Senate Forecast, 7/18: Republican Outlook Improves

FiveThirtyEight: Politics Done Right

by Nate Silver @ 9:47 PM
Bookmark and Share Share This Content

Our latest Senate simulation has the chamber convening in 2011 with an average of 53.4 Democrats (counting Joe Lieberman and Bernie Sanders), 46.1 Republicans, and 0.5 Charlie Crists. This is an improvement for Republicans from our last forecast three weeks ago, which had 55.2 Democrats, 44.2 Republicans, and 0.6 Crists. The changes, however, predominantly reflect several methodological improvements we have made rather than any particular national momentum, although the dynamics of some individual contests are certainly evolving.

The model gives Republicans a 17 percent chance of taking over the Senate if Charlie Crist caucuses with them, up significantly from 6 percent three weeks ago. If Crist does not caucus with them, their chances of a takeover are 12 percent. However, the model does not account for the contingency that someone like Joe Lieberman or Ben Nelson could decide to switch parties, which makes their chances slightly better than we suggest here.

Democrats’ chances of gaining a net of one or more seat and re-claiming a 60-seat majority are 7 percent, down from 12 percent three weeks ago. If they could persuade Charlie Crist to caucus with them, however, their chances would improve to 10 percent.

Methodological improvements. There are four changes to the model from June’s version, which are briefly described below. The first change — the likely voter adjustment — is by far the most important.

Likely voter adjustment. We now notate whether each poll is of likely voters, registered voters, or all adults, and include variables for this in the regression analysis we use to calculate pollster house effects. The regression shows that, holding house effects constant, Democrats do a net of 4 points better in polls of registered voters (with a 95 percent confidence interval of about 2-6 points) than in polls of likely voters, and roughly 7 points better in polls of all adults. So if we took, for example, the recent Ipsos poll of California, a poll of registered voters which showed Barbara Boxer with a 4-point lead over her Republican challenger, Carly Fiornia, we would expect it to show about a tied race if a likely voter screen had been applied instead.

The adjustment works in exactly this fashion: it adds a net of 4 points to the Republican candidate’s margin each time that it encounters a poll of registered voters, and 7 points every time that it encounters a poll of adults (which is very rare in state-level polling.) Polls of likely voters are left unchanged as having a likely voter screen is assumed to be the default condition.

In 2008, I had been reluctant to be overly aggressive about using likely voter polls this far out from the election. In fact, when the pollster had presented us with a choice of a registered voter and a likely voter poll, we went with the registered voter numbers until Labor Day. The reason for my change of heart is twofold. First, by the time we get to November, essentially all polls will be of likely voters. In other words, we are essentially predicting what the polls are likely to show in due time anyway, once all pollsters have switched to a likely voter model. This should lend additional stability to the forecasts, as they will converge toward their final estimates more quickly. Secondly, there is substantial quantitative and qualitative evidence of an enthusiasm gap between Democratic and Republican voters, which we would quite naturally expect to be reflected in likely voter polls but not so in polls of registered voters. (It is probably also the case that registered voter polls generally somewhat underestimate Republicans’ performance in midterm elections even when the mood of the country is fairly neutral, but that is especially likely to be the case in a year like this one, where enthusiasm is imbalanced).

Because of the way our model is calibrated, this adjustment does make a fair amount of difference: it is most of the reason that Republicans have a somewhat better forecast now than they did three weeks ago. But note that the adjustment is self-correcting; if, as other pollsters switch over to likely voter models, they do not show an enthusiasm gap, its effect will be reduced in our model.

We are tracking to see whether a similar adjustment might be warranted for polls which exclude cellphones from their samples; at the moment, there does not appear to be sufficient statistical certainty to do so.

I’m sure that some people will also wonder what impact this adjustment has on Rasmussen’s polling, which has had a large, Republican-leaning House effect this cycle. The fact that Rasmussen has always sampled likely voters, whereas some other polls survey registered voters adults, accounts for some of its house effect. However, as we reported before, it does not account for all of it: we still show a Rasmussen house effect of 2-3 points even once this is accounted for.

