Skip to content

Misleading vote reporting

Someone pointed me to this:

The top few candidates looked like what I’d remembered hearing, but I was surprised that Yang did so poorly. Just 1% of the vote.

But then I was told to carefully read the fine print at the bottom of the table:

State delegate equivalents, multiplied by 100

That ain’t cool, to label that as “Votes.”

Here’s what really happened:

Then comes Uncommitted with 0.5% of the vote and a few more candidates whose votes are in the hundreds, less than 0.2% of the vote each.

Then this:


  1. Jonathan (another one) says:

    I’m not an expert on this by any means, but I think you’re confusing the first round with the second round. In the first round, Yang got 8,660 votes (as of the count for that table). But then, any candidate who didn’t have a high enough standing in that caucus are forced to choose another candidate (of which uncommitted is one). So his 1 percent (listed in your first chart) is accurate. The last step adjusts for the number of people in the caucus to allocate delegates, and that really didn’t affect Yang much at all, but it helped Buttigieg a lot, because his caucuses get a higher weight because they were undersubscribed relative to caucusgoers.

    • Andrew says:


      It’s confusing! But it can’t be right to take “State delegate equivalents, multiplied by 100” and label this as “Votes.”

      • Jonathan (another one) says:

        The column labeled “votes” isn’t sde’s x 100. Pete ain’t going to get 550 delegates from Iowa! Iowa only sends 41 delegates based on the caucuses (plus another 8.) The sde column is the one that is all 0’s as you’ve printed it… because they don’t know the number yet. (I think!)

        • Ben says:

          SDEs are some sort of state level delegates. They are different from the national delegates. Not totally sure what they’re for, but they’re different.

          • Jonathan (another one) says:

            Actually, the Wikipedia link explains it pretty well and gives the final figures in which, for example, Mayor Pete has 564 sde’s. I think the idea is that each delegate (which will eventually meet to select the actual convention delegates) is supposed to represent somewhere in the vicinity of 100 voters, though for everyone at Steyer and above, 100 x sde > votes, while for Bloomberg and below 100 x sde <= votes. (Bloomberg is at exactly a factor of 100, as is Deval Patrick, but he had 0 votes.)

  2. Terry says:

    Well this is all a ringing endorsement of democracy in America!

    I have heard that the value of the Iowa and other caucuses is that it rewards candidates with the energy and infrastructure to figure all this stuff out.

  3. Anonymous says:

    The best comment on this amazing shambles was just before the caucus:

    Democratic Party @DNC – 22:16 UTC · Feb 3, 2020

    For three years, we’ve been preparing for the process that officially
    kicks off tonight in Iowa: the Democratic presidential primary. Today
    our chair, @TomPerez, reflects on the reforms we’ve made to make this
    the most transparent primary in our history: How We Prepared for 2020

    ‘Nuff said.

  4. joshua pritikin says:

    Based on what I read, this is due to inconsistent application of rank choice voting. It’s good that they used some rank choice voting, but it would be better if they applied it state-wise instead of locality-wise.

Leave a Reply