2020 Washington Electoral Research Recap
2020 Washington Electoral Research Recap

Over the past six­ty hours, there’s been a lot of dis­cus­sion con­cern­ing how polit­i­cal polling this elec­tion cycle missed the mark, espe­cial­ly in states like Maine, where the avail­able body of pub­lic opin­ion research incor­rect­ly sug­gest­ed that entrenched incum­bent Repub­li­can Susan Collins was toast.

Fire the poll­sters has now become some­thing of a ral­ly­ing cry on Twit­ter and oth­er social media plat­forms, in vary­ing flavors.

“Fire poll­sters, hire local reporters,” tweet­ed Car­o­line Darya Framke.

“Fire poll­sters and replace them with film pro­gram­mers,” tweet­ed Adam Piron.

“Fire all poll­sters. Let’s not both­er with them in 2024,” tweet­ed Dale Maharidge.

“Fire all poll­sters and go back to read­ing ani­mal entrails the way God intend­ed us to pre­dict the future,” joked Kyle Smith.

“Polling seems to be irrev­o­ca­bly bro­ken, or at least our under­stand­ing of how seri­ous­ly to take it is,” Mar­garet Sul­li­van wrote in a Wash­ing­ton Post op-ed.

“The polling indus­try is a wreck, and should be blown up,” Politi­co declared.

While our team at NPI can cer­tain­ly under­stand the frus­tra­tion with impor­tant results not turn­ing out the way the research indi­cat­ed they would, not every poll­ster and not every poll got it wrong this year.

Con­sid­er our poll from three weeks ago.

Pub­lic Pol­i­cy Polling of North Car­oli­na sur­veyed six hun­dred and ten like­ly Wash­ing­ton State vot­ers for us on Octo­ber 14th and 15th. 45% par­tic­i­pat­ed via land­line, and 55% via text; yield­ing a mar­gin of error of +/- 4.0% and a con­fi­dence inter­val of 95%. We asked about every sin­gle con­test­ed statewide race on the bal­lot except for ESJR 8212, a pro­posed con­sti­tu­tion­al amend­ment, and found:

In each race, the can­di­date that we found to be ahead in our Octo­ber 2020 polling is win­ning, with no excep­tions, not even in the clos­er races, like Sec­re­tary of State or Super­in­ten­dent of Pub­lic Instruction. 

The races that were clos­est in our polling have all been called, either by the Asso­ci­at­ed Press or by observers (includ­ing Sec­re­tary of State, the clos­est, which pit­ted Kim Wyman against NPI’s senior board­mem­ber Gael Tarleton.)

In races where we had a low num­ber of unde­cid­ed vot­ers, even the mar­gins in our polling are pret­ty close to the cur­rent elec­tion results.

For exam­ple, in the pres­i­den­tial race, we found Biden with 60% sup­port, Trump with 37% sup­port, and 2% not sure. In the actu­al elec­tion, as of the third day of count­ing, Biden had 59.13%, Trump had 37.79%, and oth­er can­di­dates had less than 2%. That’s an almost per­fect cor­re­la­tion with our poll finding.

In the guber­na­to­r­i­al race, mean­while, Demo­c­ra­t­ic Gov­er­nor Jay Inslee cur­rent­ly has 57.74%; our poll had him at 56%. Repub­li­can chal­lenger Loren Culp achieved 40% in our poll and present­ly has 41.95%% in the elec­tion; it appears that he and Inslee split the small num­ber of vot­ers who were undecided.

It’s not unusu­al for there to be some devi­a­tion between poll find­ings and actu­al results; the fact that there is bare­ly any in this case shows that our sam­ple tru­ly was rep­re­sen­ta­tive of the 2020 Wash­ing­ton State electorate.

Just like it has been for sev­er­al elec­tion cycles now.

NPI’s part­ner­ship with Pub­lic Pol­i­cy Polling has been going strong for over half a decade, con­sis­tent­ly pro­duc­ing data that has fore­shad­owed elec­tion results.

Take the 2018 election.

In a poll con­duct­ed six months before the elec­tion, NPI’s research found Maria Cantwell with a six­teen point lead over Susan Hutchi­son.