Research 2000 polls removed. We no longer include any polls from Research 2000 because, in my considered professional opinion, the preponderance of evidence suggests that some or all of their data may have been falsified, or that if it was not falsified, it nevertheless reflects highly unorthodox and unscientific polling practices. Although I take no position on whether Research 2000’s polling has been proven to be falsified “beyond the shadow of a doubt”, this is not a court of law and we have no reason to apply that stringent a standard. If Research 2000 made even a rudimentary effort to provide us with the raw, interview-level data from its polls, or records of its transactions with its call centers, I would probably give them the benefit of the doubt, but so far they have not. (Note that I have not physically removed Research 2000’s polls from my database — they are still in there, but they are assigned a special flag that gives them a weight of zero. If Research 2000 were able able to make me reasonably comfortable that it in fact conducted the polls that it claims to have conducted, I would of course consider re-including them.)

Note, however, that in the case of Research 2000, this actually has a fairly minor impact, as their polls had already been assigned a low weighting because of their mediocre track record, and because they were already being substantially adjusted because of their large, Democratic-leaning house effect.

Use of partisan identification rather than PVI. Our forecasts are based, in part, on a regression model which seeks to predict the polling average in each state based on a number of variables, like the incumbent status of each candidate and their favorability or approval ratings. One of the variables in this analysis had been a state’s Partisan Voting Index (PVI), which is a measure of how it voted relative to the rest of the country in the Presidential elections of 2004 and 2008. However, growing out of some concerns with the numbers the regression model was producing in West Virginia, I decided to test an alternate measure of a state’s partisan orientation, which was the number of people who identified as being a member of each party in Gallup’s tracking throughout 2009. The partisan identification measure turned out to have substantially more predictive power than PVI (in fact, PVI adds no additional value once it is accounted for), so we will now be using party identification in PVI’s place. This helps the Democrats in states like Kentucky, which has relatively high levels of Democratic identification relative to their voting in the past two Presidential elections, but hurts them in places like Colorado, for which the opposite is true.

West Virginia now included. We now include a forecast for West Virginia, where Robert Byrd died last month and where there may now be a special election to replace him. We assign a 25 percent likelihood to each of the following scenarios: there is a special election between Joe Manchin and Shelley Moore Capito, there is a special election between Joe Manchin and Betty Ireland (these two matchups have been polled), there is a special election between a generic Democratic and a generic Republican candidate (the forecast for this scenario is based solely off of regression analysis), and there is no special election until 2012. Overall, the model figures that there is a 12 percent chance that the Republicans take over the seat in West Virginia, mostly based on the scenario where there is a special election but Manchin does not run.


As noted above, it is these methodological changes that account for most of the improvement in the Republicans’ standing in this month’s forecast. We actually show the national trendline — based on an analysis of generic ballot polling — having been exceptionally flat for at least the past 4-5 months, with Republicans having gained only about half a percentage point during that period. (This contrasts with Barack Obama’s approval ratings, which have shown some additional signs of erosion.)

Still, this is by all means looking like a very bad cycle for Democrats, and they are running out of time to right the ship. Republicans clearly have enough states in play to become the majority party. On the other hand, they are playing defense in a number of places, and some pickups that looked fairly certain before (Nevada, Delaware) look less so now. Therefore, they will probably need some additional national momentum of their own to have a clear shot of taking over the Senate.

Detailed results follow below. (Note: our earlier post this evening had incorrectly omitted a PPP poll of Wisconsin. We put the Republicans’ chances of taking over the state at 26 percent, not 24 percent. The other numbers in this article have been corrected accordingly.)

…see also 2010, likely voters, methodology, party identification, rasmussen, research 2000, senate, senate polls, senate rankings, west virginia


CDS said…

July 18, 2010 9:57 PM
Alex Power said…

I realize that this reflects the polls, but I’m not quite sure I believe Dan Coats is a 95% favorite in Indiana (both in that it is better odds than Ron Wyden in Oregon, and since a lot can change in 3 months). If there’s any places I can bet on Ellsworth at these odds, I would like to wager a large amount of money.
July 18, 2010 10:01 PM
Matt said…


I agree, and it’s a point that can be made for many of the races, particularly Nevada, Pennsylvania, and Colorado. But Nate’s method is a quantitative one that looks at polling. Realistically, Nate’s method probably isn’t going to show us extremely useful results until the September version (where it will start being extremely valuable, in my view).

