ŷ������

In this live taping of the in Washington, D.C., Nate and Galen break down the current for the Senate, House and gubernatorial races.

Then Nathaniel Rakich and Amelia Thomson-DeVeaux join to discuss how abortion has played a role in elections this year and when we should know the results of next month’s midterms. Lastly, they all play a round of D.C.- and midterm-related trivia.

Where Our Model Thinks The Polls Might Be Biased

By

Oct. 28, 2022, at 9:52 AM

ILLUSTRATION BY EMILY SCHERER

With less than two weeks until Election Day, Republicans now have a 48 percent chance of taking the Senate, according to FiveThirtyEight’s Deluxe forecast, their highest figure since late July and putting the battle for Senate control . But whether that shift will continue or has peaked is in the eye of the beholder.

Most recently, the GOP hasn’t seen the rapid gains in the forecast that they did a week or two ago, and if you scroll through our , you can find of that have good news for Democrats along with that look strong for Republicans. Still, the current picture leaves Democrats without much margin to spare. Unless they can pull off an upset in North Carolina, Ohio, or Wisconsin, Democrats will need to win two of the three closest Senate races �� Georgia, Pennsylvania and Nevada �� in order to maintain their majority, while also holding Arizona and New Hampshire.

In the House, meanwhile, Republicans have an , putting Democrats in a tough �� although hardly impossible �� position. (, it’s as though they have pocket 5’s against pocket aces.) Unlike in the Senate, Democrats will have to do more than just win the majority of toss-up races to control the House. In fact, there are now 219 seats that our model rates as lean Republican, likely Republican or solid Republican, one more than the 218 needed for a majority. So even if Democrats won all the toss-ups, it wouldn’t quite be enough. To keep the House, Democrats will need for our model to be systematically underestimating them.

That can happen �� . That said, what the polls say and what the FiveThirtyEight forecast says are not quite the same. Most importantly, our Deluxe forecast already assumes the Republicans probably will beat what polls currently show.

Below, you can find a comparison between the FiveThirtyEight for each Senate race where we calculate a polling average, and the projected margin of victory or defeat in the same races according to the Deluxe forecast:

The Deluxe model expects Republicans to beat their polls

FiveThirtyEight polling average and projected margin of victory (or defeat) from FiveThirtyEight Deluxe forecast as of Oct. 27, in 18 Senate races

State Polling Average Deluxe forecast New York +15.5 +18.8 Illinois +14.7 +16.3 Connecticut +12.4 +15.8 Colorado +8.8 +7.6 Washington +8.6 +9.4 New Hampshire +5.3 +5.0 Arizona +3.6 +3.7 Pennsylvania +2.5 +1.5 Georgia +1.7 +0.3 Nevada +0.5 -0.3 Ohio -1.4 -4.5 North Carolina -2.6 -4.4 Wisconsin -3.4 -4.2 Florida -7.0 -9.2 Iowa -7.1 -13.5 Utah* -7.8 -13.2 Missouri -11.5 -18.1 Oklahoma�� -15.7 -25.3

* In Utah, the figures reflect Mike Lee’s margin against Evan McMullin, an independent.

�� In Oklahoma, figures reflect the regular Senate election; Oklahoma also has a special Senate election.

Senate races with a polling average on are included.

In Ohio, for instance, Democrat Tim Ryan trails by only 1.4 percentage points in our polling average, but is projected to lose by 4.5 points, which implies a 3-point polling bias against Republican J.D. Vance.

That’s on the high side for the competitive Senate races, but the Deluxe forecast does expect Republicans to somewhat overperform their polls in most of the other competitive races: by 1 to 2 points in Georgia and North Carolina, and by about 1 point in Wisconsin, Nevada and Pennsylvania. It doesn’t expect much polling bias in Arizona and New Hampshire, by contrast.

In some of the less competitive races, the Deluxe forecasts thinks the bias could be more severe. Check out Iowa, Missouri and Oklahoma, for instance, where Republicans are projected to do much better than their polling �� although there are relatively few recent polls of Missouri and Oklahoma. Meanwhile, Democrats are expected to outperform their polls in blue-state races such as Washington, Connecticut, Illinois and New York.

To some degree, this , although not perfectly so. But one fairly good heuristic is simply that Republicans tend to outperform their polls in red states, while Democrats do so in blue states �� see the map below for an example of this from 2016.

Why do Republicans do better in the forecast than the polling average? There isn’t any built-in or assumption of polling bias in the model. Rather, it’s a bunch of little things that tend to add up:

Our forecast makes additional adjustments to the polls beyond what is shown in our polling averages. In particular, this includes a likely voter adjustment and a timeline adjustment, both of which help Republicans at the margin. The likely voter adjustment helps Republicans because they tend to do better in polls of likely voters than registered ones. The timeline adjustment accounts for how the generic ballot has shifted since polls were conducted, and since it’s , that tends to help the GOP too.In some states �� Rust Belt swing states such as Pennsylvania, Ohio, Wisconsin, and redder ones like Iowa and Missouri �� the “fundamentals�� our model considers are more favorable for Republicans than the polls. The most important of these is simply . Republicans have done fairly well in these states in recent years, so you might expect them to do well again in what is likely to be a fairly good year nationally for the party.In addition, the model considers macro-level fundamentals, such as the poor historic performance of the president’s party at the midterms, and . So the model endorses the idea that Democrats . These factors zero out in the model by Election Day, though, so they don’t have too much effect now given how close we are to the election.Finally, the that the Deluxe model uses are more bullish for Republicans in several key Senate races than polls and other objective indicators.

