 54 percent to 21 percent among 17- to 29-year-olds, and they virtually tied among voters aged 30-44. However, voters between 45 and 64 went for Biden 47 percent to 18 percent, while Biden routed Sanders 55 percent to 8 percent among those 65 or over.
What The Delegate Race Looks Like At The District Level In Arkansas
In Arkansas, polls have closed and our forecast suggests Biden has about a 7 in 10 (69 percent) chance of winning statewide, though Bloomberg has a 1 in 6 (17 percent) shot and Sanders about a 1 in 8 (12 percent) chance. Biden has a forecasted statewide vote share of 35 percent, on average, ahead of Bloomberg’s 23 percent and Sanders’s 25 percent. But the results by congressional district will also matter for delegate allocation purposes. And in Arkansas, it’s possible that Bloomberg might outpace Sanders in some of these districts. Here is our forecasted vote share for each candidate in each district:
Biden looks fairly strong in Arkansas
Average forecasted vote share for the top four Democratic presidential candidates in Arkansas congressional districts, according to the FiveThirtyEight model as of 9:30 a.m. on March 3
| District | Biden | Bloomberg | Sanders | Warren |
|---|---|---|---|---|
| AR-01 | 37% | 24% | 22% | 14% |
| AR-02 | 36 | 25 | 22 | 15 |
| AR-03 | 29 | 25 | 25 | 17 |
| AR-04 | 37 | 24 | 22 | 14 |
| State | 35 | 25 | 23 | 15 |
Nate pointed out that Warren is winning women in Massachusetts, but she’s not leading with men. According to exit polls, in each of the first four states to vote, Warren did better with women than she did with men — 4 points better in Iowa, 3 points in New Hampshire, 5 points in Nevada and 3 points in South Carolina. This notion of gender affinity (that women support female candidates) seemed weak early in the primary, as Amelia reported, but I think it’s fair to suggest that more women than men have stuck with Warren’s campaign as she’s lost ground, overall, over the course of the primary season.
I can’t help wondering if touchscreen voting machines are germ vectors. Yet another argument for hand-marked paper ballots!
My own coronavirus anecdote: While exit polling, I had someone refuse the survey because of concern about germs, which is a first.
They’re talking about the coronavirus on ABC right now and we’ve gotten a few reader questions about its impact on this election as well. Here are Micah and Galen with their perspective on how the virus could affect the primaries:
