Chris Christie’s campaign has been showing signs of life in New Hampshire of late. He could already point to his rising net favorability rating among Granite State Republicans, and over the weekend he added the New Hampshire Union Leader’s endorsement. The backing of the state’s biggest paper has historically been a leading indicator of how New Hampshire Republicans are likely to vote. Whether that will prove true for Christie is an open question.
Up until the paper backed Newt Gingrich four years ago, every candidate the Union Leader endorsed since 1980 gained in the polls afterward. Here are the candidates the Union Leader endorsed, their average in New Hampshire polls before the endorsement and their vote percentage in the state’s primary:
|ELECTION||ENDORSEE||STANDING IN POLLS BEFORE ENDORSEMENT||PRIMARY RESULT||DIFFERENCE|
|1988||Pete du Pont||4||11||+7|
Note that even when pre-Gingrich candidates didn’t win, they picked up considerable support after the endorsement. On average, the Union Leader’s preferred candidate outperformed their polls by 8 percentage points (that includes Gingrich).
Of course, 8 percentage points isn’t exactly winning the lottery. Christie currently has about 5 percent in New Hampshire polls. An extra 8 percentage points would put him at 13 percent. That probably won’t be enough to win the primary. Christie would need to pick up closer to the 21 percentage points that John McCain did in the 2008 campaign to win. And even if Christie did that, 26 percent would still be less than the percentage that every single New Hampshire primary winner received since 1972.
Also, the Union Leader’s endorsement didn’t foretell a Gingrich surge. After he was endorsed by the paper, Gingrich lost 10 percentage points.
So where does that leave us? Should the Christie campaign be popping champagne corks? Quietly optimistic? Indifferent? It depends on how you interpret the data.
- The Gingrich flop may show the newspaper’s endorsement didn’t mean anything to begin with and it was all just statistical noise. We’re working with a sample size of seven, after all. As FiveThirtyEight editor-in-chief Nate Silver wrote after the paper endorsed Gingrich, “a regression analysis on historical data is not really the same thing as a prediction of how these factors will play out in the future. Fairly often, a relationship that is found to be highly statistically significant in past data will prove to be unreliable when applied out-of-sample.”
- Maybe the Union Leader endorsement did pack some punch in the past but has lost its power as newspaper circulation has dropped. Voters can now receive their news from many sources. That means fewer Republican voters will read the Union Leader, let alone take the paper’s advice.
- On the other hand, the previous correlation between the Union Leader’s endorsement and candidates’ improvement in the polls could have been indicative, not causative. That is, as Nate put it four years ago, the paper may “replicate in some way the thinking process that some segment of New Hampshire voters will [eventually] go through, whether or not they pay any attention to The Union Leader itself.” In 2012, the endorsement may have been more indicative of conservatives’ thinking about Gingrich before he collapsed in Iowa, which may have sapped his momentum in New Hampshire. (Christie likely won’t have the problem of underperforming expectations in Iowa — he is polling at just 2 percent there right now.)
- Maybe 2012 was a fluke and the Union Leader endorsement is a good sign for Christie, whether it’s indicative, causative or a mixture of the two.
At the least, the Union Leader endorsement doesn’t hurt Christie in his quest to convert his high net favorability ratings in New Hampshire into votes, and it could be a good sign that he’s on his way to doing so.