In our era of polarized politics, downhill races are increasingly driven by partyism. However, we keep our attention on each candidate – in the Senate races, Joni Ernst’s inability to name the soybean price, or Cal Cunningham’s out-of-marriage relationship. Can races be simultaneously nationalized (ie, track party association) and separated from the presidential race?
This is evidence that both can happen at the same time. It points to the possibility in close Senate races like South Carolina, Iowa and Montana, last-minute phone banking, money transfers and even (for Montana) voter registration the same day can be of particular value in moving a race.
Above is a chart showing the probes obtained from PEC’s public archives for the 13 major Senate races are currently within 10 percentage points. Montana and Mississippi are ignored because there is very little data from the summer. Each state is graphed against President data for the same day (“EV” file in the archive).
The matching red line has a slope of 0.86, which means that for every 1 point of difference between states in the Presidential polls, the corresponding Senate poll is 0.86 points in the same direction. The number is close to 1 and shows that during the summer, Senate races mostly reflect the presidential race. The party leading the presidential race also led the race to the Senate in all but two states, Iowa (where Greenfield did better than Biden) and Georgia (where Perdue outperformed Trump). So in August the coattail effect is quite strong.
By the end of October, this effect was much weaker. The slope of the fitted black line is 0.43, only half that of August. This meant that Senate races were much less dependent on the presidential race.
Initially, one might conclude that as the fall unfolds, voters start paying attention to the Senate candidates for their own worth. However, there is evidence that partisan still has problems. The leading party for the President and the Senate is currently a perfect fit (although Iowa, at Biden + 0.1% and Greenfield + 0.5%, is actually a draw). This is reflected in two parameters: the correlation coefficient r is high, +0.88 in August and +0.81 in October. In addition, both matching lines go almost all the way.
What does that mean? To me it says that while the Senate and Presidential races are somewhat separate from each other, the two are still driven by a deeper factor: partisan loyalty.
When we react to individual races, we might think we are forming stories that are fundamentally driven by events. That’s certainly how political news sounds like that. For example, it is said that Jon Ossoff (D-GA) brought Perdue (R-GA, incumbent) into their debate last week. But to what extent are those stories just post-graduate reasons to support our fundamental partisan biases? What would a Perdue advocate say? This also explains why news of Cal Cunningham’s (D-NC) adulterous relationship has so little appeal. If Cunningham’s party was a bit popular in North Carolina, where partisan polarization was, then his texting habits could be a problem.
Having established party as key, let’s get back to the question of when will more people begin to track races into the individual Senate.
The separation between the Senate and Presidential races has been gradual, starting in July. On International Labor Day (July 25 this year), the Senate slope relative to the President has almost reached its present value.
Senate heads are outperforming their party’s presidential nominees in most states. The exceptions are Texas, where Cornyn (R) is safe but Trump is losing within 1 percentage point, and Arizona, where Kelly (D) is 7 points ahead but Biden is only 4 points ahead. So overall, it seems that Senate candidates should bind themselves as closely as possible to their own presidential candidate. That, for example, explains why Sen. Perdue (R-GA) chose to participate in a Trump rally rather than engage in a second debate with Jon Ossoff (D). The aforementioned public response to the first Ossoff-Perdue debate may also play a role.
All of this is a fancy look at the fact that in recent weeks Senate races have narrowed even more to Presidential races. It means there are only some races where a state can end up electing candidates for the Senate and presidents of the opposition parties. But there are a few.
Here are the races to the Senate that are error polling. Based on data from past Presidents’ years, Senate polls could drop by as much as 3-4 percentage points overall. The states where individual candidates do the best in overcoming the partisan leanings of their states are Montana and South Carolina. If you use phone banking this weekend, these are great places to gather. And of course Iowa, basically a tie.
(And if you are interested in state legislature and Redistricting Moneyball, don’t forget Texas and North Carolina!)