Friday, October 26, 2012

A Bit More on Momentum

I had a post up yesterday over at Plum Line talking about the myth of momentum, and why we shouldn't expect change in one direction in presidential general elections to build on itself. You should also read John Sides, who talks about media coverage, and Nate Silver, who supplies a useful definition and a bit of empirical evidence. What I add to the mix in my post is why we shouldn't expect the groups who are responsible for changes in polls to yield continued changes.

Just one quick addition to all of that: just because momentum is unlikely in the current election doesn't mean it can never happen. Kevin Collins points out that momentum makes sense in multicandidate contests when strategic voting comes into play. That's true.

Momentum also may be relevant in primary elections. Not just because of strategic voting (learning that my favorite candidate has little chance may shift my vote to my second-favorite, in order to beat the candidate I really don't like), but also because of the information environment and the nature of the electorate. First, we have an electorate of generally unattached voters who could vote for any of a number of candidates. Then, a candidate who wins a primary typically get the lion's share of press coverage immediately after that event. I may, as an unattached voter, gravitate towards that winner just because she's the only candidate I'm really aware of that week. Indeed, that dynamic could easily play out in a single-election primary (such as a Senate primary) with a positive poll, debate performance, or financial report serving as the event which allows one candidate to dominate the information environment. Note, however, if parties are able to dominate the information environment, then you should get a lot less momentum.

The main idea, at any rate, is that for "momentum" to work it requires the first set of changes to then lead to further changes -- which requires some pot of voters who will react either will react, but only more slowly, to the cause that produced the first set of changes, or a pot of voters who will react to the change itself. So find conditions where that may work, and you'll find possibilities for momentum.

23 comments:

  1. People assume the polls are correct in that they measure actual votes... They don't even likely voter polls just show enthusiasm. When people respond to polls only the most over the top supporters respond. How do we know this? Because only 3 to 4% of all people called respond to polls. It makes the loudest voice the most heavily polled voice. It's why Richard Nixon kept winning with the silent majority yet polls were all over.

    The media will report on Nov 7th that all the undecideds must have broke for the President , but the fact is that if you look at the projected turnout there are more people who favor the president over Romney , but simply are not gung-ho. They'll still vote, but they get weeded out of Likely voter models.

    It's the turnout machine and the overall status of the nation being about 60% left leaning that will carry the election.

    ReplyDelete
  2. The reason the specious "momentum" concept beguiles political-news media is because, by intellectual disposition and by institutional incentive, they overwhelmingly evaluate public matters according to imputed campaign competence and marketing appeal. To use other paradigms would make more obvious the ways in which they disclaim the obligation to engage with policy or non-hysterical reasoning. Image is everything, and success no matter how ephemeral is assumed to be self-sustaining.

    ReplyDelete
  3. Momentum probably makes more sense in a place like the UK where we have a 2.5-3 party system. There are so many areas where instead of a Labour vs. Conservative race it's one of those two vs. Liberal Democrats so tactical voting can make a big difference.

    It's also dangerous though, after our first ever TV debates, the Lib Dem leader got a huge bounce and saw his party jump into first place in the polls but this caused them to divert scarce resources into areas which really weren't winnable, resulting in them overstretching themselves and actually loosing seats.

    ReplyDelete
    Replies
    1. Hang on. Firstly, our polls don't really screen for likely voters very well. Clegg's big jump was almost entirely among groups that have very low turnout. Presumably these were low-information (non-)voters who were drawn in by the novelty of the very first TV debates. This is backed up by my anecdotal experience as a teller.

      Secondly, I don't think they lost seats due to putative "overstretching." The Conservative vote increased substantially, and that was always going to win seats from the Lib Dems. In fact, considering their false position from the 2005 election, with the effect of the Iraq War, I am amazed the Lib Dems did as well as they did in 2010, and I think a lot of that comes from the positive effects of "momentum" from the debates.

