Losing by Winning

Jon Chait takes on the topic of elections it's better to lose, noting that liberalism would probably have been in much better shape had Gerald Ford been re-elected in 1976 leaving the GOP, rather than Jimmy Carter, saddled with the essentially unsolvable problems of America in the late 1970s:

And the elections that people think don't matter often do. Moderates and liberals widely regarded the 2000 presidential campaign as a snoozer. Apathetic liberals held "shadow conventions" that year to highlight the stultifying timidity of the two major parties. The implicit premise of Ralph Nader's 2000 candidacy was that it was as good a year as any for liberals to make a protest statement and throw the election to a Republican. We now can see that the radicalism of George W. Bush, then half-concealed, along with the rallying effect of Sept. 11 made the 2000 election incredibly consequential.

This consideration of the little-appreciated-at-the-time significance of the 2000 election, however, is the reason why I don't think it ever makes sense to do anything other than try your best to win. The 2000 election turns out to have been incredibly important primarily because of 9/11. The giant external shock created a public expectation of dramatic policy shifts which, of course, made it much easier than it otherwise would have been to implement such shifts and far harder to obstruct or prevent them. Thus, a kind of latent radicalism inside the Bush administration was unleashed in a way it otherwise wouldn't have been. Events, in short, are incredibly important and one of the main things presidents do is respond to them. Meanwhile, it's just not possible to know in advance which four-year periods are the ones that are going to feature dramatic events.