FiveThirtyEight’s updated Senate forecast is out, taking into account the results of various primaries — most significantly Nevada, where the victory of Tea Party favorite Sharron Angle has turned a likely GOP pickup into a tossup (and where “Harry Reid’s favorability/approval numbers have also improved some, and are now merely godawful rather than mind-bogglingly wretched”), and Arkansas, where choosing incumbent Blanche Lincoln was actually a bad choice, electability-wise, for the Democrats — as well as the odd developments in Florida, with Charlie Crist becoming the Republicans’ Lieberman (and becoming, according to the polls, a pretty solid favorite to win).
Bottom line: there are now four Senate seats that are extremely likely to flip from blue to red: North Dakota (100% chance), Arkansas (99.7%), Indiana (88%) and Delaware (80%). All the other battlegrounds are pretty much tossups. So, conceding those four states to the Republicans, we’re looking at a baseline of Dems 55, GOP 45 for the 112th United States Senate, with 5 presently Dem-held seats and 6 GOP-held seats clearly in play (and a few others marginally so). More details after the jump.
The presently GOP-held “tossup” and “lean” states — i.e., the potential Dem pickup opportunities — are Florida (where there’s presently a 66% chance of either Crist or Meek winning), Ohio (57% Dem chance), Missouri (42%), New Hampshire (35%), North Carolina (30%) and Kentucky (29%). The latter two are significantly more Dem-friendly because of primary results. The Dems also have marginal chances in Arizona (16%) and Louisiana (9%), though it’s hard to believe the national environment is going to somehow swing in their direction such that those would become true battlegrounds.
The presently Dem-held “tossup” and “lean” states — i.e., the potential GOP pickup opportunities (above and beyond the four near-certain pickups mentioned earlier) — are Pennsylvania (53%), Nevada (52%), Illinois (52%), Colorado (50%) and Washington (26%). If this turns into a “wave” election like 1994, California (18%) and Wisconsin (11%) could be in play, too. But the trend isn’t moving toward such a “wave”:
[W]e show essentially zero change in the national environment over the past several months, and only a net gain of one or two points for Republicans since the start of 2010. In contrast, Democrats lost about 12 points on the generic ballot over the course of 2009. They are not really climbing out of the hole the dug themselves, but on the other hand, it does not appear to be getting worse.
Of course, if the current rumblings of economic PANIC!!! turn into a distinct plunge toward double-dip territory by November, we’d be looking at a totally different map. But for now, things seem pretty stable nationally, and local conditions (i.e., primary results) have improved things markedly for the Dems:
Democratic fortunes were improved by the primaries in Nevada and Pennsylvania, California, North Carolina, and Kentucky, and worsened probably only in Arkansas (and South Carolina, which they had almost no chance of winning anyway.) This accounts for most of the movement in the rankings. Whereas, as of our last update, our simulations were projecting an average of 54.0 Democratic and 46.0 Republican seats, we now show 55.2 Democrats, 44.2 Republicans, and 0.6 Charlie Crists.
Heh. I like the concept of 0.6 Charlie Crists. It’s like he’s Schrödinger’s Senator. 🙂
Anyway, the GOP’s odds of reaching 50 or 51 seem pretty slim. In addition to North Dakota, Arkansas, Indiana and Delaware, they’d need to sweep the Dem-held 50-50 tossup states (PA, NV, IL, CO), and hold onto both of their own 50-50 states (OH, MO), and either pull an upset in Florida or convince Crist to caucus with them — or convince Lieberman or Nelson to switch parties — just to reach 49. Hitting 50 would require a mild upset in Washington, or an additional seat from the Lieberman/Nelson/Crist trifecta. A major upset, most likely in California or Wisconsin, would probably be necessary to reach a majority.
Of course, there is once scenario in which this is all quite a bit more plausible:
[The Republicans] can hope that Rasmussen is right and everybody else is wrong. Right now, we are identifying about a 4.5-point Republican-leaning house effect in Rasmussen’s polls, relative to a robust average, weighted based on pollster quality. But suppose for a minute that Rasmussen is right and everyone else is wrong, and we calibrate to Rasmussen’s average instead. In that case, the projected number of Republican seats goes from 44.2 to 47.4, and their chance of taking over the Senate more than triples to 22 percent, counting Charlie Crist as a Republican.
