2018 Senate highlight race: Arizona and Nevada


Incumbent: Jeff Flake (R)

Last party margin: 49.2-46.2

Senator Jeff Flake is one of the most endangered Republican incumbents in the 2018 cycle. Ever since his vote against Manchin-Toomey amendment (expanding universal background check on all gun purchases), his approval rating has been trending low. A poll conducted by Public Policy Polling in 2013 has showed Flake having one of the worst approval ratings in the nation–with just 32% approving and a whopping 51% disapproving. A Morning Consult poll also showed Flake to be one of the least popular senators.

Also conducted by Public Policy, a poll this May showed Flake up only 2 points with Kyrsten Sinema, a prominent Arizona Democrat. Although Sinema is a member of the Blue Dogs coalition, her voting record is nowhere near Jim Matheson, a prominent Utah Democrat and potential candidate to run for the open seat in 2018. According to Ballotpedia, her voting record is moderately left of center and largely voted along the party line. Combining Flake’s one-of-lowest-approval in the nation and his lackluster polling number, we rate this race Leans R pending Trump’s approval numbers for any further rating.



Incumbent: Dean Heller (R)

Last margin: 45.9-44.7

Unlike Flake and McCaskill, Heller’s approval rating is more modest. In the same poll that showed Jeff Flake down 19 points in approval rating, it showed Heller up three points 44-41.  However, in his 2012 campaign, Heller promised to be bipartisan and independent. Voting records have not shown that. Heller’s records indicate his stance is much to the right of Shelley Moore Capito, who is seen as a “moderate conservative” in her party, but Morning Consult approval poll shows Heller up 26 points with 48 percent approving and 22 percent disapproving. (Morning Consult and PPP are different polls, and they yield vastly different results.)

I used to comment this race as strongly leaning Republican, as I thought Clinton winning the presidential election was all but assured, pushing a R-wave considering her less-than-modest approval numbers, but Trump winning completely changed Senate calculus for 2018. This race starts as a Tossup. 

Canadian Voting Index–An Introduction


Partisan Voting Index is a concept developed by American political scientist Charlie Cook. While Mr. Cook would definitely not be happy me stealing his concept of partisan voting index to Canada, it’s not going to be effective in a Canadian context anyway. Canada has three major political parties–the Conservatives (or the Tories), the Liberals, and the New Democrats (or the Socialists)–PVI fails to work in a three-party context. Therefore, an alternative method to measure political party support needs to be developed. Therefore, I am introducing the Canadian Voting Index, basing on a slight variation of Mr. Cook’s legendary work.


Canadian Voting Index measures each political party’s support level in a given geographic area, rather than a positive-negative number that measures two party’s support level in an area. Essentially, this is altering from Cook’s methodology, just converting a two-party index into a three-party index. Method of calculating CVI is identical to that of PVI, except in a three-party context.

CVI is calculated by:

\frac{\%_{r1}+\%_{r2}-\%_{n1}-\%_{n2}}{2} (1)

Where \%_{r1} denotes a party’s popular vote in a riding in election 1, \%_{r2} denotes that party’s popular vote in that riding in election 2, \%_{n1} denots that party’s nationwide popular vote in election 1, and \%_{n2} denotes that party’s nationwide popular vote in election 2.

For example, if political party B has 35% of votes in riding A in election 1, 33% of votes in riding A in election 2, 25% of national popular vote in election 1, and 27% of national popular vote in election 2, its CVI is:



Canadian Voting Index could be used to project a political party’s performance in a riding in an upcoming election. For instance, we can apply CVI number to a nationwide poll to project a party’s expected vote share in a riding. Note that CVI is not a comparison between political parties, as that comparison would have no practical application aside from building a statistical table.

Canadian Voting Index could also be used calculate a region’s electoral shift from one election to another, by comparing CVI from different elections.


Creating a CVI table would involve collecting 2011 and 2015 election results for all 338 electoral districts across Canada. 2011 election result needs to be in redistributed format to make such result most accurate. A table based on equation 1 then needs to be computed to output final CVI data for each riding.

Preliminary Data

A CVI table at the provincial level has already been completed. Preliminary findings include huge swings between one election to another in both the provinces and the ridings. For example, Gatineau swung 40 points toward the Liberals from 2011 to 2015. Based on the preliminary data, it might be difficult for a CVI-based model to accurately predict elections in Canada due to the existence of such swings.



Province NDP Cons. Lib.
British Columbia 4.05 2 -4.9
Alberta -10.95 27.4 -12.25
Sasketchewan 3.55 16.65 -12.95
Manitoba -5.35 9.65 1.4
Ontario -4.05 3.95 5.85
Quebec 9 -19.15 -4.25
New Brunswick -1.1 -1.15 7.9
Nova Scotia -1.8 -8.45 16.2
Prince Ed. Il. -9.45 -5.5 20.45
Newfoundland 1.65 -16.4 22
Yukon T. -8.2 -6.85 14.1
Northwest T. 13.15 -10.7 4.15
Nunavut T. -2.2 1.6 8.75