One additional concern I have is this. The likely voter adjustment is probably a good idea, but I don’t see any point to adding it before September. There’s no evidence to suggest that likely voter models have any value before Labor Day. If Nate is adding it now to try to work out kinks, then I think that’s a positive. But it surely doesn’t have any real predictive value at this point.
July 18, 2010 10:12 PM
Zane said…

If the Democrats lose any of Wisconsin, California, or Washington, I will eat my hat*.

*I don’t actually wear a hat, but this should still be considered a strong statement on my part.
July 18, 2010 10:19 PM
Bart DePalma said…


Which LV polling are you using to derive the GOP +4 modifier?

Is the non-Rasmussen LV polling you are using routinely 2-3 points more favorable to the Dems than is Rasmussen?

Thanks in advance.
July 18, 2010 10:23 PM
shrinkers said…

Dems losing 4 or 5 seats is probably the upper limit. I’m still thinking more like 2 or 3.

Remember that no one predicted the Dems would have 60 seats at any time this current session (and in truth, they only had 60 for about 6 months). Even the 58 or 59 was seen as only an outside chance. Of course, 2008 was a strange year. But then, we are still in strange times.

Depending on what Crist does, we’re at almost exactly even chances of the Dems getting 60 or the R’s getting a majority. That feels pretty good.

The Dems are doing pretty good this legislative session with less than 60 — and the 60 they had for a short time was just a numerical oddity anyway, since so many of them are conservative, and voted with the party only after getting huge concessions, if at all.

I expect the filibuster rules to change next session, since the Republicans have so blatantly been abusing the procedure. So a majority of even 53 or 54 should certainly be enough to continue the work we sent them to Washington to do — especially since the Dems are going to maintain a majority in the House.

Further, since the Republicans are setting expectations so very high (all that silly tsunami talk), anything less than running the table will seem like a total failure for them. Of course, after the election, they will claim victories no matter what, and will turn up the crazy juice spigot. This will put the Dems into very good shape for 2012, since the Teaper rhetoric is so offensive to such a large segment of the American voting public.

And as Matt and Alex have already hinted, the real campaigning still hasn’t started. A few people beyond us geeks are only now starting to pay attention. I expect to see many of the numbers change starting September.

I also want to learn how the pollsters are defining “likely voter.”
July 18, 2010 10:31 PM
Juris said…

Yeah, Shrinkers. The GOP seems bent on threatening the Democrats with a horrible Japanese-made disaster, right here on American soil.

Red-coated Americans should only threaten their opponents with American-made catastophes. I mean, c’mon, Gopers, tidal waves are as American as apple pie. Why give the Japanese credit for them by calling them tsunamis? That is totally and utterly unAmerican. Just one more sellout by the Red party.
July 18, 2010 10:57 PM
Ebscer said…

I think the republicans have a real chance of taking the house. The senate seems less likely (although they will improve their standing some).
July 18, 2010 11:00 PM
Congress said…


No way filibuster rules are changed; wishful thinking on your part. Rules changes like this require a 2/3 majority… so no minority party will ever sign on to it… and nor should they. We the people would like SOME checks and balances on all of these crooks, D’s and R’s …
July 18, 2010 11:20 PM
Chaboard said…

Balancing this (essentially steady state) Senate new with the positive House fundraising picture painted here…:

….I believe on balance it’s a good news day for us Dems.

Actually, while the money news in the House was reasonably good I think the part that surprised me was that Charlie Cook currently only has 34 Dem-held seats identified as “tossup or leaning in the direction of the GOP” – and 28 of those 34 are tossup. If they split down the middle then we’re looking at only a 20-seat pickup (assuming ZERO GOP turnover). Even with a moderate wave you’d expect only maybe 75% of those to go and you’re still looking high twenties. (And 75% seems high. They didn’t even get that in ’94 when there were a boatload of Dem retirements AND beneficial redustricting, did they?)

True, if a “tsunami” materializes they’d probably all be gone as well as a few more on top. But seems at this point even Cook thinks we’re a LONG, LONG way from anything like a tsunami.
July 18, 2010 11:21 PM
birdboy2000 said…

Is a senator’s age accounted for in the model? It seems a decent proxy for “likelihood of death/retirement” (though I haven’t run any statistical analysis) and such things can throw seats from “beloved incumbent” to “open” rather easily.
July 18, 2010 11:22 PM
Congress said…


Have you done any analysis on some of the internal numbers from these polls?

