Skip to main content
ABC News
Senate 2016: The Democrats Strike Back

It may come as a shock given all the attention being paid to the presidential race this year, but the president isn’t all-powerful. In fact, the U.S. Congress is supposed to be a coequal branch of the federal government. From voting on important legislation to confirming Cabinet appointees and federal judges, the Senate matters.

Right now, Republicans hold 54 seats to the Democrats’ 46 (including two independents who caucus with the Democrats). The Democrats have a favorable map in 2016: Of the 34 seats up for grabs, 24 are held by Republicans. Democrats need to net four seats to win control of the Senate if Hillary Clinton wins the presidency and five seats if Donald Trump wins. We’ll launch our official Senate projections later this year, but it’s not too early to take a more informal tour through the races. The combination of the polling, the political lean of the states being contested and the candidates running suggests a close race for control of the chamber.

The biggest factor working in the Democrats’ favor is fairly simple: Senate election results are increasingly tied to the presidential vote in each state:


And of the eight seats most likely to change hands in 2016, six are held by Republicans in states that President Obama won twice, and one is held by a Democrat in a state that former President George W. Bush won twice. If Trump does better in the presidential race than expected, Democratic gains could be kept to a minimum, but the field is tilted in their favor.

For now, Democrats are most likely going to win three to four seats. How did I get to those numbers? I summed up Democrats’ chances in all 34 seats.

For each seat, I estimated the chance that the Democrat or Republican would win. That percentage chance comes from a simple model based on an analysis of Senate races since 2006. For seats in states where polling has been done this year, the model includes variables for an average of all polls in the state since January,1 how Democratic- or Republican-leaning the state was in the previous presidential election (2012 for this year), whether the incumbent is running for re-election, and the average generic ballot result2 nationwide since January to give us an idea of the national environment. Where there are no polls, the model is the same except without a polling variable. This makes the model less certain about the outcome but still gives us a rough guide.


Again, this is a basic model that we’re just using as a yardstick; we’ll have our official forecasts in a few months. Let’s dive into the eight seats most likely to flip, according to our model (for more context, I’m also listing race ratings from the Cook Political Report and the Rothenberg & Gonzales Political Report).


Candidates: Republican Sen. John McCain will likely face Democrat Ann Kirkpatrick
Cook rating: Likely Republican
Rothenberg/Gonzales rating: Republican favored
Chance of Democratic victory: 25 percent

This is the type of race that can sneak up on you. Longtime Republican senator McCain fended off a conservative primary challenge in 2010, and he has never lost a race in Arizona. That should hearten Republicans. McCain, though, has never really confronted a strong Democratic opponent. That’s going to change this year.

Before the general election, though, McCain has to deal with issues in his own party — he hasn’t gotten rid of his problem on the right. Although McCain will probably get past his main primary challenger, former state Sen. Kelli Ward, the fact that we’re even talking about his primary speaks to McCain’s vulnerability. He’ll have to put up with Ward through the Aug. 30 primary vote.

If he wins, McCain will face off against Kirkpatrick, a Democratic representative in the House. Arizona is a red state, and McCain is an incumbent, so he’ll enter the general election with some advantages. But Kirkpatrick is a moderate and has a history of winning in Republican-leaning districts. McCain leads in the polls, but only barely, with a 2 percentage point edge on average.

And then there’s Trump, whose presence at the top of the GOP ticket should give us pause. Clinton leads in the average poll out of Arizona at the moment, but there are still a lot of undecided voters who likely lean Republican, and Trump led in the one Arizona poll taken since he wrapped up the nomination. There are reasons to think Trump might underperform in Arizona, namely the state’s large Latino population. In 2012, 23 percent of the citizen population was Latino, though just 17 percent of Arizona voters were. We’ve already seen reports of Latinos registering in high numbers in California this year, so it isn’t far-fetched to imagine the same thing happening next door in Arizona. If it does, McCain needs to watch out.


Candidates: The Democratic and Republican candidates are yet to be decided
Cook rating: Tossup
Rothenberg/Gonzales rating: Pure tossup
Chance of Democratic victory: 64 percent

The question dominating the Florida Senate race is whether Marco Rubio — who planned not to run for re-election to pursue the White House (that didn’t work out) — decides to run after all. If he does, then he’d probably be considered at least a slight favorite against the most likely Democratic nominee, Rep. Patrick Murphy. Rubio led Murphy by 8 percentage points in a recent Associated Industries of Florida survey. Even Rubio, however, wouldn’t be a sure thing for Republicans; Obama carried Florida in 2008 and 2012, and Clinton leads Trump in most of the state polling.

