Just about every election night includes reporting errors. They can be difficult to identify, but can often be corrected during a recount or a postelection canvass. This year’s Iowa caucuses are the reverse: Errors are now easy to identify, and hard to correct.
The errors are detectable because of changes to the way the Iowa Democratic Party reports its results, put in place after the Sanders campaign criticized the caucus results in 2016. This cycle, and for the first time, the party released three sets of results corresponding to different steps in the caucus process. The rules are complex and thorough, and they create conditions in which the results can be obviously inaccurate or inconsistent within a precinct.
First, caucusgoers express their preference for a candidate upon arrival, and these votes are recorded in a “first alignment.” Then, candidates with limited support at a precinct, usually less than 15 percent, are deemed not viable; their supporters get a chance to realign to support a viable candidate. The preference at this point is recorded as well, and it’s called the final alignment.
Viable candidates can’t lose support on realignment, but there were more than 10 cases where a viable candidate lost vote share in the final alignment, even though that is precluded by the caucus rules.
No new voters are permitted to join the caucus after the first alignment. But in at least 70 precincts, more than 4 percent of the total, there are more tabulated total votes on final alignment than on first alignment.
Many of these cases could be simple tabulation mistakes in a precinct caucus that otherwise went smoothly, like a West Des Moines precinct that reported the first alignment results only for the “nonviable” candidates (those who didn’t meet the 15 percent threshold) but still reported final alignment results for the viable ones. Others appear to be more serious.
At the next step in the process, each precinct allots county delegates based on final preference, and these county delegates are reported to the news media as “state delegate equivalents,” which approximate the number of delegates won at the state convention. Each precinct caucus gets a set number, but a handful of precincts allotted more state delegate equivalents than they had available.