Table 1: Preliminary Data

2018 Senate Ratings

Safe D Likely D Leans D Tilts D Pure Tossup Tilts R Leans R Likely R Safe R
Feinstein (CA) Stabenow (MI) Nelson (FL) Manchin (WV) Tester (MT) McCaskill (MO) Cruz (TX) Wicker (MS)
Murphy (CT) Menendez (NJ) Casey (PA) Heller (NV) Flake (AZ) UT Open (Hatch) Barrasso (WY)
Carper (DE) Kaine (VA) Baldwin (WI) Heitkamp (ND) Fischer (NE) Corker (TN)
Hirono (HI) Brown (OH) Donnelly (IN)
King (Ind-ME) Tester (MT)
Cardin (MD)
Warren (MA)
Klobuchar (MN)
Heinrich (NM)
Gillibrand (NY)
Whitehouse (RI)
Sanders (Ind-VT)
Cantwell (WA)

Democrats are favorable in 2016, numerous pundits predict. Recent polling suggest the Democratic candidate is on track of winning in WisconsinIndiana (although there’s only one poll)New HampshirePennsylvania, (either NH or PA has an outlier poll saying large-margin Republican lead) and Democrats have a very good chance at the quasi-abandoned Illinois. At R+0.2, Nevada is considered tossup by most pundits. Also North Carolina (R+2.5), Missouri (R+4.8), Florida (R+5.7), Ohio (R+7.5) Arizona (R+8.0), and even Georgia (R+8.0) could potentially flip Democratic if this election becomes a wave for Hillary Clinton.


If the Democrats were to win all Lean D states, picking up one tossup states (NV and NC), and one Lean R state, Democrats would have a 52-48 majority in the upper chamber. The lead is by no means large; in 2008, the Democrats had a filibuster-proof majority of 60 seats. Now they have 46, losing 14 seats between their supermajority and now.


Enough discussion about 2016 Senate. Let’s move onto our topic–2018 Senate. Republican and Democratic strategists alike point out the math is very unfavorable for the Democrats. 23 Democrats, 8 Republicans and 2 independents are up for re-election (or retirement) in 2018. Like this year with the Republicans, the 2018 map for the Democrats is incredibly bad.


Info por favor @ en.wikipedia CC BY-SA 3.0

Info por favor @ en.wikipedia CC BY-SA 3.0


Without wasting words, let us start by doing state-by-state analysis.



Incumbent: Jeff Flake (R)

Last party margin: 49.2-46.2


Senator Jeff Flake is one of the most endangered Republican incumbents in the 2018 cycle. Ever since his vote against Manchin-Toomey amendment (expanding universal background check on all gun purchases), his approval rating has been trending low. A poll conducted by Public Policy Polling in 2013 has showed Flake having one of the worst approval ratings in the nation–with just 32% approving and a whopping 51% disapproving. A Morning Consult poll also showed Flake to be one of the least popular senators.


Also conducted by Public Policy, a poll this May showed Flake up only 2 points with Kyrsten Sinema, a prominent Arizona Democrat. Although Sinema is a member of the Blue Dogs coalition, her voting record is nowhere near Jim Matheson, a prominent Utah Democrat and potential candidate to run for the open seat in 2018. According to Ballotpedia, her voting record is moderately left of center and largely voted along the party line. Combining Flake’s one-of-lowest-approval in the nation and his lackluster polling number, Flake would find himself in the tossup column, especially after Trump’s election to the presidency

Rating: Leans R




Incumbent: Dianne Feinstein (D)

Last party margin: 62.5 – 37.5


There is not much to talk about. Safe D no matter Feinstein retires or not. Potential Democratic candidates include Reps. Loretta Sanchez, Linda Sanchez, Judy Chu, Xavier Becerra, Norma Torres, California State Sen. Kevin de Leon, or California Secretary of State Alex Padilla.

Rating: Safe D.




Incumbent: Chris Murphy (D)

Last party margin: 55.1 – 43.3


Still not much to talk about. Connecticut has Democratic trifecta control even at this time. Murphy won the open seat by 11.8%, and this time he would enter the race as incumbent, adding momentum to his side.

Rating: Safe D.




Incumbent: Tom Carper (D)

Last party margin: 66.4 – 29.0


Same as Connecticut except the state is even more Democratic.
Rating: Safe D.




Incumbent: Bill Nelson (D)

Last party margin: 55.2% – 42.2%


If Nelson, 76, decides to run for another term, I see his prospects of winning re-election to be very high, as only his first Senate race was remotely competitive. However, if Nelson decides against running for re-election and decide to retire instead, it would be very difficult for another Democrat to carry the state for them. If Murphy loses 2016, he could potentially run again in 2018 against potential candidates Carlos Lopez-Cantera, Pam Bondi and David Jolly (who is likely to lose his congressional seat this year to former govenor Charlie Crist). Democratic potential candidates include Ex. Gov. Charlie Crist, but he lost two statewide races and if president Clinton has a low approval he is likely running for re-election to his House seat if he gets it this year. Under president Trump it would be a different story, with potential of Democrats sweeping every statewide office.


Rating: Leans D

September 7 update: a Public Policy poll showed Nelson only up by 4 points in a hypothetical matchup with Gov. Rick Scott. I am still standing with my Likely D prediction but it could change.




