Flawed polling in SoS race
by Brian Leubitz
With Sen. Yee dropping out of the Secretary of State’s race, the media and the polling operations have been in something of a frenzy to figure out how that will impact the race. And, so, you would think that a poll that was being conducted during that mess could have some very interesting data.
It could, but the Field Poll that was being conducted while Yee was arrested has a few very serious flaws. First, here are the up-front numbers after the Yee arrest: Peterson-R: 30%, Padilla-D: 17%, Curtis-G: 5%, Schnur-NPP: 4%, Cressman-D: 3%, Other/Undecided: 41%.
That’s all well and good, but let’s look at a few flaws in this poll.
1) The poll didn’t include all the candidates on the ballot. Ordinarily in a competitive race with just a few relatively well-known candidates, you can kind of forgive that. However, this is a different kind of race. There are a slew of unknown candidates. Even Padilla, who is the most known candidate in the race, was basically an unknown to 54% of likely voters. But the poll did not include two candidates who haven’t filed fundraising reports with the state: Jeffrey Drobman and Roy Allmond.
Now, to be clear, neither of these two will be your next secretary of state. And they won’t pick up a ton of votes. But Allmond is running as a Republican, splitting the generic Republican vote. Drobman is running as a Democrat and may cause problems for Democrats as well. However, that split of base Republican vote could be meaningful. Peterson, with his $1800 or so that he has in the bank still seems likely to grab one of the top-two positions, but that is hardly a given.
2) The poll was split between pre and post-Yee. The margin of error is higher than most Field polls, with a 5.5% pre-Yee MoE, and 6.5% post-Yee MoE.
3) Winning the June primary is essentially meaningless. We do not have head-to-head matchups in this poll
Conclusion: I normally love the Field Poll data, and some of the things about the coverage that have been bothering me have nothing to do with Field at all. The media should know that winning the June election doesn’t really make you a frontrunner, but that doesn’t stop Breitbart declaring that Peterson is “favored” to win. Yes, he is favored to win the vote totals in June, but that and a quarter will get you a gumball.
Give me data for a head to head matchup between Peterson and Padilla, and then see what we get before any leads are declared. Note that this is also an issue in the Controller’s race. SacBee declared Mayor Ashley Swearingen the leader in that race, despite the fact that Democratic vote is split. Top-2 is apparently creating a lot of confusion for both reporters and readers, but in many ways, it isn’t that different than a regular primary when it comes to vote consolidation. Most Democrats will vote for the Democratic candidate in November, so comparing June vote totals is more than a bit confusing. Perhaps headline writers could do a better job on this front?
I mean, come on, do you really think this video at the top of this post is going to push Peterson to the win?