While none of this is good news for Democrats, it should guard against one potential misinterpretation of our forecast. You might be tempted to take the FiveThirtyEight Deluxe forecast, and then mentally shade it further to Republicans because you worry that the polls could underestimate Republicans again. Heck, . But the Deluxe forecast already assumes that Republicans will do better than what current polls show, by around 2 percentage points in the average Congressional race. You may be double-counting, in other words, if you take the Deluxe forecast and then assume additional polling bias on top of it. Some of it is already baked into the model.

FiveThirtyEight   More: | |

In this live taping of the in Washington, D.C., Nate and Galen break down the current for the Senate, House and gubernatorial races.

Then Nathaniel Rakich and Amelia Thomson-DeVeaux join to discuss how abortion has played a role in elections this year and when we should know the results of next month’s midterms. Lastly, they all play a round of D.C.- and midterm-related trivia.

You can listen to the episode by clicking the “play�� button in the audio player above or by , the or your favorite podcast platform. If you are new to podcasts, .

The FiveThirtyEight Politics podcast is recorded Mondays and Thursdays. Help new listeners discover the show by . Have a comment, question or suggestion for “good polling vs. bad polling��? Get in touch by email, or in the comments.

In Part 3 of this installment of the , the team analyzes how an educational divide is currently shaping American politics and how a college diploma can influence an individuals�� beliefs and preferences.

In Part 2 of this installment of the , Equis Research co-founder Carlos Odio joins the pod to break down a new Axios-Ipsos poll that asked Latino Americans which party they favor in the midterm elections.

There are just two weeks until Election Day and according to the the race for the Senate has been a “dead heat.�� In Part 1 of this installment of the , the crew discusses whether the airtime reservations for each parties�� Senate campaign ads are impacting the forecast’s shift.

FiveThirtyEight   More: | |

There are just two weeks until Election Day and according to the the race for the Senate has been a “dead heat.�� In this installment of the , the crew discusses whether the airtime reservations for each parties�� Senate campaign ads are impacting the forecast’s shift. Then, Equis Research co-founder Carlos Odio joins the pod to break down a new Ipsos poll that asked Latino Americans which party they favor in the midterm elections.

Lastly, the team analyzes how educational divide is currently shaping American politics and how a college diploma can influence an individuals�� beliefs and preferences.

You can listen to the episode by clicking the “play�� button in the audio player above or by , the or your favorite podcast platform. If you are new to podcasts, .

The FiveThirtyEight Politics podcast is recorded Mondays and Thursdays. Help new listeners discover the show by . Have a comment, question or suggestion for “good polling vs. bad polling��? Get in touch by email, or in the comments.

With less than three weeks until Election Day, Republicans have noticeably gained ground in the . In this installment of “Model Talk,�� Nate Silver and Galen Druke discuss the reasons for Republicans�� improvement. They also explain why a dramatic shift among independent women in a recent ����dzܱ����’t be taken at face value �� but also ����dzܱ����’t undercut the poll. Finally, they answer listener questions about early voting and Fivey Fox’s statistical philosophy.

Why I'm Telling My Friends That The Senate Is A Toss-Up

By

Oct. 21, 2022, at 6:00 AM

ILLUSTRATION BY EMILY SCHERER

The Senate when we launched our forecast in June �� and after a summer in which political developments , there is now clear movement back toward Republicans. The FiveThirtyEight Senate forecast gives Democrats a 58 percent chance of keeping control of the Senate, down from a peak of 71 percent last month.

How to translate a roughly 60 percent chance into words is a tricky one, since it’s right on the threshold where you might choose to emphasize that one party is ahead or that it’s pretty close to 50-50. Poker players use the term “�� (short for “coin flip��) when two hands have about an equal chance of winning, even if the chances aren’t exactly equal. Most players would describe as a “flip,�� for instance, even though the pair of 10s would win 57 percent of the time.

In politics, the similar term “toss-up�� is often applied, although the definition is fuzzy. We’re even inconsistent in how we use it ourselves.

But let’s get real. If a friend asked me to characterize the Senate race, I’d say “it’s pretty fucking close,�� and emphasize that neither party has much of an advantage. Here’s why.

For one thing, as of Thursday afternoon, Republicans realized a slight lead (of 0.1 percentage points) in the FiveThirtyEight for the first time since Aug. 2.