That’s as opposed to a 6 percent chance under 538’s model, if Charlie Crist is counted as a Republican, or 4 percent chance if he isn’t. (Silver adds: “If Republicans could hypothetically persuade both Crist and Joe Lieberman to join them, …their chances would improve to 12 percent.”)
I trust Rasmussen’s numbers, and not necessarily because they are favorable to the GOP. I trust their methodology much better on who the likely voters are going to be, as it tracks well both intuitively to me, as well as anecdotal and empirical evidence I have run across — not to mention the historical track record of 1994 and other non-presidential general elections.
I think Silver’s skeptical (though, unlike some of his readers, not completely dismissive) take on Rasmussen is pretty compelling. See, for instance:
http://www.fivethirtyeight.com/2010/05/new-poll-of-500-political-junkies.html
http://www.fivethirtyeight.com/2010/05/wisconsin-senate-feingold-46-smithee-r.html
http://www.fivethirtyeight.com/2010/04/use-of-likely-voter-model-does-not.html
http://www.fivethirtyeight.com/2010/03/house-effects-render-poll-reading.html
Although, Rasmussen is at least more trustworthy than Research 2000! But I digress.
Regardless of which pollster one trusts, I think it’s sensible to look at the Senate midterms as starting from a 55-45 baseline. It would be utterly shocking if the Dems won in any of the four states that I’m conceding to the GOP: North Dakota because of its demographic/geographic/ideological fundamentals, the retirement of the only Democrat who could ever win there, and the bad national environment; Arkansas because of its fundamentals, the unpopularity of the incumbent, and the bad national environment; Indiana for essentially the same reasons as North Dakota but less strongly; and Delaware because of the popularity of the GOP candidate and, again, the bad national environment for Dems. So the GOP has a four-seat head start.
After that, who knows? If the economy deteriorates, I’d say the smart money is on the GOP to win five out of six in PA, NV, IL, CO, OH and FL (treating a Crist win in FL as a GOP loss). That would put them at 48 seats, +3 from the 55-45 baseline (since they’d be going from 2-4 in those states to 5-1). A sweep of those six gets them to 49, assuming they hold MO, NH, NC and KY, which I assume they would if the economy deteriorates. But I’m really skeptical that they can win WA, CA or WI, two of which they’d need — barring shenanigans with Lieberman or Crist or Nelson — to get a majority. California Repubs missed an opportunity and messed up the national calculus, IMHO, when they decided not to nominate the electable, moderate and impressive Tom Campbell, and went with Demon Sheep instead. Likewise, the candidate recruitment failure in Wisconsin is a major bummer for the national GOP. Even in a wave election, those states may be just barely out of reach. I suspect the most plausible path to 51 for the GOP is to sweep the six tossups, win Washington, and woo Senator Joementum to the dark side at last.
On the other hand, if the economy stays relatively stable and Obama avoids doing anything colossally stupid between now and November, I’d expect the GOP to end up with something like 46 or 47 seats.
Caveat: I’m also assuming such characters as Sharron Angle and Rand Paul “avoid doing anything colossally stupid between now and November.” This is not necessarily a safe assumption, and that 46-47 could potentially become 44-45 on the basis of local candidate implosions.
You mean the Research 2000 that was behind the Daily Kos’ infamous Republicans-are-insane poll?
California Repubs missed an opportunity and messed up the national calculus, IMHO, when they decided not to nominate the electable, moderate and impressive Tom Campbell
Campbell is seen as the candidate of the party machine and “business as usual”. And out here, “moderate” means “a Republican who votes with the Democrats”.
Brendan, if Tom Campbell was electable and impressive, it stands to reason that there would be more Tom Campbell-like politicians out there. There are not. You’re confusing “moderate”, “reasonable”, and “attractive to Brendan Loy” with “electable”.
As for 538, despite the liberal echo chamber that site has become, I highly respect their poll analysis. That being said, I think both 538 and you are missing the big picture here with Rasmussen, which is that Rasmussen has been incredibly successful in predicting elections, and that Rasmussen has been closest to accurately predicting what the anti-incumbent zeitgeist foretells for this year’s electoral results based on recent results (most prominently, Brown vs. Coakley, which 538 conveniently ignores when rehashing their analysis of the various polls’ “house effect”).
