Having spent waaay too many evenings behind the glass at focus groups, I'd make this observation/caveat: People in them tend to say what they think you want to hear, or just want to be all 'Alpha Dog' and be contrarian. Either way, marketers use the results for all-purpose ass-kevlar, so they can say"Science says!" As an example of how they get it wrong, the Absolut Vodka campaign (Absolut _______) died an eviscerating death in research. But the client said, 'screw it, let's do it'. It ran for almost two decades. So much for focus groups there.
Political focus grouping is largely about determining message track. Generally, it would happen prior to the writ being dropped. It's really, really hard to focus group a change in message in the middle of a campaign (That said, Kinsella's forte has largely been rapid-response as conditions change). Hence, Olivia sticking to this seemingly non-sequitur nonsense about strollers and buses and whatnot. They appear to have no Plan B and are sticking to the focus-grouped nicey-nice they gathered months ago. People would have legitimately expressed approval of high-ground, positive messaging back then. So the campaign goes with it. Meanwhile 3-4 weeks before the election, the dynamic changes, the unexpected happens, and reality roils all around them. Any responsive messaging would likely be un-tested and they may be hesitant to take a flyer.
At least, that's what personal experience suggests. YMMV.