Now that we finally have some clarity on Iowa’s results — with 86 percent of precincts reporting — we’ve turned our primary model back on, including its estimates of the potential fallout from Iowa.
The model shows former Vice President Joe Biden’s chances of winning a majority of pledged delegates being halved — from 43 percent before Iowa to 21 percent now.
Who gains from Biden’s decline? Well, a little bit of everyone. The model thinks Iowa was more good news than bad news for Sen. Bernie Sanders, although it was a somewhat close call. His chances have advanced to 37 percent, from 31 percent before Iowa, making him the most likely person to achieve that majority.
FiveThirtyEight Politics Podcast: The results in Iowa.
Pete Buttigieg’s chances are also up, to 6 percent from 4 percent before, but even after getting most of the credit for winning Iowa in the model (more about that in a moment), they haven’t improved by as much as you think. That’s because, as I explained in Wednesday night’s post, Buttigieg still has his work cut out for him in building a broader coalition; it’s going to require a big bounce in states and among demographic groups where the former mayor is not currently strong. With that said, Buttigieg is potentially quite competitive in New Hampshire, where our model gives him a 20 percent chance of winning, and that could give him a further boost.
Buttigieg has a 9 percent chance of winning the plurality of pledged delegates. The gap between his plurality and the majority odds reflects how he might be poised to benefit from the field remaining divided between several candidates.
Sen. Elizabeth Warren’s majority chances are also slightly up, according to our model, having improved to 10 percent from 5 percent before. That’s because in a more chaotic field, with at least one of the front-runners (Biden) potentially falling back into the pack, her roughly 15 percent of the vote in national polls could eventually give her some opportunities, even though none of the next three states look especially promising for her.
But the big winner is … nobody. The chance of there being no delegate majority has increased substantially, to 27 percent from 17 percent before Iowa. For reasons we’ve explained previously, the no-majority scenario isn’t quite the same thing as a contested convention, but the two concepts are closely related.
In running the model, we’re relying on Iowa results as currently reported (as of 5:15 p.m. Eastern on Wednesday afternoon) plus estimates from our friends at The Upshot on the probability that each candidate eventually wins the various vote measures that Iowa tracks. The Upshot now gives Buttigieg about a 98 percent chance of winning the most state delegate equivalents (SDEs) — the measure that the media has traditionally used to declare Iowa winners (and the only measure that Iowa has reported before this year), although it creates a bias toward candidates who perform well in rural areas.
Meanwhile, The Upshot gives Sanders a greater than 99 percent chance to win the first alignment vote — the candidate that voters initially lined up with when they entered their caucus sites. They regard the final alignment popular vote — the candidate that voters wound up with after supporters of nonviable candidates were allowed to realign — as a tossup between Buttigieg and Sanders.
As I explained on Wednesday, we hadn’t totally thought through how to handle the case where different candidates won by different metrics in Iowa. We were slightly surprised, however, at how much emphasis the media put on SDEs as opposed to the popular vote metrics. Since post-primary bounces are largely the result of media coverage, we decided that our formula for predicting bounces should reflect that.
So we looked at how media outlets that are designated as Democratic National Committee debate poll sponsors — a list that includes most of the biggest newspapers, wire services and TV networks — were covering the Iowa results and whether they were emphasizing SDEs, the popular vote measures, or some mix of both. More specifically, we looked at the first few paragraphs1 of the most prominently featured article about the Iowa results on each outlet’s website as of 1 p.m. on Wednesday afternoon.
- We found that seven news organizations (The Washington Post, Associated Press, CNN, ABC News, the Des Moines Register, Reuters and USA Today) gave very high or almost exclusive billing to state delegate equivalents.
- Four news outlets (The New York Times, CBS News, NBC News and The Wall Street Journal) gave roughly coequal billing to SDEs and the popular vote measures.
- Two outlets (Fox News and NPR) were roughly halfway in between, treating SDEs as the main measure but prominently mentioning the popular vote as an alternative.2
- No outlets gave more billing to the popular vote
- And we ignored one outlet, the Las Vegas Review-Journal, because its most prominent story about Iowa results was an AP story, and the AP was already covered elsewhere.
If you do the math, this works out to roughly 80 percent of the media emphasis being on SDEs and 20 percent on the popular vote. Thus, in calculating the bonus for who won Iowa, we’ll give 80 percent of the bonus to the SDE winner and 10 percent to the winner of each of the popular vote metrics. Overall, given the Upshot’s current probabilities in Iowa, this works out to Buttigieg getting about 85 percent of the credit for winning Iowa, and Sanders getting about 15 percent.
Keep in mind, though, that all of the wackiness around Iowa makes forecasting post-Iowa bounces much harder than usual. So our model might be wrong — and as new polling comes in, it well replace those assumptions, potentially creating big swings in our numbers. We’ll also update our model as results from the state are finalized.