Since senatorial announcements of newfound support for gay marriage have gotten dull, the hot new trend in politics is to try and predict which senator will be next to come out into the open on the issue. Some, like the Washington Post's Sean Sullivan, made predictions based on their guts. But if we learned anything during 2012, it's that data always trumps gut calls. So we looked at the data.
We had three theories of what state characteristic might serve as a predictor that its senator was about to change his or her mind. The first was how Obama fared in the state last November, that there might be correlation between how Obama did electorally and the order in which senators declared support on the issue. The second was household income, that senators' shifts correlated to socioeconomics. And, finally, that the correlation was between the density of same-sex couples in a state.
Obama's margin of victory
A quick description of methodology is in order. The blue line on each chart below (pegged to the left axis) indicates (using data very helpfully compiled by the Post) the number of days between a senator's announcement of support and the first senator to make such an announcement: Barbara Boxer, in June 2006. (This isn't quite true; Ron Wyden announced in 1996, but that threw off the curve.) So the flat part at the right of the blue curve is the recent spate of announcements. We included both interactive and static versions of the charts because we wanted to include a trend line — that dashed red line in the second chart below. (For the truly curious, that line is a fifth-order polynomial.)
And so: The relationship between announcements of support and how Obama fared in the state. What we're looking for is a dashed line that clearly descends over time.
Data from Washington Post.
It seems clear that how Obama fared in a state is a good indicator of when a senator is likely to announce support for same-sex marriage. Which isn't a surprise, of course. Politicians are very reactive to political support. And the more political support demonstrated for a candidate that supports same-sex marriage, the more likely another candidate would be to take the same position.
If this model is correct, the next five most-likely announcements would come from:
- 1. Susan Collins, Maine
- 2. Ron Johnson, Wisconsin
- 3. Dean Heller, Nevada
- 4. Chuck Grassley, Iowa
- 5. Kelly Ayotte, New Hampshire
These are all Republican senators. Which isn't a big surprise; it makes sense that states that voted for Obama but have Republican senators would most likely be what's left. The four Democrats who still haven't endorsed — Mary Landrieu of Louisiana, Tim Johnson of South Dakota, Mark Pryor of Arkansas, and Joe Manchin of West Virginia — would be the 23rd, 25th, 38th, and 39th most likely to do so under this model.
Please note: If Heller or Grassley declares his support for gay marriage, send warm coats to Hell.
Verdict: No correlation.
The line here is probably too flat to use as a guide of who might be next to endorse the concept. So we can skip our predictions.
Density of same-sex couples
Verdict: Hard to tell.
There appears to be a loose correlation between the density of same-sex couples in a state (as a percentage of all households, calculated by the Census) and how rapidly a senator will declare his or her support. This may, then, be a good test case. If one of the following five senators is next to declare an evolution, we may be on to something.
- 1. Susan Collins, Maine
- 2. Dean Heller, Nevada
- 3. Kelly Ayotte, New Hampshire
- 4. Marco Rubio, Florida
- 5. Jeff Flake, Arizona
Rubio and Flake are interesting cases. The former is going to be abnormally sensitive to tricky political situations, given his ambitions. The latter recently said he expected a pro-gay-marriage Republican presidential candidate in the future, though he likely didn't mean Rubio.
All of this data, of course, correlates to a bunch of other things, like urban density for example. Consider it more of a thought exercise than anything.
If you're curious, the Post's Sullivan made the following predictions, in order: Lisa Murkowski, Alaska; Susan Collins, Maine; Richard Burr, North Carolina; Kelly Ayotte, New Hampshire; Pat Toomey, Pennsylvania. Somehow Dean Heller didn't make the cut, but otherwise not all that different than our list.
Meanwhile, Nate Silver of 538, the man who taught us all about the supremacy of data, did his own analysis of the data. Among his conclusions:
Some senators will continue to oppose it, either because it does not yet constitute a majority position in their states (like Senator Heidi Heitkamp of North Dakota, they may say it should be decided at the state level), or because they oppose it on moral grounds, or because they are more concerned about a primary challenge than the general election.
Heitkamp announced her support today. Maybe there's something to this gut thing after all.