Incumbent: Mazie Hirono (D)

Last party margin: 62.6 – 37.4


If Hirono could beat a popular former governor by 25 points, she could beat another Republican candidate.

Rating: Safe D




Incumbent: Joe Donnelly (D)


Indiana is certainly one of the most competitive Senate races in 2018. If Bayh is elected in 2016, which would likely happen, Donnelly is more likely to get unseated. Yes, I know Joe Donnelly is one of the most conservative Democrats out there in the Senate (slightly to the left of Manchin on economic issues, and almost the same position as Manchin’s on social issues), but still doesn’t change the fact Hoosiers would probably not be satisfied with two Democratic Senators. In 2012, Indiana was a safely rated as a hold for the Republicans before Lugar’s surprising primary defeat. Mourdock, even by Indiana standards, is considered too conservative to be electable. He never lead the poll more than 5%, and was hurt further by the pregnancy from rape controversy.


Donnelly was incredibly lucky running against Mourdock, but this time around he would face a much more powerful opponent. Republican challengers are plenty. Susan Brooks has been touted as a potential candidate, as well as Greg Zoeller, Curtis Hall, Luke Messer. Although the national climate could be very favorable to the Democrats, Donnelly will no doubt find himself in a tight reelection battle.

Rating: Tossup




Incumbent: Angus King (I)

Last party margin: 52.9-30.7-13.1


Paul LePage is no Charlie Baker or Earl Ray Tomblin. He isn’t the most popular governor in an otherwise Democratic leaning state. His approval rating is also one of the worst in the nation, according to a UNH/Press Herald poll. The same poll showed him down 63-29 with King.


But Angus King, 74 on election day, could obviously decide to retire. Failed gubernatorial candidate Mike Michaud could be replacing King as the choice of moderates and progressives. But he lost to LePage, who knows. Maine is relatively nonpartisan, just like North Dakota, which would be covered later in this essay.

Rating: Safe I




Incumbent: Ben Cardin (D)

Last party margin: 55.4-26.7-16.6


Cardin retires or not, Maryland remains solidly D unless Republican candidate is Larry Hogan. Potential Democratic candidates include Donna House and John Delaney.

Rating: Safe D




Incumbent: Elizabeth Warren (D)

Last party margin: 53.7-46.2


Replace Hogan with Baker is Massachusetts.

Rating: Safe D



Incumbent: Debbie Stabenow (D)

Last party margin: 58.8-38.0


Michigan was Tennessee of 2014, enough said.

Rating: Likely D




Incumbent: Amy Klobuchar (D)

Last party margin: 65.2-30.5


Add Minnesota to the list of Connecticut, Delaware and Hawaii.

Rating: Safe D




Incumbent: Roger Wicker (R)

Last party margin: 57.2 – 40.6
Democrats in Mississippi have seem already given up. There is no credible candidate other than Jim Hood (Mississippi Atty Gen). Even if Hood runs for either Governor’s Mansion or Washington, the Republican candidate would still be favored,.

Rating: Safe R.




Incumbent: Claire McCaskill (D)

Last party margin: 54.7-39.2

Now there we have it, the most competitive (or leaning takeover) race in the nation. McCaskill was well on her route of losing re-election in 2012 until Todd Akin said this:

If it’s a legitimate rape, the female body has ways to try to shut that whole thing down.

Apparently this is a statement to appease Tea Party voters, strengthening their support for him, but why the hell in the world did Todd Akin want to strenghten his Tea Party support while turning all moderate votes to Claire McCaskill?  Tea Party voters only have one choice–him, but moderate voters could split between him and McCaskill, and in the election they sided with McCaskill, beating Akin by a resounding 15-point margin.


Ok. Enough of 2012, moving onto 2018. McCaskill is considered one of the most vulnurable incumbents because of her poor approval rating–merely 39% approving while 46% voicing their disapproval, according to a Public Policy Poll. This figure is only marginally better than Jeff Flake’s approval rating, which is one of the worst in the nation. Having these approval ratings, combined with a strong Republican opponent (Ann Wagner), Claire McCaskill would be the Republicans’ easiest target for 2018. Theoritically, McCaskill and Flake are in the same position because of their less-than-ideal approval ratings, but McCaskill would face a less ideal situation because of her Democratic party affiliation in a red state. Changing demographics in Missouri are propelling Democratic voters out of the state, and St. Louis County region became a solidly red region from a swing region. All these factors are suggesting Missouri is on its path to become solidly red.


Less than ideal approval rating, strong Republican opponent, and changing demographics all suggest Claire McCaskill would face an uphill battle for reelection. Since Missouri has statewide races this November, situations could change after that election, but McCaskill is currently on her track to defeated reelection. If McCaskill is primaried and she does lose, this race would be rated otherwise. For example, if Jason Kander is the Democratic nominee instead of McCaskill, this race would be Tilt D at most.

