By Robert Romano

One conventional wisdom headed into the 2018 midterms was that Republicans would have a very poor night and lose races they might otherwise win because females, specifically, suburban Republican females, were abandoning President Donald Trump and down ballot candidates.

There was only one problem. On election night, it didn’t actually happen. In states that are evenly divided, like Florida or Iowa, Republicans did about as well as Trump did in 2016.

According to 2016 CNN exit poll in Florida, Trump garnered 52 percent of men and 46 percent of women.

In a Nov. 2 St. Pete Polls survey that correctly predicted the outcome, Gov. Rick Scott (R-Fla.), running for Senate against Sen. Bill Nelson (D-Fla.), did comparably well along gender lines. Scott got 53 percent of men and 46 percent of women. Almost exactly the same.

In the same poll, DeSantis — who overperformed the poll’s result when voting actually happened — garnered 49 percent of men and 44 percent of women.

If there was some exodus of suburban Republican women from the GOP, it should have proven fatal to Scott and DeSantis in Florida, a state that could not be more closely divided politically.

In similar, statewide outcomes, Republicans held governorships in Iowa, New Hampshire and Ohio, all closely divided states. Surely, a flight of Republican women or just women generally from the GOP would have killed Republicans in those races, too.

In an Emerson poll that correctly predicted Gov. Kim Reynolds (R-Iowa) would be reelected by about 4 points, Reynolds garnered 57 percent of men and 43 percent of women. In the 2016 CNN exit poll in Iowa, Trump garnered 61 percent of men and 44 percent of women, practically the same.

In New Hampshire, Gov. Chris Sununu (R-N.H) garnered 56 percent of men and 45 percent of women in the last Emerson poll before the election. Sununu won. In the 2016 CNN exit poll, Trump got 53 percent of men and 41 percent of women. If anything, there, Republicans improved among both men and women since Trump got elected.

The Emerson poll in Ohio incorrectly predicted Richard Cordray would beat Mike DeWine, 49 percent to 46 percent. It was a bad sample that was too heavy-Democrat. It did show a drop off in support for Trump among men and women. There was just one problem. On election night, DeWine won handily, 50.7 percent to 46 percent. Trump won Ohio 51 percent to 43 percent in 2016. Pretty similar outcomes.

The same trend played out nationally, too, Manzanita Miller, an associate analyst with the Market Research Foundation, told me in an emailed statement, pointing to CNN’s 2018 exit poll.

“Of female voters who were already registered Republican, 93 percent of them voted GOP, and 6 percent broke for Dems. This is nearly equal to registered Republican men, where 94 percent voted GOP, and 6 percent broke for Dems,” Miller said, saying that she was “not seeing an indication of suburban females shifting to Dems at all.”

Which, it turns out, was better than Trump did nationally two years ago, where he garnered 89 percent of Republican men and 88 percent of Republican women in 2016, according to the CNN exit poll. No slippage there.

“The GOP won an equal share of suburban voters, and a greater share of rural households,” Miller added.

To be fair, there was some slippage overall, but it affected men and women equally. Whereas Trump got 52 percent of men and 41 percent of women nationally, Republican House candidates got 51 percent and 40 percent, respectively. The shift, if there was one, occurred among independents, where Democrats got 51 percent of independents who are men, and 56 of independents who are women.  Republicans got 44 percent of independent males, and 39 percent of independent females, compared to Trump’s 50 percent of independent men and 42 percent of independent women.

So, perhaps, there is some flight of suburban Republican women from the GOP that is hurting Republican candidates down ballot somewhere in America, it’s just that when you look for them in a poll or at the ballot box in 2016 and 2018, they’re nowhere to be found.

If the predictors are supposed to be party affiliation and gender, so far the phenomenon of Republican women leaving the GOP remains hypothetical.

They are certainly not appearing in some of the most closely divided and contested states with huge suburban presences that Republicans depend on to win where you would expect to see an impact.

Which makes it unicorns. A fantasy, more or less. Or, a goal or hypothesis of Democrats of how to flip Republican women to vote Democrat because they’re supposed to be just so disgusted with the President.

Here’s an idea, maybe if Democratic leaders and pundits stop calling their husbands racists and sexists — just maybe — Republican women will consider voting for the Democratic candidate. Just a thought.

Narratives don’t vote. People do. Something to keep in mind as we head into 2020.

Robert Romano is the Vice President of Public Policy at Americans for Limited Government. You can read more of his articles at