Votes are still being counted, particularly out west where mail-in ballots are the predominate form of voting. But we are down to a handful of races that are still too close to call.
There are three things to consider at this point. The first is how many ballots remain to be processed and where those ballots are. The second is that Alaska uses ranked-choice voting. While the U.S. does not have much experience with ranked-choice voting, Australia does. The Australian experience has taught us is that second and third choices tend to be for the candidate closest to the first choice but there will be some drop off and a minority will go to the “non-similar” candidate. As such, it is hard for a trailing candidate to close the gap much, and the trailing candidate usually wins only if the original margin is small, there are a significant number of votes for the “also-ran” candidates. Third, the vote counting machines are very accurate. There will be some voter errors that are only caught by visual inspection, but recounts rarely change the final count by much. Closing a 1,000 vote margin in a recount is almost impossible.
Turning to what is still outstanding, depending on the media site that you use, the only outstanding Senate race is Pennsylvania. The real issue in Pennsylvania is provisional ballots. The current margin is 18,000. Senator Bob Casey is receiving about 60% of the vote from provisional ballots so far, and most of the remaining provisional ballots are in areas that favor Senator Casey. There will be a recount, but, as noted above, it is unlikely that the recount will change the vote total by more than 1,000. Senator Casey’s chances depend entirely on how many additional provisional ballots are counted.
There will be recounts in several house races — Maine’s Second District, Iowa’s First Disrict, and Ohio’s Ninth District. In Maine’s Second District, Representative Jared Golden led by around 2,000 on first choice and grew that lead to 2,500 when preferences were applied (picking up around 1,000 votes to around 500 with the remaining ballots not ranking either candidate on the approximately 13,000 ballots that voted for a write-in candidate as first choice). It is unlikely that the Republican challenger will win the recount, and most media sites have now projected this race for Representative Golden.
In Ohio’s Ninth District, there are apparently 8,000 outstanding absentee and provisional ballots. Representative Marcy Kaptur currently leads by around 1,100 votes. That is probably large enough of a lead that the recount should not alter the results, but if the margin declines as the last 8,000 ballots are processed, there is a chance that the Republicans could flip this seat.
In Iowa’s First District, the initial count has been certified. Representative Marianne Miller-Meeks is currently ahead by just over 800 votes. There is a slim possibility that the Democrats could gain this seat but that margin would represent a very large swing compared to most swings from a recount.
In short, the current leaders in all three races should win which means no additional changes of seat.
The “still being counted” is primarily California. Depending on which media projection desk that you follow, either two or three seats are outstanding. Many have called the Twenty-first district for Representative Jim Costa. He has a lead of 8,000 votes with about 90% of the vote counted. The Republican is unlikely to make up that gap. The two close races are the Thirteenth District and the Forty–fifth District. Both are currently held by Republicans. Representative John Duarte is currently ahead by about 2,000 votes. Democrats tend to do better in “late arriving” ballots, but it is unclear if there are enough outstanding ballots in Democratic areas for the challenger to catch up. Representative Michelle Steel has now fallen thirty-six votes behind her Democratic challenger. The remaining votes are probably mostly from Orange County where Representative Steel has done slightly better. In short, this race could be headed to a recount and it could be close enough that either candidate could win. In short, it is almost certain that Democrats will pick up at least one of these seats and could pick up all three. My hunch says two of the three. If the Democrats sweep all three, that would be a gain of two which could be significant because of the last race outstanding.
The last race is Alaska. As in Maine, Alaska has ranked choice voting. There are still some votes left to process but Representative Mary Peltola is about 9,000 behind her challenger Nick Begich (who finished third in 2022). The other two candidates have around 16,000 votes between them. However, getting a 50-50 split of the that vote would be an accomplishment for Representative Peltola. Getting 80% of that vote is probably impossible. In first preference votes, Representative Peltola has over 45% of the vote. Given how red Alaska is that result is amazing. In all probability, this seat will flip to the Republicans. It was always going to be an uphill battle. The great job that Representative Peltola has done is reflected by her strong performance (beating the presidential ticket by about 5%), but any sane Republican should be able to win Alaska. Representative Peltola won in 2022 because Alaska Republicans prefered the not-sane Sarah Palin over Mr. Begich. When the choice was between Representative Peltola and Palin, voters went with Representative Peltola. If the Republicans had opted for Mr. Begich over Sarah Palin, he would have won in 2022.
The bottom line for the House is that depending on what happens in the two close districts in California, the final breakdown seems likely to be 220 Republicans to 213 Democrats. If the Democrats win both districts, they will have gained one seat. If the Republicans win both seats, they will have gained one seat. So much for presidential coattails or a red surge. President Trump has named several House members to the cabinet. That is going to mean some vacancies for the first part of next year. The Republicans are likely to keep all of those seats, but it will take several months to hold those elections. And that is going to make it even harder for the Republicans to pass any legislation. If the Democrats do pick up both seats in California, it will only take three Republicans to defect to defeat any extreme legislation. I can easily see three votes against much of what the Freedom Caucus wants. The issue going forward is whether President Trump and Mike Johnson can get the Freedom Caucus to accept bread crumbs to keep 218 Republicans on the same page. If not, not much of significance is going to pass the House. And, unless Republicans abolish the filibuster, nothing that the Freedom Caucus wants is going to get through the Senate outside of reconciliation bills.
UPDATE (11/23):
As expected, the margins in Pennsylvania were too much for Senator Casey to close the gap and win. While Representative Peltola gained some ground when preferences were distributed, she did not gain enough to keep her seat. And the last votes appear to have pushed the margin in Representative Kaptur’s race beyond the margin for a recount.
The recounts in Iowa and Maine appear to be ongoing. The early results from Iowa seem to indicate that the recount is, so far, matching the election day results. For the reasons noted above, that is only a mild surprise. But Iowa apparently does a machine-based recount (i.e. it runs the ballots through the machine and only looks at the ballot that the machine can’t read) rather than a hand recount (in which every ballot is individually examined). A machine recount is less likely to have deviations than a hand recount (in which writing on the ballot may cause the judges to read the ballot differently than the machine does) and will only catch errors in processing the ballot (e.g., failure to run some absentee ballots through the machine or running the same batch twice).
That leaves the two close races in California where we are getting down to the last ballots to be processed. It is hard to know for sure how many ballots are out in the two districts as both districts are multi-county districts with most of the counties in each of these districts being split among multiple districts. That makes it hard to know for sure how many of the ballots remaining in each county are from these two races. Assuming a random distribution of ballots in each county (in other words that, among the outstanding ballots, the share of ballots from these two races will be similar to the share of ballots from these two races and that the votes in these two races will be similar to the votes already counted in those two races), both races are still very close. In the Thirteenth District, the Republican is up by 204 votes, but (based on those assumptions), the Democrat is likely to close the gap significantly. This race could be decided by less than fifty votes, and it will depend on how accurate those assumptions are. in the Forty-Fifth District, the Democrat is up by over 500 votes. While the Republican incumbent might gain some votes from what remains, she will need to do better than she has been doing (and late arriving mail-in votes tend to be slightly more Democratic than the votes processed early). The last votes in this race are unlikely to move the needle significantly. In short, it looks like the split in the next house (excluding vacancies) will be 220-214 with one race (the Thirteenth District) still very much up in the air.
