For reasons I cannot comprehend, few pollsters have surveyed Iowa voters since the first presidential debate. Even fewer Iowa polls have come out since the release of a 2005 videotape sparked the latest Donald Trump meltdown.

Forty years of data indicate that third presidential debates “have had less of an impact on the polls” than earlier debates. (Dan Guild reviewed here how first debates have affected previous presidential races.)

In lieu of a time-wasting “curtain-raiser” about things to watch for at tonight’s big showdown in Las Vegas, let’s look at what the latest opinion polls and election forecasts say about chances for Trump or Hillary Clinton to win Iowa’s six electoral votes. Last time Bleeding Heartland covered this territory, several analysts had shifted Iowa from “lean Democrat” to “toss-up” or from “toss-up” to “lean Republican.”

Aside from the latest Selzer & Co poll for the Des Moines Register and Mediacom, which Bleeding Heartland discussed here, just three Iowa polls of the presidential race have come out in October.

The Times-Picayune/Lucid 2016 Battleground State poll sampled 1,089 Iowans “who report both being registered to vote and likely to vote this November” between October 7 and 10, shortly after the infamous Trump video became national news. The methodology is unusual, so you may want to take this survey with a grain of salt:

Respondents are selected from among the nearly one million survey-takers who pass through Lucid’s technology each day. Results are post-stratified using simple survey weights to reflect Iowa’s demographics according to the US Census.

All respondents on the Lucid platform have volunteered to participate in a survey and were not chosen for participation in the Iowa poll through random selection. No sampling error can therefore be calculated.

Unlike reader polls that survey respondents who happen to be reading a specific article or website, Lucid’s platform finds survey-takers through a marketplace that diversifies where respondents come from and blinds the subject matter of the survey that each respondent will take. Lucid’s findings also takes into account each survey-taker’s demographics so that the model is representative of the population at-large.

Times-Picayune/Lucid’s survey found the largest lead for Clinton seen in any Iowa poll since the summer: she’s ahead of Donald Trump by 42 percent to 37 percent, with Libertarian Gary Johnson at 10 percent, the Green Party’s Jill Stein at 2 percent, and 9 percent undecided. UPDATE: On October 20, Times-Picayune/Lucid released corrected numbers from this poll: Clinton 41 percent, Trump 37 percent, Johnson 10 percent, Stein 2 percent.

SurveyMonkey and the Washington Post polled fifteen battleground states between October 8 and 16 (entirely after the 2005 videotape release) and found Trump leading Clinton in Iowa by 49 percent to 44 percent in a two-way race. Trump also led by 5 points in a four-way field: 45 percent to 40 percent, with 10 percent for Johnson and 2 percent for Stein.

UPI/CVoter sampled 322 likely Iowa voters between October 2 and 9 (some before, some after Americans heard Trump boast that he “can do anything” to women he finds attractive). Clinton barely led Trump by 48.7 percent to 47.1 percent, with 4.2 percent of respondents preferring unspecified other candidates.

I’ve always expected a close presidential result in Iowa, and I would not be surprised to see either candidate win. I’d feel a lot more confident if Democrats were putting up bigger absentee ballot numbers.

At last Friday’s Iowa Democratic Party “2016 Gala” (the event formerly known as the Jefferson-Jackson dinner), a common topic of conversation was how mortifying it would be for Iowa to go to Trump, especially since Clinton could carry a few states that went for Mitt Romney in 2012, notably North Carolina and Arizona.

At this writing, the only major election forecasting site to see Iowa (barely) leaning to Clinton is FiveThirtyEight.com, which updates its predictions multiple times a day. Today FiveThirtyEight’s “polls-only” forecast gives Clinton a 57 percent chance of winning Iowa. She’s at 53.6 percent in the “now-cast” and at 51.5 percent in the “polls-plus” forecast, which factors in economic indicators and historical data.

(OCTOBER 22 UPDATE: Reuters/Ipsos published a new electoral college “estimate of turnout” that rates Iowa a light blue, “low confidence Clinton” state, based on data that find her slightly ahead of Trump by 46 percent to 43 percent.)

Three forecasters currently see Trump or Clinton as equally likely to win Iowa. The Rothenberg & Gonzales Political Report rates Iowa as a “pure toss-up,” along with Florida, North Carolina, and Ohio.

As of October 14, NBC News puts Iowa in a much larger toss-up group, including Arizona, Georgia, Florida, Nevada, Ohio, and Utah (!), along with the second Congressional districts of Maine and Nebraska.

NPR’s Domenico Montanaro published his latest electoral college map on October 18, with the following toss-up states: Iowa, Arizona, Georgia, Florida, Nevada, Ohio, North Carolina, and the second Congressional districts of Maine and Nebraska.

UPDATE: On October 20, Larry Sabato released a new new “Crystal Ball” map, moving Iowa from “lean Republican” to toss-up, along with Utah and Maine’s second district.

Most national election forecasters now rate Iowa as “lean Republican”:

• Princeton Election Consortium (October 19)

• CNN (October 19)

• The Fix at the Washington Post (October 17)

• Associated Press (October 16)

• Larry Sabato’s Crystal Ball (October 13) (see update posted above)

• ABC News (October 5 version)

• Governing.com (moved from toss-up to lean Republican on October 5)

• Cook Political Report Electoral College Scorecard (September 29)

Spin your own scenarios about the presidential race in this thread.

UPDATE: Forgot to mention that the Talking Points Memo Electoral Scorecard also rates Iowa “lean Republican.”