      Delete
  4. It's interesting to read a technical discussion of "momentum" in electoral politics. When the Romney "momentum" theme first appeared a week ago I took it mildly seriously until after the final debate when 1) commentators noted that there were no more opportunities for either candidate to reach a huge national audience and 2) there was no evidence of movement in the polls. Now it's clearly becoming a joke... Jennifer Rubin's twitter feed has been an excellent example. Now when I hear "momentum" I hear "Romney is losing."

    ReplyDelete
  5. But what about Joementum?

    ReplyDelete
  6. A few weeks ago JB said that the obsession with describing politics as "right" or "left" is kinda silly because using terms from the French Revolution to talk about American politics in the early 21st Century isn't particularly helpful. And I think the same thing could be said about use of terms from the study of physics. Using my thoroughly unscientific method of using a post it note pad as a straight edge to look at how Obama has been going up on on the 538 "Nowcast" since the Biden debate and then projecting where that line will end up on election day, I'd say Obama has the momentum.

    Joementum! Yes!

    ReplyDelete
  7. Because the echo chamber is wrong in asserting that Romney's early October momentum will continue indefinitely, it follows that momentum doesn't exist? Heck, even the most sabermetricy baseball fan knows momentum doesn't work that way.

    What does your team do when your opponent gains momentum? Maybe you go to the rally caps. If you're on the road, you try to take the crowd out of the game. In several ways, you attempt to take back the momentum. So it likely is with political campaigns, in fact perhaps the lack of persistent momentum reflects the other side's efforts to change/blunt momentum, a campaign effect. Its easy to imagine that Team Obama was getting a bit complacent in late September that this thing was over; its also easy to imagine they've been acting with quite a sense of urgency for a few weeks. The opposite might well be true for the Romneys; thus momentum ebbs and flows.

    Collins' observation of momentum persisting in a multiparty primary may also reflect campaign efforts to stop momentum (or not). For example, if you're Romney and Santorum is gaining momentum in the run-up to the Super Tuesday confederate primaries, you may well encourage the Santorum mania as a way to knock out Gingrich before a southern embarrassment. The decision calculus is complicated about responding to one of several competitors gaining momentum in a multicandidate race - head to head its always best to directly confront the other's momentum. I'd guess that's what campaigns, like professional sports teams, do - and so momentum rarely lasts in a campaign, as it often doesn't in sports either.

    ReplyDelete
    Replies
    1. I sincerely right now wish that momentum existed in baseball, but I've never seen any evidence at all that it does. In baseball, or football, or basketball. I mean, obviously teams have streaks both in-game and between-games, but there's no evidence I'm aware of that anything like momentum exists.

      And to get to what Couves said below -- before the first debate is when all that unskewed nonsense was preventing despair. I'm sure Team Romney feels a lot better now, and obviously they have a better chance to win when they're even in the national head-to-head than when they were down 3-5, but other than that I'm not sure why we should believe that it's helping them.

      Delete
    2. So other than giving Romney "a better chance to win," momentum doesn't matter?

      Delete
    3. But what is "momentum?" Romney is doin better in the polls now compared to a month ago... But he isn't doing better now compared to a week ago. So where's the momentum? And, in terms of judging why he's doing better now compared to a month ago, do we blame that on momentum or on the first debate/better public outreach?

      Delete
    4. Momentum is the wrong word. It's possible in a situation like this that there's a tipping point, or a preference cascade, or whatever you want to call it. It's possible that, for an incumbent, once they've lost voters this far in, there's no hope of getting them back and it becomes a matter of how well they can hang onto other marginal voters. But the word "momentum" would be the wrong way to describe it, even though that's the kind of thing some people may mean when they say it.

      Delete
  8. Romney is definitely benefiting from momentum. Before the first debate, Republicans were pretty much in despair. You can't expect to win if your own supporters have lost hope.

    ReplyDelete
  9. What you say about his supporters has some merit, Couves. You could also say that unless Democrats are terrified that a Bush clone will seize power again, that they won't bother to go to the polls. Let's agree that the most crucial aspect of momentum has to do with breaking the 50 percent mark in enough swing states to take th EC. Lacking that, Romney doesn't have "momentum." whatever it may be.