Rating: Tossup/Tilt R




Incumbent: Jon Tester (D)

Last margin: 48.6-44.6

Sitting representative Danny Rehburg failed to unseat Tester in 2012. President Obama only scored 41% of votes in the state, indicating Tester ran almost 7 points ahead of Obama in his state. And Rehburg was a reasonably popular representative, representing the state for six terms.  Sure, it was 2012, but in Montana Obama was not positioned favorably and Tester is reasonably more favorable than Obama in the state. In 2018, to defeat Tester would require Hillary Clinton be really unfavorable if she becomes president, but if Donald Trump, whose favorability is almost guaranteed to be low, becomes president, Jon Tester would virtually be guaranteed to hold the seat for another six years. Some people see this race as very competitive and current trends show that to be true, but Tester would start with a major advantage.


Rating: Lean D



Incumbent: Deb Fischer (R)

Last margin: 57.8-42.2

Fischer could find herself with some nominal opposition with Rep. Brad Ashford, who was just defeated for reelection as Trump carried the district by 2 points, but that’s not enough to bring a serious challenge to her.

Rating: Likely R



Incumbent: Dean Heller (R)

Last margin: 45.9-44.7


Unlike Flake and McCaskill, Heller’s approval rating is more modest. In the same poll that showed Jeff Flake down 19 points in approval rating, it showed Heller up three points 44-41.  However, in his 2012 campaign, Heller promised to be bipartisan and independent. Voting records have not shown that. Heller’s records indicate his stance is much to the right of Shelley Moore Capito, who is seen as a “moderate conservative” in her party, but Morning Consult approval poll shows Heller up 26 points with 48 percent approving and 22 percent disapproving. (Morning Consult and PPP are different polls, and they yield vastly different results.)


In comparison (to Jeff Flake and Claire McCaskill), these approval ratings are not bad at all.  If there is no significant demographic shift (such as influx of Hispanics or other strong D-leaning population), Heller should be pretty safe here.


The above texts were written pre-election. With Trump’s unexpected election to the presidency of the United States, Heller became the only Republican Senator who sits in a state that both voted for Obama twice and Hillary Clinton, Heller’s uncompromising record, if touted by a strong candidate (Jacky Rosen or Ruben Kihuen are among contenders here), can have detrimental effects on him.


Rating: Leans D


New Jersey

Incumbent: Bob Menendez (D)

Last margin: 58.9-39.3

Menendez has been covered up with scandals lately, but those scandals should not be strong enough for him to decide against seeking a third term. If Menendez doesn’t retire, he should be safe. If he does, unless the Democrats run a lackluster candidate, Democrat should be fairly safe as well.


Rating: Likely D


New Mexico

Incumbent: Martin Heinrich (D)

Last margin: 51.0-45.3

Lately, New Mexico has been trending Democratic. Heinrich, 47, is young and ambitious. Although he hasn’t been featured on the national spotlight, according to Morning Counsult, Heinrich is up 25 points in approval rating, making him on the same level as Heller. New Mexico is six points more Democratic in comparison to Nevada Republican (Nevada is D+2, New Mexico is D+4, so there’s a six point difference between Heller and Heinrich). If Heller is likely, NM should be safe for the Democrats if Republicans only run a generic candidate.


But New Mexico’s Republican bench is solid. Susana Martinez and Gary Johnson are both exceedingly qualified candidates to take New Mexico’s senate seat. Johnson’s popularity in New Mexico has been reflected in recent pollings. If he could get 20% in a presidential race, he could win the Senate race with 50-55%. Martinez is very popular as well, taking 57% while Udall (Democratic nominee for Senate) took 55%, so at least 10% of the voters must have split their ticket. A clash between two (and potentially three) popular politicians in NM would be very interesting to watch.

Rating: Safe D


New York

Incumbent: Kirsten Gillibrand (D)

Last margin: 72.2-26.3

See Minnesota.

Rating: Safe D


North Dakota

Incumbent: Heidi Heitkamp (D)

Last margin: 50.3-49.2


North Dakota’s senatorial race would be very interesting to watch. Heidi Heitkamp narrowly defeated one-term representative Rick Berg in the 2012 election. I watched a debate between the two candidates and found out Heitkamp to be a very emotional and appealing speaker. Representing an overall conservative state with a large liberal-conservative shift like North Dakota, Heitkamp’s much more to the ideological center than his predecessor is–Kent Conrad was around 30th most liberal, while Heitkamp is around 50th.  She has an A rating from NRA and consistently ranked among the most bipartisan senators. She is also pretty conservative on social issues, making the Republicans unable to attack her from this point. North Dakota’s general electoral history is incumbent protection, and incumbents usually get reelected with large margins–Kent Conrad and Byron Dorgan all won their elections by wide-margins. But Heitkamp’s popularity is not nearly as high as Conrad and Dorgan’s, so that raises some questions for this North Dakota incumbent protection theory. Entering 2017 we might see some polls done in this state, and they may confirm or reject my theory that Heitkamp is narrowly favored but a strong Republican could beat Heitkamp.


Heitkamp is in a better position than McCaskill mostly because of her approval ratings. Heitkamp, although not Conrad/Dorgan/Hoeven level, is reasonably popular, at least much more so than McCaskill is. That made her to be not in too much of a disadvantage.