I’ve been following politics and polls since middle school, and it’s always fascinated me how media and the polls have trouble pinning down conservative / Republican voters. There have been excellent theories here and there — the “shy Tory” / “silent majority” is a good one — as well as recognition that white Republicans vote far more consistently and thus their electoral strength carries more weight in primaries and non-presidential general elections. But none of these theories have sufficiently explained results such as what happened in 1994. It seems to me Rasmussen is reading the tea leaves and is artfully tweaking his process to account for this “1994” effect, which explains his strong pro-Republican “house effect”, and thus far, he has been fairly dead on.
Rasmussen has been incredibly successful in predicting elections
Five Thirty Eight’s recent pollster ratings — the most comprehensive example of their poll analysis, which you say you highly respect — showed Rasmussen, based on an objective, across-the-board analysis of all their polling (as opposed to subjective, selective recollection of individual polling success), to be only slightly above average as a pollster, with decreasing reliability in recent years (coincidentally as the Dems have done better at the polls in ’06 and ’08).
But to Silver’s credit, I think, he’s acknowledging the possibility that what you’re saying is essentially right, that Rasmussen could indeed be “artfully tweaking his process to account for [a] ‘1994’ effect” that will in fact materialize. The problem, though, is that there’s no objective data to support the hunch that it will materialize (sorry, but one special election in Massachusetts does not a data set make) — it’s just a theory, an educated guess, a hunch. It may be right, but there’s no particular objective statistical reason to believe it’s right. If you share Rasmussen’s preconceptions, you’ll think it’s right. If you don’t, you won’t. So when you say “thus far, he has been fairly dead on,” I don’t really know what you mean. We’ll find out in November, but not until then, who’s “dead on” and who’s dead wrong.
if Tom Campbell was electable and impressive, it stands to reason that there would be more Tom Campbell-like politicians out there
Not necessarily. I can think of at least one hypothesis that would explain the dearth of Tom Campbells without positing that they’re unelectable and unimpressive. It would go like this:
Hypothetically, suppose we had a two-party system whose nominating processes had a natural tendency to favor nominees who appeal to the “base” rather than the “center.”
Further suppose, hypothetically, that we had a political culture in which centrist voters are generally disinterested in engaging in party politics (in part because of the feedback loop creating by their own disengagement — i.e., they aren’t involved, so the parties are non-centrist, so they’re turned off by the parties, so they don’t get involved), leading to a huge mass of self-described unaffiliated or independent voters with essentially no say in choosing the two major-party nominees.
Also suppose, again hypothetically, that our two-party system were structured in such a way that there are enormous barriers to entry which make it very hard for centrist voters and centrist candidates to form successful minor parties that could effectively challenge the above-described hypothetical status quote.
Finally, suppose hypothetically that one of the major political parties recently suffered a tremendous decline in its image, or “brand,” among centrist voters, such that even MORE centrist folks joined the unaffiliated/independent/uninvolved group, leaving this hypothetical party even MORE dominated by its ideological “base” and even LESS influenced by the center.
If Tom Campbell were a member of that party, then it would stand to reason that he, and folks like him, would — hypothetically — have a tough time getting anywhere in politics, despite the fact that they ARE impressive and, in theory, electable… if only they could find themselves a workable platform from which to showcase that impressiveness and electability to the (largely centrist) masses.
The only remaining question is whether this hypothesis is true. I maintain that it is.
Tom Campbell doesn’t have enough of an established relationship with the voters to pull a Lieberman or a Crist, basically giving the two-party system the finger and going straight to the general electorate. It’s rare to find a politician who does. But that doesn’t mean, if we redesigned our electoral system to more accurately reflect the general electorate’s desires, we wouldn’t have more Liebermans and Crists and Campbells. It just so happens that our present system creates incredibly high barriers to entry for such people.
Rasmussen had the same problem in 2006 and 2008 that most other pollsters had in accounting for the rise of young and minority voters. For instance, most polls had Obama winning in 2008, but nobody really predicted that his margin would be so comfortable. Since Rasmussen appears to disproportionately rely on the “shy Tory” vote in his likely voter models, he was especially thrown off when those voters largely revolted in 2006 and 2008 by staying home or voting for other candidates.
