by Jeffrey H. Anderson
Americans will soon get to cast their first votes since the science–denying COVID mask and vaccine mandates, the second wave of COVID-related blowout spending and subsequent inflation, and the COVID-related school closures that allowed parents to see what the public schools are really teaching their boys and girls – including that they can choose whether they are boys or girls. With all of these matters implicitly on the ballot, how are things shaping up going into Election Day?
Starting with the House of Representatives, six months ago Amy Walter of the Cook Political Report projected “a GOP gain in the 15-25 seat range.” At the time, I responded, “While things could change over the next six months (although the cake is probably largely baked), a GOP gain of 30 to 40 House seats appears more likely at this stage of the contest than Walter’s projected GOP gain of 15 to 25 seats.”
Fast-forwarding six months, Cook now projects (as of November 6) Republican gains of between 0 and 35 seats, with a midpoint of 17.5 seats – so, slightly below the midpoint of its range of projected outcomes in April. Nate Silver’s FiveThirtyEight now says that the most likely scenario is a Republican gain of 15 House seats, and it still maintains that Democrats have a 17 percent chance of holding the House.
RealClearPolitics, meanwhile, is now projecting that Republicans will gain between 14 to 48 House seats, with a midpoint of 31 seats. Since Republicans need only 6 seats to obtain a majority, RCP’s projections amount to something like a 99 percent chance that Republicans will gain control of the House. In line with RCP, I’ll stick with my earlier projection of a Republican gain of 30 to 40 House seats, which would produce a GOP margin of about 50 to 70 House seats.
In Senate races, an expansive map suggests a range of possibilities spanning from having Democrats narrowly maintain control of that chamber to Republicans moving perhaps halfway from their current tally of 50 seats (if we’re counting Mitt Romney and Lisa Murkowski as Republicans) to a filibuster-proof 60 seats – with a favorable map to come in 2024.
The Cook Political Report, as I have previously explained, can be a valuable resource for projecting Senate races, but this requires using a handy decoder to account for Cook’s consistent left-leaning bias. Over the past four federal elections, Republican Senate candidates have won a whopping 72 percent of Cook’s “toss-up” races, posting a win-loss record of 23-9. At the same time, they have posted an 11-0 win-loss record in Cook’s “competitive” races that merely “lean” Republican, winning by an average margin of 14 percentage points – 6 points higher than Democrats’ 8-point average margin of victory in competitive races that Cook says “lean” their way.
Cook currently lists four Senate races as toss-ups (Arizona, Georgia, Nevada, and Pennsylvania). If Republicans win three of those four, effectively matching their usual winning percentage in Cook’s “toss-up” races, and if other states go as Cook projects, that would give Republicans two pickups, 52 Senate seats, and control of that chamber.
FiveThirtyEight, meanwhile, is projecting (as of November 6) that the most likely scenario is for Republicans to pick up one Senate seat and gain control by a 51 to 49 margin. Offering a different assessment, RealClearPolitics projects that Republicans will gain three seats and end up with 53 senators.
Two weeks ago, I highlighted what appeared to be 11 competitive Senate races. One of those no longer qualifies, as Republican incumbent Chuck Grassley’s (R-Iowa) lead has ballooned from 3 points to 12 points in Des Moines Register polling. That leaves these 10 competitive races, with Republicans needing to win five to take the Senate and Democrats needing to win six to hold it:
10) Ohio (Advantage: Republicans): Republican J.D. Vance hasn’t trailed Democrat Tim Ryan in a poll listed by RCP since late September, and he currently (as of November 6) leads by 7.5 points in the RCP average of recent polling – after leading by less than a point just three weeks ago. In 2020, polling grossly underestimated Donald Trump’s support in Ohio – being off by 7 points. Vance will prevail unless this year’s polling is off by even more in the opposite direction, which isn’t likely.
9) North Carolina (Advantage: Republicans): Republican Ted Budd has led in every poll listed by RCP since the start of October, and the Tar Heel State was almost 6 points to the right of the nation in 2020 (as Joe Biden lost there by 1.3 points while winning by 4.5 points nationally.) It would be something of a shocker at this point if Democrat Cheri Beasley (who trails by 5.2 points) were to win, but this remains a dark-horse competitive race.
8) Wisconsin (Advantage: Republicans): In 2016, Ron Johnson performed 6.1 points better than the RCP average (winning by 3.4 points when polling had him behind by 2.7 points). This time, he’s up by 2.8 points versus Democratic challenger Mandela Barnes. Unless anti-incumbent sentiment is even stronger than polling indicates, Johnson appears to be in the driver’s seat.
7) Colorado (Advantage: Democrats): Democratic incumbent Michael Bennet looks beatable, and Colorado remains a swing state (albeit one that leans Democratic). But Republican Joe O’Dea – who praised Bennet’s role on immigration as a member of the Gang of Eight during a recent debate – might not have drawn enough of a contrast for voters. The RCP average has Bennet up by 5.3 points. Trafalgar has O’Dea within 2 points, however, and this race could still surprise.
