1) This is probably a missed opportunity for the Romney campaign. There are really two things that a vice president can do: Move the needle a few points in a key state (LBJ in 1960, Lieberman in 2000) or reinforce a message (Gore ’92). With a choice like Sen. Rob Portman of Ohio or Gov. Bob McDonnell of Virginia, Romney really could have made a difference in a swing state that could have put him over the top against Obama. Maybe Ryan will reinforce a positive message for Romney that allows him to win, but I’m still skeptical.

2) This probably improves Obama’s chances of winning. The Ryan plan doesn’t exactly have a great track record winning elections: It played a large role in the Republicans’ defeat in a special election in upstate New York in early 2011. Romney’s path to victory involves winning an outsized share of downscale white voters, and this plan presumably makes that task more difficult. Again, maybe the focus-group testing shows that opposition to the president is so strong among these voters that it just doesn’t matter, but I don’t think I’d bet the farm on this.

3) In fact, it opens up an Obama landslide scenario for the first time. I’ve always thought that Obama wouldn’t be able to win more than a two-to-three-point re-election victory, mainly because a president almost never wins the votes of people who disapprove of the job that he is doing, and Obama’s approval rating is unlikely to be much above 50 percent on Election Day. But, while I don’t think it’s guaranteed, this really does give Democrats an opportunity to make Romney so radioactive that people who don’t like the president nevertheless vote for him. If the white working class revolts at the prospect of the Ryan plan, Obama really could match, or even exceed, his 2008 showing.

4) These types of picks rarely end well. When we think back on the “bold” or “unexpected” picks in history, they rarely have good outcomes. Agnew in ’68, Eagleton in ’72, Ferraro in ’84, Quayle in ’88, Kemp in ’96, Palin in ’08 are all looked back on with disfavor. The “good” choices were almost always “safe” choices.