As for the second part of your argument, I fully concede 538’s critique of Rasmussen’s polling methodologies and I agree that objective data is largely missing to support Rasmussen’s lean. But what I am trying to say above all else is that the point of polls is not to be objectively bulletproof in their methodologies, their goal is to be proven correct. Folks like Silver and yourself tend to be disproportionately emphasize the former as a way to achieve the latter, but when it comes to human behavior, it simply isn’t possible to come up with a bulletproof methodology that works for every election and every demographic. At some point, you have to bake into your methodology assumptions and presuppositions about how voting behavior this election will mirror or not mirror past examples, and Rasmussen is no more guilty of this than any other pollster — he just acknowledges the art of his craft more readily than folks like Silver (who prefer to drill into the statistical and scientific aspects) prefer to accept. But again, Rasmussen’s track record suggest his polls ought to be taken very seriously despite the fact that their numbers are relatively outside the normal range of other polls as tracked by 538.
Brendan, your analysis at #8 is cute, but only serves to reinforce my point that “You’re confusing ‘moderate’, ‘reasonable’, and ‘attractive to Brendan Loy’ with ‘electable’.” The system we have requires candidates to both appeal to the base and be able to appeal to enough of the center and other side to successfully be elected and re-elected. Tom Campbell clearly has failed this test and is objectively not “electable”.
I should add, gripe about Rasmussen’s methodologies if you must, but at least Rasmussen isn’t being sued for fraud by its customers and benefactors.
AMLTrojan – but … but … but … isn’t Reason2K’s results fake but accurate ?
(grin)
Brendan, hypothetically would you give us a few specific hypothetical Examples that illustrate the hypothetical Enormousness of these Veryhard Incrediblyhigh structural barriers in the present Electoral System.
( 2 items suggested for hypothetical Omission from your list because neither is a structural Design component of our Electoral System: [a] the Commerically-spawned ideological polarization of the Media & Internet politicoculture; and [b] the temperamentally-rooted Unmotivatedness of the Centrist-&-Moderate-minded segment of the citizenry to get Out there and Utilize the existing mechanisms for 3rd-party/new-party/independent politics, leaving that fertile field largely to plowing by Fanatics the intensity of whose Fervor is unrequited even by today’s Democrats & Republicans. / Hypothetically speaking, of course. 🙂
(On the Other hand, Gerrymandering, being legitimately Structural, IS List-eligible. / Provided you include a few sample Maps of districts hypothetically re-drawn in such a way as to give the new Reasonable Centrist Moderate Party [RCMP ;] a Fair opportunity to have Its
horsecandidate win the race. 🙂 (Hm? Oh, you’re most welcome. Far be it from Mee to unduly Restrict the permissible Answers. 😉#13 above, Blockquote corrected to properly reflect my Tendentious boldface Emphases ;} —
As our president would say, let me be clear. I’m not talking about barriers to entry for third-party ballot access. I’m talking about barriers to entry for a third party to permanently join the political milieu in a serious way, with a real chance of meaningful electoral success. Off the top of my head, three of the biggest barriers are:
1) The Electoral College, which basically depends on there being exactly two major parties — if there were three major parties, we’d constantly have elections going to Congress. But we never get to that point, because the E.C. has the effect of rendering most third-party candidacies somewhere between useless and counterproductive (see: Nader, Ralph). Unless and until a third party can establish itself as a major player, to the point where they can threaten to replace one of the two major parties, they can’t expect to get anywhere in presidential elections. And good luck “establishing yourself as a major player” as a party without the ability to mount a viable presidential ticket. It’s a chicken-and-egg problem.
2) “First past the post” voting, as they call it in Jolly Old England, which has many of the same effects, turning votes for third parties into “wasted votes” and forcing us to think about elections in a binary way (two parties, not three or more). The main difference between us and England in this regard is the strength of regional parties over there; that’s what gives them more of a multi-party system than we have. But, as shown by the mathematical near-impossibility of the LibDems ever getting a majority even if they were to win a huge plurality landslide in the popular vote, they really DON’T have a multi-party system, functionally. It’s Labour vs. Conservatives vs. Potential Coalition Partners For Labour And The Conservatives In The Rare Event Of A Hung Parliament. We could have this too, if ever a third party would gain traction at a regional level (which is probably not something we should hope for, but that’s a separate question). But it’s not the same thing as a true national three+ party system. The LibDems’ only hope of real power would be to replace Labour, not to become a permanent national third-party force. The same is true of third parties here, although none of ours are anywhere near as strong as the LibDems. Bottom line, there is really no such thing as a permanent national multi-party system if you have first-past-the-post voting. Such system naturally gravitate toward a two-party equilibrium. And it’s incredibly hard, nowadays, for a third party to replace one of the existing two parties, in part for reasons you allude to, and in part because of…
3) The enormous cost of competing in major elections. Again, a chicken and egg problem. Good luck raising any sort of funds if you can’t first demonstrate the ability to have serious & sustained electoral success, and good luck having any sort of serious & sustained electoral success if you can’t raise funds. This serves to reinforce the other strucutral issues. Unless you’re independently wealthy (see: Perot, Ross) or have a tremendous personal “brand” (see: Lieberman, Joe), you ain’t got a prayer. Money, honey, money is the only thing.
