With just less than four months to go before the November elections, we felt it was time to roll out our PAAR ratings for the US Senate races occurring this year. After the 2010 election cycle, we created our PAAR model (Percentage Above Anticipated results) to create a baseline of expectations for the upcoming election. This model is based upon our thesis that as elections are more and more nationalized and voters are increasingly locked into their partisan preferences, demographic changes within the electorate ultimately play the deciding factor in electoral outcomes. PAAR does not attempt to make exact predictions for each race, but instead is a measurement that shows what a candidate should receive based on their party affiliation. We believe that by recognizing those candidates and campaigns that did or did not live up to expectations, we can help identify new strategies for upcoming elections and manufacturing future wins.

Said more directly, we need to stop viewing wins as the only measure of success. Similar to baseball moving away from home runs, batting average, and runs batted in as the only measures of a player’s worth, the parties, and their related infrastructures, need to move away from only viewing a well-run campaign as a campaign that wins, with no other context placed around the results.

As a quick refresher, unlike most political models, our model uses no polling results and does not take the specific candidates or their backgrounds into account. Instead, our model is based heavily on demographic patterns and past election results.

In 2014, PAAR projected four months ahead of the November elections that the most likely outcome was Democrats losing majority control of the Senate, with the Democrats holding 47 seats versus 53 seats for Republicans (actual result was a 46-54 split). As we discussed in our look back, Democrats on average did -2.5 points worse than they should have, while Republican candidates averaged 0.3 points better than expected. Overall, just six Democratic candidates over-performed expectations (seven if you count Greg Orman who ran as an independent in Kansas), compared to 15 Republican candidates that did better than expected.

While Democrats are still feeling the fresh burn of the 2016 Presidential election, many have been motivated by the promising sign of special elections swinging comfortable red seats into close contests or blue victories. However, as we have said before, Democrats should not expect a blue wave to save them; they need to put in the time, money, and work needed to capitalize on a wave if it develops, and be prepared for one not forming. The map for the Senate races this year is certainly not in the Democrats favor as members who were elected in a year with President Barack Obama at the top of the ticket are now up for reelection, many in states won by President Donald Trump two years ago. Indeed, our PAAR model points to the expected outcome being Democrats losing 6 seats and gaining 1 seat this November, leaving the Senate partisan makeup at 56-44 in favor of the Republicans.

Again this is not a prediction, but the baseline expectation if candidates from both parties run true to form.

Trump State Democrats

While many talking about the House races are focusing on Republican incumbents in Clinton-won districts, those discussing the state of the race for the Senate side are focusing on Democratic incumbents in Trump states. One of the best examples of this is Senator Heidi Heitkamp, who is in for a rough bid for reelection when you consider she only won her first term by less than 3,000 votes, and Trump won the state in 2016 by a whopping 36 percentage points. The PAAR model notices this struggle, and has a generic Republican winning North Dakota in November, 64% to 33% over a generic Democrat. Senator Heitkamp is anything but a generic Democrat and we would expect her to perform well above this baseline. If this does in fact happen she, and her campaign, will deserve immense credit for performing above what a typical Democrat should.

Based on the PAAR projections, five other Democratic seats are set to lose this November: West Virginia, Montana, Indiana, Ohio, and Missouri. All of these, not surprisingly, as states that Trump won handily. West Virginia and Montana, similar to North Dakota, have the Democratic candidate losing by double digits according to PAAR. This is unsurprising, considering Trump’s success in both states in 2016. Indiana, although to a smaller degree, is a similar story.

Ohio and Missouri, however, are a bit different. PAAR’s calculations have a Republican winning there by 2.2 points or less, setting up each to be a close race. The fact that there are Democratic senators in North Dakota, West Virginia, Montana, and Indiana shows that candidates can make a difference, and that there are exceptions to any rule or model.

There are another four Democratic seats in Trump states that PAAR projected to remain the same after November: Pennsylvania, Wisconsin, Florida, and Michigan. The most likely outcome of the first three should be similar to Ohio and Missouri in that they will be very close races; Senator Debbie Stabenow has some room to breathe, according to PAAR, in Michigan with her advantage of nearly 10 points.

Other Close Races

Three other Senate races have a spread less than 5 percentage points; all three are in states that are purple with recent tendencies to go blue. The race will be close for Maine if it narrows down to two candidates; however, with a handful of candidates from the Democratic, Republican, Libertarian, and Independent parties running against the incumbent Senator Angus King, the result may look a lot like his 2012 election, where he received 51% of the vote.

Virginia was certainly a hot topic during the 2016 presidential race, and many questioned up until the final call whether the state would stay blue or go back to red, a position that was more familiar to the commonwealth before 2008. And while many have said that a combination of Virginia’s instability in going blue in 2016 and Tim Kaine’s position in the campaign as Clinton’s VP pick might make his reelection campaign this year become more difficult, Republicans have struggled to choose a formidable challenger for him in November. Nonetheless, PAAR shows a Virginia Democrat winning by about 2 percentage points, 50.5% to 48.1%. Given Senator Kaine’s opponent this year, 48% seems incredibly high, and if we were to make a prediction Corey Stewart is likely to fall well short of this mark.

And finally, in New Mexico, where Senator Martin Heinrich is running for reelection, a close race is not totally out of the norm; Heinrich won in 2012 by about 6 points. However, the expectation is for the Democrats to retain the seat.

Seats Most Likely to Flip to Blue

Here is where our tale of caution truly comes into play: based off of PAAR’s projections, only one seat is predicted to be close to moving in the Democrats’ direction. In Nevada, Senator Dean Heller represents the most vulnerable Republican incumbent and while he holds a slightly higher expectation, it is just 0.3 percentage points. Hillary Clinton won the Silver State by 2 points (48% to 46%) in 2016, and Senator Cortez-Masto won with 47% of the vote. Democrats have a strong candidate in Jacky Rosen, and many expect this seat to flip. However, if this does happen, it will be because of a campaign that exceeded the expected results.

Other Seats

Once we move past Nevada, Democrats should not technically be competitive. Again, this is not to say they won’t be, but they shouldn’t be. Democrats are looking to three states as their chance to take over the majority (assuming they hold all seats currently hold) and by rights, none of these seats should be competitive this year. Arizona has been trending in the Democrat’s direction over the past eight years, but the expected results is still well below 50% for the Democratic candidate.

Texas and Tennessee are even further behind, and the fact that these seats are even remotely considered to be competitive speaks volumes about the candidates in these two races, as well as the overall political landscape we are currently in.

Implications

By all accounts, Democrats are likely to over-perform expectations this year. The combination of the Trump Administration, along with a set of exceptional candidates certainly are putting the Democrats in a much strong position than they technically should be. We have little doubt that Democratic incumbents will over-perform and many will hold seats that they should not technically be competitive in. The big question for November is how much of a boost do the challengers get and is it enough to propel Democrats into the majority.

This is why we hold elections, and why campaigns can make a difference. It will be certainly interesting to see how much of a difference the campaigns in AZ, TX, and TN can make, and how far above PAAR they end up.

Appendix