    ReplyDelete
    Replies
    1. Anon, it's possible to gain momentum, and still come in second place.

      Delete
    2. You mean, you gain momentum only to smack into a wall? Ouch.

      I think the use of the word "momentum," like the use of the word "surge," are largely for dramatic effect. They sound so much more hopeful than "reversion to the mean." So we're back to your original post. The beneficiaries, though, are not the Republican rank and file. They are the independent Republican campaign groups, who most certainly must show some results if the money is to keep rolling in.

      Delete
    3. Reversion to mean? You make it sound like Romney's improving poll numbers were inevitable. They were not. They're the result of 60 million people watching the first debate and many of them saying something like "Hey, Romney looked pretty good, but Obama doesn't even seem to want the job anymore." Romney starts going up in the polls, Republicans see victory within their grasp and redouble their efforts, the media narrative begins the reflect the new reality of a tight race, and so on... A single event set in motion other events that would not have otherwise occurred. That's what distinguishes a campaign with "momentum" from one that follows some sort of natural and predictable course.

      Delete
    4. Agree with general narrative you lay out but with the caveat that political scientists were saying that the race was going to tighten. Yes those people were impressed by Romney and may have switched vote as result. But they would have switched to Romney eventually no matter what due to economy, fundamentals of the race, etc. to think about it another way: if the economy had been growing at a faster pace last year, those folks would have watched Romney, been impressed, and still voted Obama.

      Delete
    5. So Romney was destined to get a bump in the polls? Ok. Since there's no way to prove or disprove that, I guess we'll just have to agree to disagree.

      Delete
    6. To the Couves comment at 4:15:

      I agree that something happened, probably debate-related (although it's not certain, but likely) between late September and about October 10 -- but it may have been just the debate, happening rapidly over October 4-6.

      What I disagree with is that "and so on" part of it. Whatever movement to Romney there was ended. In fact, the current HuffPollster trend line shows that Romney moved into a tie on October 5, and that's it; since then the two have been almost dead even.

      I don't see how that fits "momentum."

      Delete
  10. I missed the last part of this argument, but I find the assertion that momentum doesn't exist a bit strange. We know for certain that variance exists, yes? For example, Posey had a .957 OPS this year in 610 PAs, and we know that .957 OPS was the sum of 610 or so individual OPS's; that few if any plate appearances of Posey's are as a ".957 OPS" guy; .957 is the mean of all the "different" Poseys that came to the plate 610 times.

    We also know that sometimes the good Buster is correlated with his other teammates doing well, sometimes his success is related to his teammates doing well, sometimes its just him, sometimes its other things. I've no doubt its virtually impossible to measure this sort of thing.

    So you can say that momentum is difficult or impossible to measure. But as long as there is within-subjects variance across games and weeks and months, momentum is no worse an explanation than any other.

    ReplyDelete
  11. You know what else is interesting about momentum? Assuming Cain cleans up tonight, the Tigers are on the verge of a historic collapse: they recently became the 3rd (?) team to win a 7-game series without ever trailing (in the LCS), and now they are maybe five hours away from becoming the 4th team to lose with such ignominy.

    In 2007, the Colorado Rockies went on a historic run by winning 21 of 22 to take the NL pennant. Then they took on the Red Sox in the WS, and like the Tigers this year, got waxed on the heels of a historic pennant/playoff run. That's twice in six years. Random chance?

    NHL players might say differently. In the NHL there's a 'superstition' that players shouldn't touch the conference championship trophy. We don't have many examples of teams violating that superstition; one that comes to mind is the '97 Capitals, who hammed it up with the conference championship trophy. Those Capitals proceeded to get waxed by the theretofore playoff underachieving Red Wings in the finals.

    Lot of examples we can think of involving teams that get blown out in a final after celebrating/being celebrated for something other than winning it all. Is there something to that NHL superstition? What does your instinct tell you?

    ReplyDelete

Note: Only a member of this blog may post a comment.