Rating: Tossup



Incumbent: Sherrod Brown (D)

Last margin: 50.7 – 44.7


Brown’s opponent Josh Mandel never led the poll in 2012, but that doesn’t mean he doesn’t have any power to unseat Brown. Mandel first started as an underdog (trailing Brown almost 15 points in the polls), but he almost equalized that polling near the end of the race. Brown ultimately prevailed, but in 2018 Ohio’s Republican bench would be even larger. Republicans hold every statewide elective offices and all of them are term-limited. DeWine could try to win his old seat back after an eight-year spell at OAG, and his approval rating is not 2006 level (and 2006 was such a large Democratic wave that made Tennessee a tossup) . Kasich could also be a potential Republican candidate–he ran an impressive presidential campaign although losing, but his publicity in Ohio was more than enough that denied Bernie Sanders any chance of a victory.

But keep in mind Brown is much more popular than all these Republican candidates mentioned. Brown is even more popular than Kasich is, by three percentage points according to PPP. Rob Portman, who has a negative approval rating, is expected to win his race against Strickland by more than 15 points. Brown’s approval rating is almost twenty points higher than Portman’s. he also leads by more than 10 points in a hypothetical matchup against Mandel.


Rating: Lean D



Incumbent: Bob Casey (D)

Last margin: 53.7 – 44.6

Casey faced Tom Smith, an unknown, in the 2012 general election and was posed a significant threat. Name recognition should push all independents and undecideds to the Casey side, but that didn’t happen. This is unusual because Casey’s approval rating is not remotely low–he is at Sherrod Brown level according to PPP. Considering Pennsylvania’s Republican bench, Casey would probably face another Smith in the general election (almost all statewides are held by the Democrats), but if a strong Republican comes out by any chance, Casey would be significantly threatened. If Casey faces another Smith, he is safe, but if he meets someone like Charlie Dent, the race could become competitive.


Rating: Lean D



Rhode Island

Incumbent: Sheldon Whitehouse (D)

Last margin: 65.0 – 35.0

Not much to talk about. Whitehouse is very safe as the last Republican who could even post nominal threat to this seat became a Democrat.

Rating: Safe D



Incumbent: Bob Corker (R)

Last margin: 64.9 – 30.9

Same as Rhode Island except change D to R.

Rating: Safe R



Incumbent: Ted Cruz (R)

Last margin: 56.4 – 40.6

Cruz could face a primary challenge. Actually he is very likely to face a primary challenge. His primary challengers are unlikely to defeat him, though. There is no significant difference between Cruz, Dan Patrick, Rich Perry or name another Republican. Ideologically and name recognition wise they are all the same. Texas has been trending purple, meaning a strong Democrat could make this one a swing. Potential Democratic candidates include the Castro brothers if they can boost Latino turnout significantly. If no strong Democrat (turnout-driving) is running, the Republican candidate should be very safe.

Rating: Likely to Safe R.



Incumbent: Orrin Hatch (R)

Last margin: 65.2 – 30.2

Hatch is very likely to retire. If he retires (or dies), DINO Matheson could have this seat within reach. If Matheson declines, the Republican candidate should be VERY safe, but if he runs then there could be some drama.

Rating: Likely R (Lean R if Matheson runs, Safe R if he doesn’t)



Incumbent: Bernie Sanders (I)

Last margin: 71.0 – 24.9

Safe I or D, depending if Bernie Sanders runs for reelection.



Incumbent: Tim Kaine (D)

Last margin: 52.9 – 47

Likely D if Kaine runs for reelection, tossup depending on 2017 general election result.



Incumbent: Maria Cantwell (D)

Last margin: 60.5 – 39.5

Rob McKeena (or Dino Rossi) could pose some threat to Cantwell, but not nearly enough to unseat her.

Rating: Safe D



















Post-2016 PVI

I was going to offer an installment on my response on the 2016 presidential election (no, I am not dead, and I feel proud that my model was reasonably accurate given those percentages). However, this is my final exam week and the exams have preoccupied me, so I shall offer this alternative here.


Some interesting phenomenons:

West Virginia is now tied for third most Republican state at R+19.1. Indiana and Missouri are as red as Mississippi, all at around R+8.7. Maine reduces from D+6 to D+4 (still fairly Democratic given ME-02’s margin this year). In a Republican wave, Oregon, Connecticut and Delaware are all vulnerable, at D+5, D+6, D+6 respectively. Ohio, on the other hand, despite voting heavily Republican this year, is still only at R+3. Arizona and Georgia are both down to R+4.5 from R+6 four years ago. The new EVEN states are now PA, WI and NH (all down from D+2 ish). MI still sitting at D+1 despite voting Trump this year, given the state’s huge Democratic leaning from 2012.


Download PVI table here

2016 General: The Model is Here

Hillary Clinton is favored to win tomorrow’s election since the beginning of the 2016 cycle. Since after the primaries, Donald Trump has only lead five non-Rasmussen or Ipsos, non-tracking national four-way polls. This possibly points to an overwhelming Democratic win tomorrow, but our model, largely based on state-level polling averages, suggests only a modest Clinton win.


