Friday, October 28, 2022

BTRTN Midterms Snapshot: For Dems, House is All But Lost, Senate Now a Toss Up

Tom with BTRTN’s latest on the race for control of Congress.

Our latest BTRTN midterms update highlights are as follows:

·        Electoral environment.  The electoral environment has swung back to the GOP, indicated by the generic ballot, which is now R+3, and by individual race polling, which has moved consistently toward GOP candidates. 

·        Senate.  At this point, control of the Senate could go either way, though indicators still, on balance, very slightly favor the Democrats. 

·        House.  The odds of the Democrats maintaining control of the House have fallen to single digits and the question is not whether the GOP will take the House, but by how much. 

·        Governors.  We expect the Democrats to flip two state houses rather easily, but there are only five competitive races that offer other realistic flip opportunities; if there are no other flips, state houses would be split 25/25.

·        Priorities.  Democratic volunteers and donors should focus their efforts on the highest leverage Senate races, which could come down to three pivotal contests.

Here are the latest BTRTN’s odds of the Democrats maintaining each of the two chambers, the expected number of seats that will change parties, and some key metrics, all compared to prior updates.

Why has the political environment gone so sour for the Democrats in these last two weeks?  GOP messaging has sharpened around two issues that seem to be resonating among the persuadable, inflation and crime.  The October 10 inflation report – the last before Election Day – put to bed any Democratic hope that there was evidence that inflation was being tamed.  GOP candidates have gleefully kept the economic struggles of Americans front and center (even as the stock market has rebounded significantly over these two weeks, and gas prices have fallen). Crime has surged ahead of abortion in overall import to voters (the economy is still the top issue) and the GOP is viewed as the party more likely to deal with it effectively..

There is still time for this picture to change.  Remember, these assessments are “snapshots,” not “predictions,” based on today’s conditions and data.  (We’ll get into the prediction business on November 7, the day before Election Day.)   But time is running short for late-breaking “surprises” that could stop or reverse GOP momentum, and, of course, early voting is well underway in some states.


Senate battleground race polls have tightened considerably over the past four months, and the Democratic candidates’ collective lead over their GOP counterparts is now nearly negligible.

We have made five ratings changes in the Senate, all five in the direction of the GOP, though none have resulted in a change in the party in the lead. 

Democrats have to be alarmed at the narrowing of two races that seemed reasonably secure (if not solid) in New Hampshire and Arizona.  Both have followed similar trajectories and neither Democratic candidate (Maggie Hassan and Mark Kelly) has committed any particular gaffe; their woes are simply a vestige of a political environment gone bad for the Democrats.

Pennsylvania is somewhat different.  The race was narrowing even prior to the only debate scheduled between Democrat John Fetterman and Republican Dr. Mehmet Oz.  Oz relied heavily on the crime issue and heavy media spending to reduce the Fetterman lead to about three points.  But the debate was perhaps just short of a prime time disaster for Fetterman, who exhibited the effects of the stroke he suffered in May with his halting, repetitive and at times inarticulate delivery.  It remains to be seen whether the debate proves damaging to him, as Oz himself made the rather remarkable claim that abortion decisions should be left to “women, doctors, local political leaders…”, a gaffe that is already running in heavy rotation in pro-Fetterman ads.

The two GOP Senate incumbents expanded their leads, with Ron Johnson of Wisconsin opening up daylight over Democratic challenger Mandela Barnes, and Marco Rubio all but putting away Val Demings, elevating the Florida race out of battleground status with his high single digit lead. 

There are thus now eight “battleground” Senate races (races in which both parties have a legitimate shot of winning) and the Democrats remain nail-bitingly ahead in five of them.  They need to win four to maintain control of the Senate, so this current “snapshot” continues to point to the Democrats holding the Senate, even flipping one seat (Pennsylvania).  But that belies how tenuous their position has become.  Each new poll seems to demonstrate continued GOP momentum, and with fully five races in the “toss-up” category, and three others in play, the Democrats could also lose big.

The full range of potential outcomes is demonstrated below; there is a 58% chance the Democrats will maintain the Senate by controlling 50 or more seats, but the chart also shows the potential for a range of adverse outcomes.

The chart below lists all the Senate races, with a focus on the eight battleground between the two purple lines.

For Democratic volunteers and donors, our guidance would be to prioritize the races as follows

·        Control of the Senate might very well come down to whichever party wins two out of three of George, Nevada and Pennsylvania – those three would be the top priority.

·        The next tier would be to bolster New Hampshire and Arizona, to prevent race dynamics from slipping any further (the GOP candidates have made significant inroads in both thus far)

·        Volunteers should probably stop there, but if one is inclined to go beyond, Ohio might be the next bet, followed in turn by North Carolina and then Wisconsin.


To understand the dynamics of House races, the best indicator, by far, is the “generic ballot.  This polling question has, over the years, been very highly correlated with how many seats each party will win or lose, especially in the midterms (as opposed to presidential election years).