6) Washington (Advantage: Democrats): If Republican Tiffany Smiley manages to upset longtime incumbent Democrat Patty Murray, who has held this seat since Bill Clinton entered the White House, this will be a night to remember for Republicans. While Murray still leads by 3 points in the RCP average of recent polling, Smiley – who has never held elective office of any sort – is positioned for a possible huge upset. FiveThirtyEight gives that upset only a 9 percent chance of happening, and Cook still says this race is “not considered competitive.” Still, the guess here is that the odds of Smiley’s winning are notably higher than the odds of having Stephen Curry – a career 91 percent free-throw shooter – miss a foul shot.
5) Nevada (Advantage: Republicans): Republican Adam Laxalt leads Democratic incumbent Cathy Cortez Masto by 2.4 points in the RCP average. Perhaps the greater concern for Cortez Masto is that polling shows her with only 44.9 percent support in an environment in which it seems unlikely that a lot of late-deciding voters will break toward incumbents of the same party as the president, whose approval rating (per RCP) is 42.4 percent. This is a state, however, in which polling has sometimes overstated Republican support. Trump fared 3.2 points worse than expected in 2016 (losing by 2.4 points when favored by 0.8 points), and Republican Dean Heller fared worse than expected in 2018 (losing by 5 points in what was supposed to be a dead heat), although polling was quite accurate in 2020.
4) Georgia (Advantage: Republicans): Republican Herschel Walker, the former Heisman Trophy-winning Georgia Bulldog, was behind by 5.2 points in the RCP average on October 9 but is now running neck-and-neck with Democratic incumbent Raphael Warnock (with Walker leading by 0.4 points). Warnock sits at 46.8 percent support in the RCP average, so – like Cortez Masto – he would need to pick up a fair number of late-deciding voters. If neither candidate gets a majority of the vote, this race will go to a runoff in December.
3) Pennsylvania (Advantage: Republicans): Mehmet Oz and John Fetterman are separated by just 0.1 points in the RCP average. Oz, who was down by 4 points a month ago, appears to have some late momentum, perhaps especially in the wake of the candidates’ October 25 debate – but early voting in the state was well underway by then. Polling in Pennsylvania underestimated support for Trump by 2.6 points in 2016 – he won by 0.7 points after being behind in the final RCP average by 1.9 points – but was accurate in 2020. This contest could easily go either way, but Oz’s recent surge, combined with the likelihood that Biden’s low approval rating won’t help Fetterman with late-deciding voters, appears to give Republicans a slight edge as this race heads to the wire.
2) New Hampshire (Advantage: Democrats): The Republican establishment has tried hard to lose this race by refusing to fund the GOP primary winner, Don Bolduc, a retired Army general and political outsider. According to Open Secrets, Democratic incumbent Maggie Hassan has outraised General Bolduc by an astounding margin of 17-1. Yet Bolduc, who trailed Hassan by 5.4 points in the RCP average as recently as October 20, is now within 1 point (48.4 to 47.4 percent), with all of the polls in the current RCP average having been taken after the candidates’ sole debate.
Cook still calls this race Democratic-leaning, but don’t be surprised if Cook moves it to toss-up status by Election Day. New Hampshire is a quintessential swing state and is the only state to have been within 4 points of the nation (in either direction) in each of the past seven presidential elections. The Granite State also still believes in Election Day – not Election Month – and this is anyone’s race, with Hassan appearing to hold just a sliver of an advantage based on recent polling.
1) Arizona (toss-up): The Republican establishment has also tried to lose this race, but Republican Blake Masters has closed a 5.5-point deficit on October 15 versus Democratic incumbent Mark Kelly to just one point in the current RCP average (48.2 percent to 47.2 percent). Masters should also be helped by the late withdrawal of libertarian Marc Victor, who dropped out on November 1 and endorsed Masters – a development that is only partially reflected in current polling. Don’t expect even Fox News to call this race early.
So, of these ten competitive races, Republicans appear to have the advantage in six and Democrats in three, with one race not favoring either party. If each party wins the races in which it now looks to have the advantage, Republicans would end up with 52 Senate seats. To keep control of the Senate, therefore, Democrats would need to win the three competitive states where they currently appear to have the advantage, win the toss-up state of Arizona, and win two of the six competitive races in which the GOP now appears to have at least a narrow edge.
RCP’s Sean Trende in January highlighted his model for Senate races, which is based primarily on the sitting president’s approval rating. Trende’s model indicated that if Biden had an approval rating of 42 or 43 percent on Election Day – his current approval rating is 42.4 percent – Republicans would pick up between two and five Senate seats, giving them between 52 and 55 seats. That sounds more plausible than FiveThirtyEight’s current claim that the most likely range is between 49 and 52 Republican seats – even though FiveThirtyEight has gotten to incorporate about nine months’ worth of additional information in making its projections. The guess here is that the GOP will end up with between 51 and 54 Senate seats. We’ll soon see the American people’s verdict.
– – –
Jeffrey H. Anderson is president of the American Main Street Initiative and served as director of the Bureau of Justice Statistics at the U.S. Department of Justice from 2017 to 2021.
Photo “Blake Masters” by Gage Skidmore. CC BY-SA 2.0. Photo “Mark Kelly” by Senator Mark Kelly.