The model has a simple methodology: combining all polls published on Real Clear Politics conducted within 2 weeks (for the same pollster, only their three most recent surveys are included; RCP is not used because of its extremely narrow inclusion guidelines) and adjust their two-party numbers by their historical partisan leanings (provided at FiveThirtyEight). If such number is not provided, the poll’s self-reported partisan registration number is compared to statewide partisan registration numbers * 0.75 to “unskew” the poll (factored at 0.75 considering LV/RV gap). If partisan registration numbers were not provided in a poll, such poll is discounted from the model. On top of this first adjustment, average two-party polling error in a specific state is adjusted as well, according to numbers published by Real Clear Politics for the 2008 and 2012 elections.



Two party average vote is calculated using equation 1 (P denotes party, D for Democratic, R for Republican)



Systematic polling average error is adjusted by equation 2: (TPA denotes true polling average)



State AvgPE
Arizona 3.23
Colorado -3.64
Florida -1.68
Georgia -0.72
Iowa 1.7
Maine -2.85
Michigan -3.68
Nevada -4.9
New Hampshire -1.1
North Carolina -0.9
Ohio -1.03
Pennsylvania -2.1
Virginia -2.7
Wisconsin -2.58

Table 1: Average polling errors in several states


Table 1 provides us how off polls were in these battleground states. A positive number indicates poll in favor of Democratic candidate; a negative number indicates poll in favor of the Republican candidate.


From table 1, we developed table 2 (which is not shown here due to its massive size). Table 2, containing our full methodology  can be downloaded in the link below. Due to time constraint (it’s 1AM Pacific on November 8), I cannot to explain my full methodology here.


In fact, my prediction model is fundamentally flawed because of its reliance on RealClearPolitics data. If my model included more polls, it would be more predictive. My model would be most predictive in Florida, Pennsylvania and Virginia because of these states’ abundance of polls, and less predictive in Wisconsin, Ohio and Maine, where polls are not as abundant.

Here are my numerical predictions for each battleground states:

AZ: 18% D, 82% R (HRC 46.55 DJT 53.44)

CO: 86% D, 14% R (HRC 54.05 DJT 45.94)

FL: 64% D, 36% R (HRC 51.29 DJT 48.71)

GA: 32% D, 68% R (HRC 48.09 DJT 51.91)

IA: 31% D, 69% R (HRC 48.15 DJT 51.85)

ME: 88% D, 12% R (HRC 54.19 DJT 45.81)

MI: 91% D, 9% R (HRC 54.77 DJT 45.23)

NV: 69% D, 31% R (HRC 51.93 DJT 48.07)

NH: 58% D, 42% R (HRC 50.77 DJT 49.23)

NC: 52% D, 48% R (HRC 50.07 DJT 49.93) — Pure Tossup

OH: 29% D, 71% R (HRC 47.94 DJT 52.06)

PA: 76% D, 24% R (HRC 53.29 DJT 46.71)

VA: 90% D, 10% R (HRC 54.20 DJT 45.80)

WI: 86% D, 14% R (HRC 54.51 DJT 45.49)

Raw Data

My electoral college ratings:


%e5%be%ae%e4%bf%a1%e6%88%aa%e5%9b%be_20161108005714 %e5%be%ae%e4%bf%a1%e6%88%aa%e5%9b%be_20161108005837


Enjoy today and live with whoever we get tomorrow for four years!

2016 Senate Ratings

U.S. Senate
Safe D Likely D Leans D Tilt D Tossup Tilt R Leans R Likely R Safe R
Blumenthal (CT) Bennet (CO) Johnson (WI)  Toomey (PA) NV Open (Reid)  IN Open (Coats) Rubio (FL) McCain (AZ) Murkowski (AK)
Schatz (HI) Kirk (IL)  Ayotte (NH) LA Open (Vitter)  Isakson (GA) Shelby (AL)
MD Open (Mikulski) Blunt (MO) Grassley (IA)  Lee (UT)
Schumer (NY)  Burr (NC) Paul (KY)  Crapo (ID)
Wyden (OR) Portman (OH) Moran (KS)
Leahy (VT)  Boozman (AR) Hoeven (ND)
 Murray (WA) Lankford (OK)
 CA Open (Boxer)


Scott (SC)


Safe D:

Connecticut, Hawaii, Maryland, New York, Vermont: No explanation needed.

Oregon: Republican bench in Oregon is pathetic. Every statewide official is a Democrat, and the only U.S. Representative represents a conservative (R+10) district. Wyden, a four-term incumbent with reasonably high approval rating, should win reelection without any trouble. Safe D

Washington:  Murray, although not well-liked by her constituents, should easily beat his token Republican opposition given she won reelection in 2010 against a powerful Republican opponent. Safe D.

Note on WA: While Cathy McMorris Rodgers, the fourth-ranking Republican in the House, has no trouble facing reelection in his moderately Republican 6th district, she would lose to any Democrat by landslide margins because of her hardcore positions on social issues. The same applies to Jaime Herrera Beutler. The only Republican with a realistic chance of winning a Senate seat in Washington is Rob McKeena, but only if the seat is open.

California: Kamala Harris is arguably the most powerful Democrat in California as the state’s attorney general. Were Loretta Sanchez, the more conservative Senate candidate, to run a campaign on courting Republican voters while holding about 25% of the Democrats, she would win this race. Yet, polls have indicated Republicans are just inclined to skip this race altogether; therefore, Harris would safely win this seat and perhaps a lifetime Senate term. Safe Harris.

