Former Vice President Joe Biden didn’t just have a good night in South Carolina on Saturday; he had a great night, winning by a larger margin than anticipated. With 100 percent of precincts reporting, Biden took 48 percent of the vote in the Palmetto State, while Sanders took 20 percent. Billionaire Tom Steyer received only 11 percent; having staked his campaign on a strong showing in South Carolina, he dropped out of the race late last night. Former South Bend, Indiana, Mayor Pete Buttigieg at 8 percent, Sen. Elizabeth Warren at 7 percent, Sen. Amy Klobuchar at 3 percent and Rep. Tulsi Gabbard at 1 percent brought up the rear.
The FiveThirtyEight forecast in South Carolina was expecting Biden to win big, forecasting on average that Biden would win 40 percent of the vote. Biden outperformed that estimate by 8 points, but because our model had already anticipated he would do well, our forecast didn’t drastically change (and some of the changes were actually due to some new polls we got over the weekend, not the South Carolina results). In fact, the biggest takeaway from our national forecast is just how chaotic the race for the nomination remains — the most likely outcome at this point is that no single candidate will win a majority of pledged delegates.
As of Sunday at 11:10 a.m. Eastern, there is a 3 in 5 (59 percent) chance that no one will win a majority — up from 1 in 2 (52 percent) before South Carolina. Sanders has a 1 in 4 (27 percent) chance of getting a majority, slightly worse than his 1 in 3 (32 percent) chance on Saturday morning. And Biden has a 1 in 7 (14 percent) chance of getting a majority, little changed from where he stood before South Carolina (1 in 6, or a 16 percent chance). All other candidates, frankly, have few paths to the nomination, and South Carolina didn’t change that.
|Candidate||Avg. projected delegates||Change||Chance of majority||Change|
Some of those changes are attributable to a number of good polls Sanders picked up over the weekend. According to Marist College, Sanders is beating Biden 34 percent to 19 percent among likely voters in Texas, with former New York City Mayor Michael Bloomberg (15 percent), Warren (10 percent) and Buttigieg (8 percent) within shouting distance. However, after adjusting for Marist’s house effects, our model interprets this poll as more like Sanders 32 percent, Biden 20 percent. A Dallas Morning News/University of Texas at Tyler poll agrees that Sanders (at 29 percent) has a lead on Biden (at 19 percent), but it also shows Bloomberg doing well, at 21 percent. However, a YouGov/CBS News survey shows a closer race — Sanders 30 percent, Biden 26 percent. Warren (17 percent) and Bloomberg (13 percent) are further behind, although once you account for house effects, they swap positions: Bloomberg is closer to 15 percent and Warren is closer to 14 percent.
And in North Carolina, both Marist and East Carolina University find a close race between Sanders and Biden, with Bloomberg, Warren and especially the other candidates further back. Marist puts Sanders at 26 percent among likely voters, Biden at 24 percent, Bloomberg at 15 percent, Warren at 11 percent, Buttigieg at 7 percent and Klobuchar at 5 percent; with house effects stripped away, though, it’s Biden 25 percent, Sanders 24 percent, Bloomberg 15 percent, Warren 11 percent, Buttigieg 6 percent and Klobuchar 4 percent. ECU puts Biden at 29 percent, Sanders at 25 percent, Bloomberg at 14 percent, Warren at 11 percent, Klobuchar at 5 percent and Buttigieg at 4 percent; house effects don’t meaningfully change the picture.
Finally, YouGov/CBS News also released a poll today confirming that Sanders has a solid lead in California: He gets 31 percent support, while Biden gets 19 percent, Warren gets 18 percent, Bloomberg gets 12 percent and Buttigieg gets 9 percent. Considering house effects, though, Warren is only barely ahead of Bloomberg, 15 percent to 14 percent. However, all these polls were obviously taken before the South Carolina results are known, and they do not take into effect any bounce Biden might (or might not) get.
All this new data means it’s high time to check in again on the outlook in the 15 states and territories1 that vote just two days from now, on Super Tuesday. While Sanders has maintained his grip on the Northeastern and Western Super Tuesday states — most importantly, delegate-rich California — Biden is now favored to win all seven Southern Super Tuesday states except Texas, which is a virtual toss-up.
And the average number of pledged delegates that we are forecasting each candidate to receive on Super Tuesday goes to show why we think it’s so likely that no candidate will win a majority of pledged delegates. Sanders should still win the most delegates (540 on average), but Biden shouldn’t be too far behind (395 on average). And even Bloomberg and Warren could pocket a significant number. That’s important because, by Wednesday, 38 percent of all pledged delegates will be committed to a candidate (unless those candidates drop out), making it increasingly difficult to avoid a contested convention.
We’ll close with a word of warning. The FiveThirtyEight model is probably its most uncertain immediately following a big primary day; it’s really just making an educated guess about what post-South Carolina polls will say, so we’d urge you to wait for some new polls before taking the model’s predictions too literally, as we did after Iowa, New Hampshire and Nevada. Of course, the complicating factor here is that the next election is in just two days, so there probably won’t be time for many new polls to be conducted between South Carolina and Super Tuesday. That may mean we are in for some surprises.