Reminder: Generic Ballot Polling Is Worthless

I'm only using Erick-Woods Erickson as the example because I happened to come across it this morning. His general advice to Republicans isn't terrible, it's the information he's throwing out as an example of how poorly Republicans are doing that's the problem.

Republicans take 41 percent on the generic ballot while the President sits at 37 percent approval.

 So what?

1. That's national polling. There are no national elections this November. There are 435 US House elections, 35 US Senate seats, and 36 state governorships up for election.
2. That's generic party preference polling. It tells us nothing about how many people, in which of those polities, like which candidates, none of whom are named "Donald Trump."

Every two years, generic ballot polling gets used to predict a "wave" for one party or the other, but it's not a reliable predictive tool for that, even to the extent that the polling is reliable and accurate.

Republicans are going to out-perform their party's national generic ballot polling in Alabama.

Democrats are going to out-perform their party's national generic ballot polling in California.

In close/"swing" districts, it's going to come down to the individual candidates and their campaigns.

Remember that every time you see a "blue wave" prediction. While I expect the Democrats to pick up seats in both House and Senate, and maybe even take majorities in one or both, we're talking about a handful of seats changing party.