Illinois (Updated 11/02/2016 from Likely D): Before Kirk’s inappropriate comment regarding Duckworth’s family’s heritage and military service, Kirk looked all but finished. While the comment might not move the needle (or probably even in his favor) in Southern states, the national attention generated would reduce his chance in Illinois to be effectively nil. An Emerson poll confirms my hypothesis that Kirk would be behind further than he was three weeks ago. Safe D.


Likely D:

Colorado: Glenn is a weak candidate and polling indicate a massive distance between the two candidates. Glenn could potentially luck out and get a victory if something very wrong happens. Likely D.



Leans D:

Wisconsin: Johnson was all but written off bythe national Republicans. NRSC canceled its spending there long ago but there are signs of a Johnson rebound going on. Johnson never led a poll this cycle but there are signs of a close race going on there. If Feingold is absolutely safe, DSCC wouldn’t pump in $2mil to help him, just to push him over the top. Feingold should be fairly safe, likely winning by 7-8 points in a week, but if Comey effect gets more severe, Johnson could sneak out a victory here. Leans D.

Pennsylvania (Updated 11/02/2016 from Tilts D): Katie McGinty has led 10 consecutive polls conducted after October 20, including two polls from Republican-leaning pollsters (Gravis and Emerson), averaging a 5-point lead ahead of incumbent Sen. Pat Toomey in RCP average. While McGinty is clearly not a strong candidate, Toomey is more vulnurable than he was ever. Leans D.

Tilts D:

New Hampshire: Polls show a close race with slight edge going to Ayotte, but given New Hampshire’s electoral votes will almost inevitably go to Clinton, straight-ticket voting might be able to push Hassan over the top. Tilts D.

Pennsylvania: See New Hampshire.

Missouri: Like Bayh (who we can find in the Tilts R column), Blunt’s status as a “Washington Insider” is helping him in no way. Blunt started the race with a 20 point advantage (just like Bayh), but his lead has largely diminished because of Kander’s attacks over his ties to the lobbying industry (lobbying is his family business with his wife and sons all being lobbyists). This would be this year’s closest Senate race with a potential recount. Tilts D.

North Carolina: Richard Burr has made headlines recently regarding his comment on assassinating Hillary Clinton. However the state is so polarized that any token Republican and Democrat would take at least 45% each. With polling showing a deadheat, third party would be determinative here. Tilts D


Pure Tossup:

Nevada: With polls showing a deadheat between Cortez Masto and Heck, the presidential winner (of Nevada)’s party would likely carry this seat. Polls show a close race between Trump and Clinton here. Tossup.


Tilts R:

Indiana: Bayh’s collapse largely has to do with his ties with the Washington lobbyists, and in this anti-establishment year, “Washington insider” status certainly do not help candidates win election. Bayh started the race with a 25 point advantage, but Young’s powerful attacks and his “anti-establishment” tone has put Bayh into a Blunt position. He and Blunt might end up as victims of anti-establishmentism. Tilts R.


Leans R:

Florida: Rubio started with a major advantage–Florida Cubans. According to a PPP survey, Clinton leads this block by 65-22 but Rubio leads by 50-41. There is no way for Murphy to win unless he could court the Hispanic voting bloc onto his hand. Leans R

Louisiana: This one is more likely than just a lean, given Louisiana’s tremendous Republican leaning. But Caroline Fayard and Foster Campbell are both strong candidates, and in the unrealistic situation of them facing each other in the runoff, Democrats would prevail. In every other case (Republican vs. Democrat or Republican vs. Republican), the Republican candidate would almost inevitably win.


Likely R:

Arizona: This race was only “lean” for a long time mainly because of McCain’s volatility in the primary. With McCain the Republican nominee he is most likely safe. Likely R

Georgia: This race would be competitive were the Democrats to run Nunn or Carter, but unfortunately this is not the case. With a Berniecrat on the ticket, Isakson is safe barring extremely unlikely conditions. Likely R

Iowa: Grassley is very popular among Iowans and the state is shifting to the right with Trump and Clinton neck-to-neck in this Obama firewall state.  Likely R.

Kentucky: The state has shifted heavily in the Republicans’ favor. While Paul looks potentially vulnerable with scarce polls showing him only up slightly (and Jim Gray being a perfect Kentucky Democrat), the state’s Republican leaning (and difficult of polling) gives Paul a significant (and almost insurmountable) edge. Likely R

Ohio: Portman initially looked vulnerable with him and Fmr. Gov. Ted Strickland neck-to-neck in the pollings before late April, but he was able to shake off all Strickland attacks by rebuking Strickland’s remarks as completely untrue (which is indeed the case; Portman supports gay marriage [largely because his son is gay] and is moderate on several other issues). Portman also highlighted Ohio’s recessionary economy during Strickland’s governorship, effectively shifting the race in his favor. Recent polling has shown slight sign of Strickland rebound but I remain skeptical. Likely R.

