Last week, I looked at the history of Senate races in presidential years. As I noted, there was some good news for Republicans even if Donald Trump loses: The average losing party loses two Senate seats. Among losing parties that got at least 48 percent of the two-party presidential popular vote on an unweighted average across states with Senate races, the average was basically break-even, with the losing party gaining Senate seats four times in six elections. And Donald Trump’s unweighted average in the two-party vote in polls across those states with 2020 Senate races is 52 percent. The bad news? The average losing party goes 13–21 in Senate races. The Republicans, however, are defending 23 seats, so they need a lot more than 13 wins to retain control of the Senate. And the two races that saw the losing party do terribly in the Senate despite doing well in the popular vote across states with Senate races were 2008 and 2012 — the last two times Republicans lost a presidential race.
Let’s update the charts from that piece. Here are the states with contested Senate races, and where they stand:
Remember: Assuming Doug Jones loses in Alabama, Democrats need to gain four Republican-held seats to get the Senate to 50/50 (which they would then tenuously control if Kamala Harris is vice president), five to get to 51 Democratic senators, and more than that if any other Democrat loses (Tina Smith in Minnesota and Gary Peters in Michigan being the two conceivable upsets, with both below 50 percent in the final poll averages, as are several Republican incumbents who lead in the final polls). At the moment, Democrats hold a fairly strong position to gain three of those — Arizona, Colorado, and Maine — and lead in North Carolina, while trailing in Iowa, Kansas, Alaska, South Carolina, Montana, and Texas. In most of these states, the Senate race is at least broadly consistent with the state of the national race, though not lockstep with it. The real wild card is Georgia, particularly the Perdue-Ossoff race, where it is possible that a bad night for Republicans could pull Jon Ossoff over 50 percent and avoid a January 5 runoff in which Republicans would likely have an edge. It is not outside of the question, however, for Perdue to avoid a runoff. The multi-way race in the other Georgia campaign seems less likely to get to 50.
It is not hard to imagine, in worst-case terms, how a really big Biden victory could lead to the sort of blowout we had in 2008, with Democrats gaining 8–10 seats. But the terrain does not favor that, not with Trump likely to either win or remain closely competitive in North Carolina, Georgia, Texas, Iowa, Kansas, Arizona, Montana, South Carolina, Alaska, and Kentucky. Biden’s coattails in states without Senate races won’t matter much in those places. If Joni Ernst wins but Thom Tillis loses, the likeliest outcome is a 50–48 Democratic Senate with the two Georgia seats headed to runoffs.
To complete the picture, the uncompetitive races, none of which seems likely to produce an upset: