The first thing I recall learning in grad school that was truly eye-opening was the extent to which individuals' evaluations of the economy, foreign policy, and just about anything else are a function of partisanship. Like most people I had always assumed that with all the proxies available – gas prices, bank rates, one's own paycheck, businesses opening and closing locally – an adult could easily develop a "sense" of how the economy is doing.
Granted, that kind of evaluation isn't a measurement. It's a feeling. And that's fine, because what survey questions ask is how the person *feels* the economy is doing. And there's nothing simpler than announcing your seat-of-pants guess at how things are.
Yet when you look at the research, one of the most consistent findings in public opinion is that even these rough feeling type evaluations are almost purely partisan. Few voters even use the easiest proxies ("Gas prices are going up! Economy bad!"). It really is as simple, for many voters, as my party is in power therefore the economy is great. The other party is in power so no matter how many raises I get the economy is bad and my taxes are going up.
Brian Schaffner posted this neat time series from the Cooperative Congressional Election Studies (CCES) – my favorite data to "play with" in all of social science – showing partisan evaluations of the economy from 2006 (regrettably, the first year of the CCES) to 2017. The results are unsurprising and demonstrate the basic point I made above.
Note that after a rare period of tri-partisan agreement that things were going to shit in 2007-08, Democrats' sense of the economy rebounded more sharply and rapidly once Obama took over. Now look at 2017.
Another thing political science has a large amount of evidence for is the contention that people who identify as independents often have highly partisan voting habits. They might not say they're Republicans, but they very well might vote exclusively for Republicans. A lot goes into Independent identification; it expresses dissatisfaction with both parties (Many Americans, reasonably, have concluded that they don't like either irrespective of policy positions) and boosts a sense of oneself as an open-minded, free thinking, doesn't take orders type of badass.
In practice, of course, most of these people vote a party line or close to it because that's what people do. They get their ballot, realize they don't know anything about 99% of the names on it, and vote using the easiest available cue.
The CCES data here demonstrate pretty clearly a point I'm often trying to make, which is: Don't "independents" in this data look an awful lot like Republicans? Like, suspiciously so? Almost like Independents are really Republicans who don't want to, for whatever reason, think of or announce themselves as Republicans?
The rise of Independent identification in survey data – and it has risen sharply in recent years – is the worst thing ever to happen to political consultants. They see the word "independent" and believe that it signifies a large mass of undecided voters just waiting to be hit with the right message to sway them. For some independents that may be so; various estimates in research suggest that maybe 1 in 10 voters falls into this category and genuinely does not make up their minds until late in an election. For the majority, though, there is no meaningful independence. It's just a label.
Chuck Schumer famously said in 2016, “For every blue-collar Democrat we lose in western Pennsylvania, we will pick up two moderate Republicans in the suburbs in Philadelphia, and you can repeat that in Ohio and Illinois and Wisconsin.” That election helped demonstrate the flaws not only in targeting opposing partisans (hint: moderate Republican is still Republican. It says Republican right in the name) but in assuming that Independents are Independent. Generally they aren't. This is well understood.
So, any election strategy you hear that references independents, swing voters, or any synonym for a big group of persuade-able voters is not a strategy that makes sense. Any winning candidate who credits it fundamentally misunderstands what happened, and losing candidates who appeal to it are likely to repeat their mistakes.