OPINION — Did the 2018 midterm electorate break new political ground as the media had predicted for months or was it déjà vu all over again? The answer is both. 

In my last column before the election, I suggested that four key measurements would tell the story of this year’s midterms: party ID, ideology, turnout by unique voter groups (young voters and women), and how independents break.

So, what happened? Two key stats didn’t change much — party ID and ideology.

Party ID

In recent elections, the party out of power in midterm elections has usually done slightly better than in presidential years when it comes to the party identification of the electorate, with 2002 being an exception due to 9/11. 2018 was no different. The party ID margin in 2016 was +3D (36 percent Democrats to 33 percent Republicans). Two years later, the party ID was quite similar at +4D (37 percent to 33 percent). Not exactly a blue wave.

Also Watch: Poll Volunteers of All Ages, Backgrounds Give Back on Election Day

Ideology

We didn’t see evidence of a blue wave in self-defined voter ideology either. In 2016, 38 percent of voters identified as moderate, 35 percent as conservative, and 26 percent as liberal. In this election, those numbers were almost exactly the same: 37 percent moderate, 36 percent conservative, and 27 percent liberal. However, this is both the lowest number for moderates and the highest number for liberals looking at every election since 1984.

Women and young voter turnout

If turnout data could send Republicans a message, it would be, “Houston, we’ve got a problem.” Three problems, actually. Women, young voters, and independents. But those problems don’t derive from turnout. Yes, the 2018 turnout looks to be larger than recent midterms, and the data appears to show that key demographic groups, across the board, voted in higher numbers. But here’s what’s important: The composition of voter groups as part of the national electorate was roughly the same.

Take female voters, for example. More women turned out to vote than in 2014, but their percentage of the electorate — normally around 52 percent — remained roughly the same. The 18-29-year-old vote, which has historically accounted for about 13 percent of the electorate in midterm elections, also seems to have hit roughly the same mark this year.

But that’s where the similarity ends. This year, women voted Democratic by 19 points (40 percent R to 59 percent D). Going back to 1984, this tops the previous largest Democratic margin among women by 5 points and almost doubles the 2016 margin of 10 points (44 percent R to 54 percent D). This is also in dramatic contrast to 2010 when Republicans won the House from Democrats by actually winning the women’s vote 49 percent to 48 percent.

Likewise, 18-29-year-olds voted Democratic by a huge margin of 35 points (32 percent R to 67 percent D). This lopsided result even surpasses the 29-point margin seen in the 2008 election of Barack Obama, and was two and a half times larger than 2016 when young people voted Democratic by 14 points (42 percent R to 56 percent D).

Independents

In 2016, Republicans won 241 seats in the House, carrying independents by 6 points (51 percent to 45 percent). In 2018, Republicans appear to have lost them by 12 points (42 percent to 54 percent), costing them about 30 seats — with some yet to be called. We’ve seen this movie before in 2006 when Republicans lost independents by 18 points (39 percent to 57 percent) and with that, the House and Senate.

In 2006, Republicans relied on a negative base strategy that was ineffective in connecting with independents. This year was a repeat. In a number of my previous columns, I argued that Republicans needed to put their focus on voters and the economy to make what has almost always been “the No. 1 issue, the No. 1 issue.”

Yet, Republicans split their issue emphasis, concentrating on the economy and immigration. In the exit poll, from a choice of four issues, 23 percent of voters cited immigration as their top issue and 22 percent opted for the economy. This decision gave Democrats an opening to focus on health care, which ended up the top issue of the election at 41 percent and allowed them to avoid discussing the economy.

Republicans found themselves at a strategic disadvantage without a persuasive legislative record on the rising cost of health care or an effective response to the Democrats’ health care argument. With significant presidential focus on immigration, they ceded their best argument for re-election — a booming economy — relying instead on a tough immigration message and anti-Pelosi ads, which work great with the base but not as well with independents and women.

According to the exit polls, 68 percent said the economy was either excellent or good, in contrast to 2016 when only 36 percent rated the economy positively. But because Republicans didn’t engage the electorate on the economy at scale, they only managed to win the 51 percent of voters who cited the economy as being good.

When it comes to the midterms, some variables remained the same in 2018. Party ID, ideology and the composition of key voter groups as a percentage of the electorate. However, Republicans losing independents by 12 points, women by 19, and younger voters by 35 is not sustainable. They can turn this around by focusing on Republican policies that are fueling a robust economy, creating jobs and higher wages. This would be a good first step in creating a positive context for 2020.

(Note: This column is based on the 2018 exit poll numbers that have been reported but are still being finalized and may vary slightly when completed.)

David Winston is the president of The Winston Group and a longtime adviser to congressional Republicans. He previously served as the director of planning for Speaker Newt Gingrich. He advises Fortune 100 companies, foundations, and nonprofit organizations on strategic planning and public policy issues, and is an election analyst for CBS News.