I have seen two consecutive polls now from Ras on the Wisconsin race where the headline doesn’t match the internals, unless the respondents are heavily weighted towards the Dems (and yes, I know it’s a so-called “blue” state, but it ain’t THAT blue!).

If you look at the gender cross-sections, and the Independent voter cross-sections, there’s no way Feingold is within a point; he should be 4-5 behind overall.

Other polls by other pollsters in other races are showing similar issues where the internals just don’t give the overall result.

Any thoughts on this?
July 18, 2010 11:24 PM
Will said…

Having trouble believing some of these. Dino Rossi has a 47% chance of winning in Washington? His only credibility comes from two failed governors’ races.
July 18, 2010 11:36 PM
Steve Rogers said…

“No way filibuster rules are changed; wishful thinking on your part. Rules changes like this require a 2/3 majority… so no minority party will ever sign on to it…”

Theoretically, a simple majority could set a new precedent and amend the senate rules if the so called “nuclear option” were invoked.
July 18, 2010 11:57 PM
Wolf of Aquarius said…

After all the talk of a “tsunami” election, it would be immensely funny if the Democrats actually regained their filibuster-proof majority. Alas, it’s probably too good to be true (7% chance, sigh). But with ACORN gone, the GOP would have to find somebody else to accuse of stealing the election … the New Black Panthers, perhaps? (Actually, PPP found that 52% of Republicans thought that Obama had used ACORN to steal the 2008 election).
July 18, 2010 11:58 PM
Matunos said…

I think the predictions based on Crist caucusing with the Dems, or Lieberman or Nelson caucusing with the Repubs, are backwards.

The effort the Democratic party is going to invest to entice Crist or keep the other two is going to be based on how they do in the election. For example, if the Dems do particularly well (a vote away from 60), or if they do particularly poorly (a vote short of 50), and Crist could put them over the edge, they’re going to go out of their way to nab him. If Crist caucusing blue doesn’t help as much, then they’re likely not to go out on a limb for him, content to deal with him more as a Scott Brown.

I think, therefore, the right question is: what are the probabilities where Crist (if he wins) comes into play for the Dems, or where keeping Lieberman and Nelson becomes an imperative.
July 19, 2010 12:22 AM
Jim said…

@Nate: This looks slightly out of date already.

(a) In Louisiana, Vitter has recently drawn both primary opposition (so is he still 100% to be the nominee?) and a conservative 3rd party challenger.

(b) In West Virginia, comments have suggested that Capito will not run against Manchin, but might well run *otherwise*, so you should probably have a category for Manchin sits it out, and the Republican is *not* generic. (In that case, the Dem is likely to be a state legislator or equivalent.) Also, is the “no election” possibility still live?
July 19, 2010 12:29 AM
Devin Mitchell said…

Everyone in the DSCC should be fired if the Democrats lose the Senate.

They already have 40 seats from senators who aren’t up for re-election, and they can get to the 50 they need for a majority (w/Joe Biden’s help) with 10 Democrats that are very safe in November in Blumenthal, Boxer, Feingold, Gilibrand, Inouye, Leahy, Manchin, Schumer, and Wyden.

They could lose everything else and still be in control of the chamber unless Nelson and Lieberman caucus with the GOP, which I don’t think they’ll do.
July 19, 2010 12:30 AM
Wolf of Aquarius said…

Everyone in the DSCC should have already been fired after the Coakley fiasco. Talk about asleep at the wheel. And the DSCC (and DNC) should be REALLY careful about resource allocation. No money for Dems like Nelson who keep filibustering the party’s platform. And no money for Meek. He’s just a spoiler at this point.
July 19, 2010 12:39 AM
shiloh said…

(If) Crist wins, besides the obvious ie the Rep party screwed the pooch in FL re: a totally safe Rep seat …

It will also mean an Ind moderate did really well w/Ind and Dem/Rep crossover voters which bodes well for Dems overall in the mid-terms, as like Ohio, consider FL a bellwether state also.

