As I suggested, those New Hampshire numbers look quite unwelcome to me. It's a small state, not a lot of electoral votes. But if it's tight that could be critical. It was in 2000. If Gore would have won New Hampshire, Florida would not have mattered.
But here's another basic question we're going to find the answer to in a few days. Ground games and field operations are very important. In recent years, Democrats, through traditional organizing strategies turbocharged by data analysis, have become the masters at this. This isn't a boast from the Democrats. Professionals in both parties recognize this. By and large Clinton has taken on the Obama data and organizing infrastructure and team - if not always the same people, the same approaches, technologies, strategies.
But this year there's something different. It's not just that Democrats have a stronger ground game. It seems like in critical ways the Republicans don't have one at all. They say this is not the case. But even though Trump has basically outsourced field to the RNC, that's just not how modern campaigns work. It seems like the Trump/RNC field and data operation is probably somewhere between close to non-existent and significantly below the level of McCain and Romney. In other words, there are a lot of reasons to believe it's going to be a total mismatch on the ground.
Now, let's be clear. Most people who are going to vote know to vote and they do vote. Ground and field operations aren't magic. If you're down ten points you are just definitely going to lose. But they can realistically grind out 1% to 3% in a final outcome. In a close election that's the world, even in a 4 or 5 point race it can be the world.
What we've never seen is a modern Obama-era field operation go up against no operation at all - or one that was put together late and on the cheap. We don't know the latter is the case - all the but no operation versus just a weak one. But if the GOP field operation is as underpowered as a lot of evidence suggests, that's a wild card that could defy expectations.
Let me be clear on this point. I am not saying ignore the polls because there's this great field operation and it will make everything okay. That is not the case. If you ignore the story the aggregated poll data is telling you you're fooling yourself. The real data we want to see is the campaigns' internal polling which is more sophisticated and I suspect more accurate. But we don't have access to that. That gives us the public polls, which are unideal but a fairly good measure when aggregated together.
But just keep this field operation wild card in mind as we go down to the wire.