This was the very first statewide poll pit­ting Cantwell against Hutchi­son; it was con­duct­ed with­in one hun­dred hours of the close of fil­ing by PPP for NPI.

Though half a year elapsed between the poll and the cer­ti­fi­ca­tion of the midterms, Cantwell went on to win by.… wait for it… 16.86 points.

The num­bers:

NPI’s research (May 22nd-23rd, 2018)

QUESTION: If the elec­tion for U.S. Sen­ate were held today and the can­di­dates were Demo­c­rat Maria Cantwell and Repub­li­can Susan Hutchi­son, who would you vote for?


  • Maria Cantwell: 52%
  • Susan Hutchi­son: 36%
  • Not sure: 12%

Election (November 6th, 2018)


Maria Cantwell: 58.43%
Susan Hutchi­son: 41.57%
Total Votes: 3,086,168

At NPI, we care about the qual­i­ty of data.

If the inputs are bad, then the out­puts will be bad (garbage in, garbage out).

We are emphat­i­cal­ly sub­jec­tive in our views as an advo­ca­cy orga­ni­za­tion — we’re very trans­par­ent about what our val­ues, prin­ci­ples, and pol­i­cy direc­tions are —  but as a research orga­ni­za­tion, we aggres­sive­ly strive for objec­tiv­i­ty when we field a sur­vey or anoth­er research instru­ment. A sci­en­tif­i­cal­ly accu­rate poll needs to have a sam­ple that is rep­re­sen­ta­tive of the electorate.

And neu­tral word­ing is essen­tial. Essential.

We can­not know what peo­ple’s views real­ly are on the issues if we were to ask loaded ques­tions that sug­gest their own answers.

Tim Eyman’s push polls — those false­ly-labeled “advi­so­ry votes” that keep show­ing up on Wash­ing­to­ni­ans’ bal­lots — are a per­fect exam­ple of how not to take the pulse of the elec­torate. Those are pieces of pro­pa­gan­da… utter­ly use­less for ascer­tain­ing what vot­ers real­ly think.

And by the way… most Wash­ing­ton vot­ers who have an opin­ion about Eyman’s push polls want them gone. We know because we’ve asked. Repeatedly.

Each time we have asked, it has been in the form of a neu­tral­ly word­ed ques­tion that pit­ted Tim Eyman’s best honed argu­ments against our own. Our argu­ments have beat­en his every sin­gle time, as this pre­sen­ta­tion explains.

Dur­ing the past few weeks, we’ve had more than one right wing com­menter show up and insist that our research was not cred­i­ble. For exam­ple, Mar­ty Wentz wrote: “your [sic] so full of it. Ins­ley [sic] has about as much a chance to win as I do.” Nick Kel­logg sneered: “I think this if [sic] a fake news and a fake poll.”

I imag­ine that Novem­ber 3rd was a rough night for Mar­ty and Nick.

Like oth­er Trump and Culp fans, they sad­ly dwell in their own “Upside Down” alter­na­tive uni­verse, where any­thing that does not fit with the par­ty line is imme­di­ate­ly assailed as fake and fab­ri­cat­ed, and con­spir­a­cy the­o­ries are rampant.

What this elec­tion has actu­al­ly proven that our research is pro­duc­ing use­ful, cred­i­ble data that can help any­one inter­est­ed in Wash­ing­ton State pol­i­tics under­stand how vot­ers per­ceive issues and candidates.

Even in a year of polling miss­es, our polling here in Wash­ing­ton State was on the mark. If you’d like to help fund our next sur­vey, we invite you to become a mem­ber of the North­west Pro­gres­sive Insti­tute or make a con­tri­bu­tion.

About the author

Andrew Villeneuve is the founder and executive director of the Northwest Progressive Institute, as well as the founder of NPI's sibling, the Northwest Progressive Foundation. He has worked to advance progressive causes for over two decades as a strategist, speaker, author, and organizer. Andrew is also a cybersecurity expert, a veteran facilitator, a delegate to the Washington State Democratic Central Committee, and a member of the Climate Reality Leadership Corps.

Adjacent posts