It’s ironic that extremist Rubio may have put Crist in the ideal situation to be elected as an independent ie Rubio is a fish out of water empty suit.

Indeed Rubio has kept a very low profile since flip/flopping on AZ’s new racist immigration law as he originally came out against it and quickly changed his mind w/public opinion.

He’s totally confused lol


and isn’t it fun to speculate ~ eh Bart …

again Bart, pace yourself ~ TIA 😉
July 19, 2010 12:42 AM
Alex said…

Anyone have any ideas about how I can set up a pool to use to bet with my friends about what seat switch?
July 19, 2010 12:48 AM
Devin Mitchell said…

Isn’t Kendrick Meek basically the Democratic of Alan Schlesinger in the ’06 CT Senate election?
July 19, 2010 12:57 AM
brian said…

I’ve heard that Crist is doing well because of the oil spill. I suspect he’ll do well if he can focus on “good government” issues rather than raw political ones.

We’ll see what happens once national issues become a campaign focus. I am curious on him having to declare a Caucus. When does that happen?
July 19, 2010 1:03 AM
Dewey said…

Interesting analysis from Charlie Cook, regarding the House. If Democrats keep their House losses at around 20-25 seats, all that does is roll them back to the majority they had during the last two years of Bush’s term – and this time, they have a Democratic president to work with. Their agenda probably wouldn’t be as ambitious, but it’s certainly workable. On the flip side, they could still hold a decent-sized majority in the Senate even if they lost five or six seats, but unless they get rid of the filibuster, government will be paralyzed for two years. We’ll see.
July 19, 2010 1:09 AM
Dwight said…

brian said…

We’ll see what happens once national issues become a campaign focus.

Unlikely he’ll let that happen. Although he’d do fine through most of the state he probably doesn’t want to push it too hard in South Alabama (AKA FL pan handle).

brian said…

I am curious on him having to declare a Caucus. When does that happen?

Whenever he feels like it. That’ll probably be negotiations that occur in at the start of the 112th Congress in Dec ’10/Jan ’11, like what went on with Lieberman (and will likely go on with him again for the 112th). It pretty much has to happen as part of divvying up the Committee duties as that’ll be core to the negotiations, and that can’t in earnest happen till you know who all the Senator-elects are.
July 19, 2010 2:23 AM
sophia said…

The GOP has actually worsened in recent decades.Again it does not take much reasoning power to understand the long-term benefits of restructuring the energy significantly outweigh the costs in the short term.
dominos vouchers
July 19, 2010 3:20 AM
Daniel Trubman said…

I get the feeling that the last few polls of the Reid vs. Angle race haven’t been included. Is that true?
July 19, 2010 3:22 AM
belegoster said…

Dewey and Chaboard,

District-by district ratings focus on the dynamics of each race individually. It looks about as rosy with the prognoses by other teams:

CQ politics: 31 Dem seats tossup/lean GOP
Larry Sabato: 39 Dem seats tossup/lean GOP

However, this is liable to miss the boat completely in case this turns out to be a wave election (which is more likely than not). The picture dramatically changes if you predict based on national factors (pres. approval, generic ballot, off-year/party in power dynamics). The congressional ballot should look particularly scary for Dems: historically, it overstates the Dem share of the actual vote by about 6% on average. Furthermore, if you do a simple regression of Dem share of actual 2-party vote against no. of Dem seats from 1994-2008 you find that they have to do better than 50% to retain a majority (Rsquared=0.92). So the die is twice loaded against them: they have to maintain a lead on the generic ballot just to stay even, and they have to do better than even to keep the House.

Thus, even though Cook rates ‘only’ 34 Dem seats as tossups/lean GOP he has been hyperventilating about ‘Hurricane GOP’ for a good year now and thinks 30 GOP pickups is the floor. He is looking at these national indicators.