That is bad news for the Democrats, because the generic ballot took a sharp and material turn in the direction of the GOP, who now lead by +3 points, back to where it was in June.  If the election was held today, the Democrats would have only about a 1 in 10 chance of holding on to the House, according to our models.

As we have pointed out before, even when the generic ballot was a dead heat, as it was just a few weeks ago, that did not mean the race for control of the House was a dead heat as well.  Given the disproportionate GOP representation in the House (due to overrepresentation of small states and effective gerrymandering), the Dems need to do much better than even in the generic ballot to have the upper hand in controlling the House.  For this race to become truly competitive, the Dems need to build their generic ballot advantage to D+3 or even D+5.  Clearly, the Dems are a long way from that, and are moving in the wrong direction.  Should there be no change in the last 10 days, the Dems are looking at a 25-30 seat loss. 

The GOP is in a commanding position.  To hold onto the House, the Democrats have to win not only all of the toss-ups, but also most of the races that are currently leaning to the GOP.

Here is a breakdown of all the races that are potentially in play, arrayed across the categories above.  For the Democrats to maintain control of the House, they need to win all of the Toss-up R races, and most of the Lean R as well.


As we have noted before, Governors became national figures with the onset of Covid as well as with the GOP push for more state control of electoral processes in the wake of The Big Lie.  A number of governors figure prominently in 2024 presidential speculation.  Thus more attention is being paid to gubernatorial races in 2022 – and there are a whopping 36 of them at stake.

There are only five races that are truly competitive at this point, but four of them are toss-ups.  The other tight race, in Kansas, has had little public polling overall and none at all in October, so it is difficult to assess exactly where it might stand right now.  This snapshot has the governor outcome at 25/25, another reflection of the polarized nature of our country.

Among the races that are not competitive are the only two states where we see flips in this snapshot, Maryland and Massachusetts.  Both are blue states but have GOP governors who have termed out, Larry Hogan of Maryland and Charlie Baker of Massachusetts.  Both will almost surely be replaced by the Democratic candidates in the races to succeed them, Wes Moore (in Maryland) and Maura Healy (in Massachusetts).


Here are the details on those five battleground races.  The focus should be on those toss-up races and Kansas.


We’ll be back with more updates in the final 10 days, and our official predictions on November 7.  Stay tuned.


  1. I have to respectfully disagree with your analysis, I'm actually a lot more optimistic about this midterm. I assume the +3 R generic ballot average comes from RealClearPolitics? In the last month right-leaning pollsters (Rasmussen, Trafalgar, InsiderAdvantage, etc.) have been flooding the media with crappy polls that show Republican leads. RealClearPolitics openly admits to cherry-picking polls to make Republicans look as good as possible. FiveThirtyEight includes these polls in the average, which I don't agree with either, but their average only shows an R +0.7 advantage. If you look at the list of all generic ballot polls (, there is a grand total of one nonpartisan poll from the last week that shows any Republican advantage (Suffolk University - R+5).

    Even on the best polls the Likely Voter screens are based on turn-out demographics from 2018 and 2020. But there's a lot of evidence that young people and women may turn out in higher numbers than in either of those two elections, which were both good for Democrats. This is just one poll so you should take it with a grain of salt, but the Harvard Youth Poll predicts a 40% youth turn-out this year, which would be absolutely massive ( And the available registration and early voting data, which is the only real data we have, are showing massive advantages for Democrats especially in swing states (, It's possible that Republicans will match this turn-out on election day, but they better hope there's no bad weather in any swing state.

    I think there's a good argument to be made either for a blue wave, a red wave, or a pure toss-up. We'll see how it goes.

    1. Yes, Brandon, good analysis but we're dancing on the head of a pin here. 538 has the generic ballot at R+1, RCP has R+3, BTRTN now (as of this morning) has it as R+2. Any of those outcomes lead to the Dems losing at least 15 seats and probably more like 20. No one has it at D+5 and that is where it would have to be for the House to be competitive. As for the youth vote, this has been Waiting for Godot. I must have missed the huge youth rally in DC when the Dobbs decision came down, and the huge outcry and protest by youth over Jan 6. I'm a child of the 1960's, and I know outrage when I see it. I don't see it. By any measure, "voter enthusiasm" is now either about even or favoring the GOP, presumably inclusive of young people. I hope you are right but I am skeptical!

    2. One other quick thing..."GOP" polls were on balance probably more accurate than "D" polls in 2020 in swing states.

  2. Since the Dobbs decision, there is a large difference between the Generic Ballot for the House between those polls of "Likely Voters" and "Registered Voters."
    advantage LV RV
    Democrats 7 38
    Republicans 47 11
    Tie 2 5
    # of D polls (40) 27

    ** I know there are a number of different basic methods used to determine who is "Likely," plus the "secret sauce" nuances among different firms. But is there any explanation of what their assumptions would look like for participation rates of various demographic or partisan groups and the overall numbers of those voting? Any comparison between what the results are now versus 2018?

    ** Is the extensive difference between LV and RV results similar to other years? I've not gone back to check, but my recollection is in 2018 there was some distinction of margins, but both types consistently favored Democrats.


Leave a comment