If Rubio does not run, then the Republican side of the aisle looks like an absolute mess. The main contenders are Carlos Beruff, Rep. Ron DeSantis, Rep. David Jolly and Lt. Gov. Carlos Lopez-Cantera. The limited primary polling we have suggests that no one has a clear edge, nor do any of them seem to have any super compelling claims on electability in the general election. We’ll know more after the Republican primary Aug. 30.

The Democratic side isn’t that much clearer. Arch liberal Rep. Alan Grayson (who endorsed Bernie Sanders) is facing off against moderate Murphy (who endorsed Clinton). Murphy has more money and more backing from elected officials, but you wouldn’t know that from the limited primary polling we have so far; the average shows Murphy up by just 2 percentage points. Still, Murphy is probably a slight favorite.

Whoever the Democratic nominee is, I’d expect Florida’s Senate general election race to mirror the presidential race. Murphy and Grayson are leading in more polls than not, but their margins are small. The one consistency in the polling: There are a lot of undecideds, who will likely break the way they do in the presidential race.


Candidates: Republican Sen. Mark Kirk is facing Democrat Tammy Duckworth
Cook rating: Tossup
Rothenberg/Gonzales rating: Lean Democrat
Chance of Democratic victory: 77 percent

This is one of the two most likely seats to flip this year. Kirk was first elected to the Senate in the 2010 Republican wave. He has the profile of a Republican who can win in Illinois: moderate and from the lakeshore. Kirk’s problem is that Illinois is a solidly blue state, especially in a presidential year. No Democratic presidential candidate has won the state by less than 10 percentage points from 1992 through 2012.

Democrats also benefit from having a strong candidate in Duckworth, a House member and Iraq War veteran. Duckworth is about as moderate as Kirk, according to their voting records, and had more cash on hand than Kirk through the first quarter of this year,3 despite Kirk’s strong fundraising. She’s also less likely to fly off the handle — Kirk has been known to make controversial statements.

Don’t sell Kirk short, though: He’s won a number of tough races and is a fighter (witness his comeback from a stroke). But at the end of the day, he’s fighting Illinois’s Democratic lean more than Duckworth. He’s trailed in the few polls that have been conducted, including being down 3 percentage points in a survey released by his own campaign. If Kirk holds on in Illinois, it’s probably a sign that Republicans are keeping the Senate.


Candidates: Democrat Catherine Cortez Masto will likely face Republican Joe Heck
Cook rating: Tossup
Rothenberg/Gonzales rating: Pure tossup
Chance of Democratic victory: 51 percent

How sweet would it be for Republicans if they could win the Democratic minority leader’s seat? Six years ago, many political observers thought just that would happen, but in 2010 then-Majority Leader Harry Reid won re-election by 6 percentage points, despite polls showing him trailing. The unpopular Reid, who is retiring, won because Republicans picked a disaster of a candidate in Sharron Angle.

Angle is running again but will almost certainly be defeated in the Republican primary by Heck, a member of the U.S. House. Heck is about as strong a nominee as Republicans could hope for. He represents a swing district just outside of Las Vegas, the 3rd.

Heck’s Democratic opponent is likely to be Masto, a former state attorney general. She’s Reid’s hand-picked successor and probably one of the best options the Democrats had. Masto, who has Mexican heritage, will be relying on Nevada’s growing Hispanic population, which has taken the state from a Republican lean to fairly easy Obama victories in 2008 and 2012. Trump’s position at the top of the ticket might hurt Heck; we’ve already seen Nevada’s other Republican senator, Dean Heller, say he might not vote for Trump. Heck has led in the very few surveys of the race released so far, though polling in Nevada has tended to underestimate Democratic candidates in past elections.

Heller proved in 2012 that a Republican can win a Senate race in Nevada even as a Democrat carries the state at the presidential level. We’ll have to wait and see whether Heck can follow in his path.

New Hampshire

Candidates: Republican Sen. Kelly Ayotte will likely face Democrat Maggie Hassan
Cook rating: Tossup
Rothenberg/Gonzales rating: Tossup/Tilt Republican
Chance of Democratic victory: 41 percent

New Hampshire could be the rare swing state where, even if Trump is struggling, he doesn’t hurt down-ticket Republicans. Trump cruised in the Granite State’s Republican primary, while Clinton fell flat there against Sanders. It’s also one of only four states where non-Hispanic whites make up more than 90 percent of the population. So Trump’s lack of appeal with minority voters won’t hurt the Republican brand nearly as much in New Hampshire as in, say, Florida or Nevada.

