The Democratic caucus in Iowa is too close to call and will come down to turnout, along with other factors seen in a caucus as opposed to a primary. The final Des Moines Register/Bloomberg Poll before the Iowa Caucus showed Hillary Clinton leading Bernie Sanders with 45 percent compared to 42 percent. The three point lead is within the poll’s margin of error at four percent. Martin O’Malley trails at three percent.
The final Quinnipiac University Poll, coming out this morning, has Sanders leading Clinton by three points, also within the margin of error.
While I have often pointed out the limitations of polls before primaries, the final Des Moines Register poll is probably the most likely to be predictive. Among its virtues, it does not exclude voters based upon past lack of participation in the caucuses as many other polls do. While it has a better track record than other polls, it still suffers from the same problem of all pollsters in not knowing who will actually turn out. Traditional Democratic voters favor Clinton while more independent voters strongly favor Sanders, but we don’t know how many of them will participate in the caucuses. Higher turn out than usual would increase the chances of a victory for Sanders.
Being a caucus rather than a pure primary vote creates additional questions. A candidate has to meet a fifteen percent threshold for their vote to count towards selecting delegates in the Democratic caucus. If they do not meet this threshold, then the second choice becomes crucial. Greater support for Sanders than Clinton among O’Malley supporters nearly erases Clinton’s lead if the votes go to Sanders. However, Buzzfeed reports on how the Clinton campaign is trying to game the system by having some of their supporters back O’Malley so that he will meet the fifteen percent viability requirement to keep his supporters from going to Sanders. Of course plotting such a strategy and getting Iowa voters to go along are two different things. I recall how Clinton protested over similar actions by the Obama campaign eight yeas ago.
Bloomberg has more background on Clinton’s strategy in Iowa–basically doing the opposite of what she did in 2008.
Another question is the consequence of the difference in date for the caucus this year compared to 2008, when Obama came in first and Clinton came in third. The 2008 caucus occurred on January 3, when many college students were still on vacation, and possibly out of the state. Will having the caucus occur after students have returned to school provide an additional benefit to Sanders? On the other hand, will college students be more likely to caucus near their campus as opposed to at homes throughout Iowa. There is the danger that this will lead to Sanders having huge leads in some areas, such as Iowa City and Ames, while not doing as well as Obama did in other parts of the state. This could result in Clinton picking up more delegates statewide even if Sanders narrowly wins the popular vote. The Sanders campaign is urging students to “Go Home for Bernie” to vote throughout the state but such travel plans might be complicated by a major storm expected to hit Iowa later tonight.
With turnout so important, Sanders supporters hope that the large turnout at his events indicate a greater likelihood of turning out to caucus.
Donald Trump leads among the Republicans at 28 percent with Ted Cruz in second place at 23 percent in the Des Moines Register/Bloomberg Poll. However, Trump has a weak ground game making it possible for an upset. I also wouldn’t be surprised if candidates such as Rubio, or even Ben Carson, do better than expected. The usually confident Trump is admitting to being a “little bit nervous” going into tonight’s vote.