Obviously, a lead of a tenth of a percentage point isn’t much. The advantage may have flipped back by the time that you’re reading this. But the tied generic ballot overstates the case for Democrats. That’s because our polling average takes generic ballot polls as they come, which are a combination of polls of likely voters, registered voters and all adults. Our model, however, and adjusts polls of registered voters and adults to make them more similar to polls of likely voters, which this year have been more favorable to Republicans. So a tie on the generic ballot among all polls translates to a slight GOP lead with the likely voter adjustment.

Something else that’s made this year’s forecast hard to convey in plain English is that there are three entirely plausible scenarios: A Republican sweep of Congress, a Democratic sweep or a split Congress (which would more likely involve a Democratic Senate and a GOP House, rather than the other way around). Until recently, a split Congress had the most likely of the three scenarios, though the chances of this outcome were under 50 percent. Now, however, a GOP sweep (41 percent) is more likely than a split (38 percent):

And all of that is before getting into the chance that the polls could overstate support for Democrats again, as they did in 2016 and 2020. This is a complicated subject; I the model does a good job of accounting for this, and one should keep in mind there’s also the possibility that the polls could be biased against Democrats. But I’m , so my mental model is slightly more favorable to the GOP than the FiveThirtyEight forecast itself. It’s worth noting, though, that some of the states where people had been most concerned about the polls being wrong, such as Wisconsin and Ohio, have toward the GOP .

But the main reason why I think of the race for control of the Senate as a toss-up �� rather than slightly favoring Democrats �� is because there’s been steady movement toward the GOP in our model over the past few weeks. In principle, past movement ����dzܱ����’t predict future movement in our forecast and it should instead . (We put a lot of effort in our modeling into trying to minimize .) This year, though, the forecast has moved in a predictable-seeming way, with a long, slow and steady climb toward Democrats over the summer, and now a consistent shift back toward Republicans.

What’s produced this pattern? It’s hard to know whether it reflects the real state of the race or is an artifact of how our model works. The summer produced an unusual streak of favorable developments for Democrats, from the to overturn the constitutional right to abortion, to , to in the midterms, to a string of . There is, of course, a tendency to see patterns in random noise, but it’s as though the coin really did come up heads for Democrats six or seven times in a row.

From a modeling standpoint, another challenge is that Democrats were defying political gravity. The president’s party . There have been some exceptions and . But the model has been trying to balance polls showing Democrats having a pretty good year against its prior expectation that the electoral environment should be poor for Democrats.

As the election nears, the model relies on its priors less and trusts the polls more, so it was initially skeptical of buying into a post-Dobbs surge for Democrats. Right about the time the model had fully priced in Democrats�� improved polling, though, the news cycle shifted toward a set of stories that were more favorable for Republicans, such as and .

It’s also possible to overstate the case for Republican momentum. Midterm elections in the way that presidential elections sometimes do. And there haven’t been any self-evidently important developments in the news cycle in the past week or so. If you’re one of those people who thinks , they’ve .

Rather, this is more a case of now having more evidence to confirm that the Democrats�� summer polling surge wasn’t sustainable.

That doesn’t mean it was fake: In fact, Democrats had a string of in which they met or exceeded their polling. If you’d held the midterms in late August, I’d have bet heavily on Democrats to win the Senate. It sure would be nice to have another special election or two now, and to see how these polling shifts translate into real results. Polls can sometimes change for reasons that don’t reflect the underlying reality of the race, such as because of or .

And certainly, Democrats have plenty of paths to retain the Senate. Republicans don’t have any sure-fire pickups; Nevada is the , and even there, GOP chances are only 53 percent, according to our forecast. Meanwhile, Democrat John Fetterman is , although his margin over Republican Mehmet Oz has narrowed. The model is likely to be quite sensitive to new polling in Pennsylvania going forward. If Democrats gain a seat there, meaning that the GOP would need to flip two Democratic-held seats to take the chamber, that starts to become a tall order. Nevada, sure, but I’m not sure Republicans would want to count on Herschel Walker in Georgia or Blake Masters in Arizona.

But the bottom line is this: If you’d asked me a month ago �� or really even a week ago �� which party’s position I’d rather be in, I would have said the Democrats. Now, I honestly don’t know.

FiveThirtyEight   More: | |

With less than three weeks until Election Day, Republicans have noticeably gained ground in the . In this installment of “Model Talk,�� Nate Silver and Galen Druke discuss the reasons for Republicans�� improvement. They also explain why a dramatic shift among independent women in a recent ����dzܱ����’t be taken at face value �� but also ����dzܱ����’t undercut the poll. Finally, they answer listener questions about early voting and Fivey Fox’s statistical philosophy.

You can listen to the episode by clicking the “play�� button in the audio player above or by , the or your favorite podcast platform. If you are new to podcasts, .

The FiveThirtyEight Politics podcast is recorded Mondays and Thursdays. Help new listeners discover the show by . Have a comment, question or suggestion for “good polling vs. bad polling��? Get in touch by email, or in the comments.