Ayotte won easily in the Republican wave of 2010, and polls show her to be relatively popular in the state. She’s less conservative than most of her Republican colleagues in the Senate, and she has tried to keep her distance from Trump by “support[ing] the nominee” but not endorsing him. Ayotte also showed she can win a tough race when she fended off a conservative challenge from Ovide Lamontagne in the 2010 Republican Senate primary. She’s also led in most of the polls (though by small margins) against the probable Democratic nominee and current New Hampshire governor, Hassan.

Hassan, for her part, is quite popular, with a 52 percent approval rating versus a 33 percent disapproval rating in a recent University of New Hampshire survey. She survived the Republican wave in 2014 by a small but comfortable 5 percentage point margin. But making the jump from popular governor to senator isn’t always easy against a good opponent.

Still, I shouldn’t oversell Ayotte’s advantage over Hassan. Although the early polling shows Ayotte doing a few percentage points better than Trump, who is trailing Clinton in the state, a rising tide tends to lift all boats. Just ask New Hampshire Sen. Jeanne Shaheen, who beat John Sununu in 2008 by a little more than 6 percentage points when Obama beat McCain by a little less than 10 percentage points in the state. If Clinton wins New Hampshire by more than a few points, she’ll likely take Hassan with her.


Candidates: Republican Sen. Rob Portman is facing Democrat Ted Strickland
Cook rating: Tossup
Rothenberg/Gonzales rating: Tossup/Tilt Republican
Chance of Democratic victory: 46 percent

I guess it shouldn’t be too surprising that control over a branch of the federal government may come down to the state of Ohio. Portman, like Ayotte in New Hampshire, cruised into office in 2010 as part of the Republican wave. He’s been known in the state for some time as a former representative from the Cincinnati area and a former member of President George W. Bush’s administration. Portman comes from the moderate business wing of the party, the “Chamber of Commerce wing,” and it shows in his fundraising. He has more cash on hand than any other Republican member of Congress running in 2016.

Portman, though, won’t have as easy a ride as he had in 2010. He’s facing off against Strickland, a former governor who almost won re-election in 2010 despite super strong headwinds and a strong challenger, John Kasich. Strickland lost by just 2 percentage points as Portman was winning his Senate seat by 17 points.

The Ohio Senate race is in some ways the opposite of the presidential race: Both Trump and Clinton have historically bad net favorability ratings; Strickland and Portman both have positive net favorability ratings, 6 and 13 percentage points in the most recent Quinnipiac University poll, respectively.

So far, Portman has held a 0.3 percentage point lead in the average poll, but Strickland has also led in a number of surveys. The fact that Portman holds even a tiny edge when Clinton is leading in the vast majority of Ohio polls suggests that he could win even if Clinton carries the state. But as in New Hampshire with Ayotte, Portman probably will be in deep trouble if Clinton wins the state by more than 5 percentage points.

One thing to watch for is the regional bases for each candidate. Portman may do better in the Cincinnati area than Trump, while Strickland may do better in southern Ohio than Clinton. It’ll be interesting to see if these bases hold in a presidential year.


Candidates: Republican Sen. Pat Toomey is facing Democrat Katie McGinty
Cook rating: Tossup
Rothenberg/Gonzales rating: Tossup/Tilt Republican
Chance of Democratic victory: 18 percent (or 46 percent)

What’s the deal with the two probabilities for this seat? If you look at all polls taken in the first half of this year, Democrats have an 18 percent chance of winning. If you look only at the one poll taken since McGinty won the Democratic nomination in late April, they have a 46 percent chance. In a victory for the party machine, McGinty was able to sneak past 2010 Democratic nominee Joe Sestak. (She was endorsed by Obama.) Hard-fought primaries, as Clinton and Sanders have demonstrated in the presidential race, can keep a candidate’s numbers down.

That Toomey has led in every single poll is likely a testament to his skill as a politician. He’s one of the most conservative members of the U.S. Senate and represents a state that hasn’t voted for the Republican in a presidential election since George H.W. Bush won there in 1988. Toomey had a +15 percentage point net approval rating in the most recent Quinnipiac University poll. He also has a cash-on-hand advantage over McGinty of more than $8 million.