Personally, as a Democrat, I would actually prefer that the GOP take the House, but maybe by just 1 or 2 seats. That way we 1) give the Tea Partiers a victory to deflate their energy, but not so much as to embolden them, 2) shift the burden of governance to the GOP, but not so much that their leadership feels comfortable in letting Darrell Issa loose. I would rather be a 217-seat minority party exposing Republican hypocrisy than a 220-seat majority party blamed for incompetent governance.
July 19, 2010 3:55 AM
Dewey said…

Every election is local. If there truly was a national referendum against Obama and the Democrats, they wouldn’t have held NY-20 or PA-12 – they wouldn’t have won NY-23. The Tea Party is all hype and I don’t think will transfer over to the general election. However, we’ll have a clearer picture after Labor Day.
July 19, 2010 4:11 AM
Joe Neri said…

Which party has raised the most money?

Which party has done better in recent special elections?

Which party had more primary voters so far?

You really think any of that will change come November?
July 19, 2010 4:37 AM
shrinkers said…

No way filibuster rules are changed; wishful thinking on your part. Rules changes like this require a 2/3 majority…

No, they don’t. At the beginning of each session of the Senate (that is, in January), the Senate takes a simple majority vote to set the rules for that session. The filibuster rules are some of the rules included in that package. These rules must be renewed every session, and can be changed at that time — and as I said, that is by a simple majority vote.
July 19, 2010 7:22 AM
Chaboard said…

“No way filibuster rules are changed; wishful thinking on your part. Rules changes like this require a 2/3 majority… so no minority party will ever sign on to it…”

Theoretically, a simple majority could set a new precedent and amend the senate rules if the so called “nuclear option” were invoked.”


That’s true but even that is overstating the case. If it’s done at the beginning of the Congress it only takes a simple majority WITHOUT the “nuclear option”.

I have serious doubts as to whether Rid has the balls to do it or whether they could round up the 50 votes they’d need (with Biden’s tiebreaker)….but that is all it takes if they do it at the start.

Me, I’d like to see them reduce the number from 60 to 55….but more importantly I’d like them to restore the filibuster of old. Change the rues to require the filibustering Senator to hold the floor talking as they had to before Byrd’s 70’s reform. You force the filibustering party to feel the pain and you’ll naturally limit the use (abuse) of it.
July 19, 2010 7:30 AM
Chaboard said…

However, this is liable to miss the boat completely in case this turns out to be a wave election (which is more likely than not). The picture dramatically changes if you predict based on national factors (pres. approval, generic ballot, off-year/party in power dynamics)

Yeah, I understand that and alluded to it in my post. But it’s not a scenario I worry about much for a couple of reasons.

One, waves build momentum and crest. As Nate pointed out in the Senate forecast we’re not seeing much movement at all….things are saying flat. I think that points to the bottoming out for the Dems having already occurred.

Two, I don’t think the GOP will be able to successfully nationalize the elections. Look at the PA special held earlier this year where they tried hard and failed. Their problem is that nationally they are despised and hated more than the DEMS so if the DO nationalize the races…they lose.

Personally, as a Democrat, I would actually prefer that the GOP take the House, but maybe by just 1 or 2 seats. That way we….[snip]

Allow me to play fill in the blank and finish your statement for you….”that way we get to endure the House Republicans issuing endless subpoenas to investigate the legality of Obama’s setting his alarm clock setting”.

If they win the House we shift from mostly gridlocked to “absolutely gridlocked”. And with all the raging crisis’ in this country right now total gridlock – even for two short years – could be the coup de grace for our nation. Not something we can risk.
July 19, 2010 7:43 AM
shrinkers said…

If they win the House we shift from mostly gridlocked to “absolutely gridlocked”. And with all the raging crisis’ in this country right now total gridlock – even for two short years – could be the coup de grace for our nation.

Right here, you put your finger on the difference between a political calculation and a policy consideration.

The Republicans seem to only be capable of making political calculation — what soundbites and campaign slogans might play well. This they do without regard to the damage to the country if a) the suggested policies are actual enacted, or b) the actual remedies to our problems are not enacted. Republican publicity is mostly sloganeering, and attempts to distract from actual issues rather than to confront them.

I forget who it was who said this — Republicans don’t want to govern. They want to rule.

What they are concerned about is being in charge, so they can continue redistributing money from the middle class to the wealthy. What they are not concerned about are issues of policy — that is, what is actually good for the nation?

I hope the Democrats never fall into that same error.
July 19, 2010 7:53 AM

Post a Comment
Older Post Home
Subscribe to: Post Comments (Atom)

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to Top