Now, let me be clear again: my comment was never intended as an angry rant against the system. I’m not necessarily saying an alternative system (e.g., proportional representation of come kind) would be BETTER. I’m just saying that, factually, one of the core traits of our system is that it can really only accommodate two major parties, and it is enormously difficult for a third party to displace one of the two existing major parties. And one of the core consequences of this fact is that truly centrist politicians don’t really have a viable “home,” and thus have less potential for electoral success than in a hypothetical political system that rewards overall popular support without regard for the ability to appeal to either a significantly left-of-center or a significantly right-of-center “base.”
The historical fact that the Republicans and Democrats have held a monopoly on national major-party status since, basically, the advent of mass-participation elections, despite numerous periods of strong popular dissatisfication with those parties and with the state of American politics in general, is, IMHO, incredibly powerful evidence of the structural barriers to disrupting the two-party status quo. Likewise, the inability of any minor parties, even ones that have tremendous success in a one-off election, to maintain and build on that success, is also strong evidence of structural two-party dominance. This is certainly true nationally, with successful third parties basically limited to single-person endeavours (Teddy Roosevelt, George Wallace, Ross Perot) and quickly dying when those people leave the scene; and it’s even true in state elections (…or is A Connecticut Party still thriving, with major-party status and the #1 ballot spot? Heh. Say hello to Governor Groark for me).
I’d say more, but I gotta get back to work. 🙂
Excellent list & Exposition thereof. As expected & at the expense of salaried Work, you have Risen to my challenge. ;}
So then, the relevent respective Reforms would (hypothetically 🙂 be:
1) Ditch the EC for the NPV. (But via proper Const. Amendment, not via the horrid Interstate Compact. 😉
2) Institute the Single Transferable Vote (STV) system. (“Ireland! Ireland! Together, standing tall…” 🙂
3) Adopt Public Funding of Campaigns, a la the Connecticut Citizens’ Election Fund. (But with even-better provision$ for minor partie$ than our$, about which our minor partie$ are $till $uing. 😉
“…with successful third parties basically limited to single-person endeavours (Teddy Roosevelt, George Wallace, Ross Perot)…”
You left out Ol’ Strom. :> (But even SC Repubs decided to omit his Son from their recent primary Winners List so that’s OK. 🙂
(…or is A Connecticut Party still thriving, with major-party status and the #1 ballot spot? Heh. Say hello to Governor Groark for me).
Actually Governor Weicker has been the veritable Comeback Kid around here lately. Yes, Yer Man has abruptly Re-erupted onto the local scene, Woof-woofing all over the local Newsies about how the Gutlessness of the D’s & R’s has put our budget right down the old fiscal Crapper. Indeed Mr. State Income Tax, Himself, has declared that the coming draconian Cuts will be ‘a very cold shower for a very drunken state.’ Party like it’s 1990. :>
Brendan #15 – could it be that rather than being “incredibly powerful evidence of the structural barriers to disrupting the two-party status quo”, it is actually a sign of the strength of the mostly-two-party system ?
In the UK, the majority of the consituencies had 2 realistic candidates – comparatively few were 3-way contests … (*snort*) And the LibDems will hae a chance of forming HM’s Government just shortly after the LibDems have realistic platform planks … (Nader has more chance of being US President) …
“And the LibDems will hae a chance of forming HM’s Government…”
And here I always thought it’s the SNP wha hae the Buchan Claik.
(Braid grin… 🙂
(grin) I’ve given up on correcting that particular typo, since it *is* Guid Braid Scots, after a’ …
Pingback: Kleybanova reaches 2nd round at Budapest GPcheap coach | cheap coach