For McGinty to win, the formula is simple. First, she must hold on to her post-nomination polling bounce. Second, Clinton has to keep her advantage over Trump in the presidential election (she hasn’t trailed in a single statewide poll of Pennsylvania this year), and her voters have to also pull the lever for McGinty. If Pennsylvania votes for the Republican presidential candidate in larger numbers in 2016 than it has in the past, as FiveThirtyEight contributor David Wasserman argues it might, that would be good news for Toomey. Keep in mind, though, that Toomey won by only 2 percentage points in the 2010 Republican wave — that tells you winning re-election won’t be easy for him.


Candidates: Republican Sen. Ron Johnson is facing Democrat Russ Feingold
Cook rating: Tossup
Rothenberg/Gonzales rating: Pure tossup
Chance of Democratic victory: 85 percent

I was surprised that our simple model made Johnson such a long shot. Then I looked at the polling data. Johnson has trailed his Democratic opponent, former Sen. Feingold, in every single poll taken in this race. The closest Johnson has come to Feingold this year is 3 percentage points, in a Marquette University survey from March. All other polls have had Johnson down by more; the most recent St. Norbert College poll had Feingold up 10 percentage points.

This race is a rematch. Feingold held the seat before Johnson defeated him in 2010. Despite being out of the game for six years, Feingold has more cash on hand than Johnson, which is notable because incumbents typically dominate the money race. Feingold probably fits the state’s ideology better than Johnson, who is very conservative. Wisconsin hasn’t voted for a Republican for president since 1984, and Obama twice carried the state fairly easily.

Johnson’s best hope is to keep the race local. More Wisconsinites approve of the job he is doing than disapprove. And Trump looks, at this point, to be a big drag on the ticket: He has trailed Clinton by 9 percentage points or more in every presidential poll taken in the state so far. Trump also lost the Wisconsin primary by 13 percentage points — his last defeat of the primary season.

It would be silly to dismiss Johnson’s chances out of hand. He’s already beaten Feingold once. But the data suggests that Feingold is the favorite.

If there’s a wave

Here’s an additional seat each party could gain if almost everything goes right (or wrong, depending on your point of view).

Colorado: Democratic Sen. Michael Bennet has to be thanking his lucky stars. Two years after Democratic Sen. Mark Udall lost a re-election bid, Republicans have struggled to field a respectable candidate to take Bennet on. Republican state Rep. Jon Keyser nearly wasn’t even allowed on the primary ballot. Bennet is an incumbent with plenty of cash on hand, so he’s clearly the favorite. But without any public polling this year and because Colorado is a swing state, I’m not confident in saying this seat is entirely safe. A model based purely on the fundamentals gives Bennet an 86 percent chance of re-election.

North Carolina: Republican Sen. Richard Burr has led in every poll against Democrat Deborah Ross, though he’s never led by more than 10 percentage points this year or gotten to 50 percent in a poll. His average lead since Ross captured the Democratic nomination has been just more than 3 percentage points. The model gives Burr an 80 percent chance of being re-elected. Why? He’s an incumbent who is leading in the polls, and North Carolina leans slightly to the right on the presidential level. Ross’s best chance is if Clinton wins North Carolina. Polls released since Trump became the presumptive nominee have Clinton a few percentage points behind.

Listen to the latest episode of the FiveThirtyEight politics podcast.


The rest

Twenty-four other states are holding Senate elections this year. None of those seats seem particularly likely to change hands, but given that there are so many, it wouldn’t be surprising if one or two became competitive. States without an incumbent running, such as Indiana (where Republican Rep. Todd Young has led by a wide margin in the polls) and Louisiana with its jungle primary, or where Democrats have won federal races over the past four years, such as Missouri (where Republican Sen. Roy Blunt leads over Democratic Secretary of State Jason Kander) and Iowa (where longtime Republican Sen. Chuck Grassley awaits the winner of the Democratic primary), could potentially yield a surprise.

For now, control of the Senate is likely to be decided in the states discussed in-depth above, and it’ll probably be close.


  1. When it is unclear who the candidate will be for one or both parties, I include an average of all the polls for all possible nominees.

  2. Democrats currently hold a 2 to 3 percentage point lead on this measure.

  3. The most recent data available.

Harry Enten was a senior political writer and analyst for FiveThirtyEight.