In my last column, I wrote about Donald Trump’s job approval ratings among Republicans, Democrats and Independents, noting that his standing among all groups is important. But my primary focus for 2018 should be on Independents, who are likely to be the canary in the coal mine for the midterm elections.
How Independents vote, either on the presidential ballot or the national House ballot (depending on whether it’s a presidential year or not), turns out to be a pretty good indicator of the type of election we are going to have.
Since 2000, Independents have sometimes broken evenly between the two parties (2000, 2004, 2012 and 2016), sometimes given one party a clear advantage (2008 and 2014) and sometimes showed an overwhelming preference toward one party (2006 and 2010), according to exit polls.
When Independents break roughly evenly between the two parties, net House gains/losses are small. But when those voters overwhelmingly prefer one party over the other – as they did in 2006 when Independents heavily preferred Democrats and 2010 when they preferred Republicans – a partisan wave developed.
Since 2000 (and excluding 2002, when the exit poll was tainted), there have been four elections – all of them in presidential years – when Independents split roughly evenly.
Republicans had a narrow four-point advantage among Independents in 2016 (46%-42%), a five-point advantage in 2012 (50%-45%) and a two-point advantage in 2000 (48%-46%). Democrats had a mere one-point advantage in 2004 (49%-48%).
Not surprisingly, all four of those presidential contests were very close. Donald Trump lost the popular vote by just over two points, and Barack Obama carried it by just under four points. George W. Bush won by 2.4 points in 2004. Four years earlier, the presidential race was a virtual popular vote dead heat.
In other words, the tight contests among Independents reflected the tight races nationally.
Two elections, both midterms, were clear blow-outs.
In 2006, Democrats gained 31 House seats, in part because they carried Independents by a stunning 18-points (57%-39%), according to the national House exit poll.
Four years later, a Republican tsunami produced a GOP gain of 63 House seats. Independents preferred Republicans by 19 points (56%-37%).
Part of the reason for the two waves is that Independents acted more like partisans than they normally do. Their strong preference for one party or the other contributed to the wave but also reflected broader sentiment among the larger electorate.
Two elections, 2008 and 2014, were neither waves nor close contests.
Independents went Democratic by eight points in 2008 (52%-44%), which produced Democratic House gains of 21 seats. In 2014, Independents preferred Republicans by twelve points (54%- 42%), which resulted in a GOP gain of 13 seats.
So, how does all this history play in 2018?
Democrats need to net at least 24 House seats to take back the chamber, which means that they will need something approaching a partisan wave because of the way congressional districts were drawn at the beginning of this decade.
If you see Independents breaking narrowly toward Democrats (for example, in national generic ballot tests), you’ll know that it’s unlikely the House will flip. But if Independents are going heavily Democratic, in the 15- to 20-point range, the House is much more likely to switch party control.
Midterms have tended to produce much wider swings among Independents than have presidential contests, and that trend is likely to hold again in 2018. But we don’t yet know how big the Democratic wave will be.
Independent voters’ attitudes should be an early warning sign for both parties about the midterms. With any luck, a major media organization or two will conduct national polls with an oversample of Independents to help us track their preferences and intentions.