Arkansas: Like Kentucky, Arkansas realigned itself completely in the recent six years (with two Senate seats, four House seats, governor, statewide offices, and state legislatures all going Republican from Democratic). Boozman is very unpopular among Arkansas voters and Eldridge has run a reasonably good campaign (being a former U.S. Attorney and reasonably conservative on all issues but he’s no Kander), but this is not enough to unseat an incumbent Republican senator in a heavily Republican state. Likely R


Safe R:

Alaska: If Murkowski was able to beat Miller on write-in, she would be able to beat him again this time as the Republican nominee. Safe R.

Alabama, Utah, Idaho, Kansas, North Dakota, Oklahoma, South Carolina: no explanations needed. Safe R





Wisconsin Senate–Is Johnson Rebounding?

Sen. Ron Johnson is clearly headed to defeat after numerous polls showed him down 10-15 points in a rematch against Fmr. Sen. Russ Feingold. However, recent polling have shown Johnson rebounding a bit, even with a Loras poll showing him leading Feingold 5 points with other reputable (Marquette) polling showing him in striking distance of keeping his seat.


Incumbent profile:

Ron Johnson (R)

First elected in 2010, 1 term


Challenger profile:

Russ Feingold (D)

Former U.S. Senator, 3 terms


Sen. Ron Johnson narrowly unseated incumbent Russ Feingold in 2010’s Tea Party wave. Elected from the slim liberal-leaning state of Wisconsin, ultra-conservative Johnson replaced ultra-liberal Feingold for this class 3 seat. Johnson proved to be too conservative for this otherwise moderately progressive state (Scott Walker brought jobs to WI but Ron Johnson did not) and Wisconsin voters have been wanting to remove him since 2010.


Ron Johnson never led a single poll in this race, not even polls conducted in 2014 or 2015, until early October when Clinton is leading Trump by a margin as large as 10 points. Theoretically, Johnson’s rebound happened att an impossible time interval. The Gravis/Breitbard poll immediately before the Loras one showed Feingold leading Johnson by 12 points, so what caused this 17-point shift in just one day?


I ran through the two candidates in Google Trends, but nothing too significant is told. The only notable surge in both candidates’ search interest graphs occured at the Johnson-Feingold debate, and both candidates got a significant but temporary surge. I also tried to obtain Johnson and Feingold’s campaign activities, noting anything controversial that could move the needle by as large as 17 points, but unfortunately nothing is found.


Therefore the only explaination of that poll is that it is an outlier. I am supposing that Johnson’s right to try bill had some impact on Wisconsin’s Senatoral race. Accoding to previous experience, Johnson’s boost is most likely temporary (he refuses to take moderate positions even when the national party abandoned him). It is extremely unlikely that Johnson would win this race but be careful to put this one safely in the Democratic column. For instance, LA gubernatorial election was rated Likely R before the blanket primary (spoiler: Bel Edwards tronced Vitter); it became lean D before runoff. (In that race, Vitter was never safe but pundits locked him)


Rating: Strong Lean D


Update (10/18/16):

Feingold up 12 in a poll showing Clinton up 8, confirming my hypothesis that the boost is temporary.

Missouri Senate–Who are the Trump-Kander voters?

Presidential hopes in Missouri for the Democrats have faded long ago, but Missouri Democrats are still able to win down ballot races. Usually, due to incumbency advantage, Senators run ahead of their party’s presidential nominees (example: Claire McCaskill (D-MO) in 2012), but this is not the case for Sen. Roy Blunt (R-MO), who faces an uphill reelection battle against little-known Missouri Secretary of State Jason Kander.


Incumbent Profile:

Roy Blunt (R)

First elected in 2010, 1 term


Challenger Profile:

Jason Kander (D)

Missouri Secretary of State, 2 terms


Kander’s campaign was really a long shot (see prediction section) before some miracle happened in early July that a Public Policy poll showed Kander down only three points. The same poll showed Trump up 10 points ahead of Clinton. A St. Louis Dispatch poll conducted a few days later confirmed the race is within five points, but this poll showed Clinton ahead in the presidential race. (The poll was conducted during Clinton’s postconvention bounce period.) Later polls all showed Blunt ahead with a small lead, until this, probably the best political ad regarding gun control ever, sending a strong signal to the conservative pro-gun Missouri while also playing with the gun control message supported by almost 90% of the American population, including 80% of Republicans. After the release of this ad, both sides now recognize this race as a tossup and DSCC pulled in swing state resources to help Kander.


In addition, “Washington (or state capital name) is broken, we need to fix it” has become the standard-bearer of politicians running for office nationwide, and in this anti-establishment year, this message has become surprisingly effective. Bernie Sanders almost upset Hillary Clinton, and Donald Trump won his party’s presidential nomination. Both candidates ran as outsiders, and their messages resonate well with the voters, who disapproved Congressional performance 7-1.


Blunt’s establishment Republican position angers Missouri voters who seek change in government. Before Blunt was elected a Senator, he was House Minority whip, second ranking Republican in the House, although he departed from the position after Republicans’ successive losses in the House. A Monmouth poll shows Blunt only having 80% of Republican support (11% goes to Kander), while Kander has 90% Democratic support, with only 5% going to Blunt. The same poll shows Kander with a 62-32 lead over Blunt in moderate voters’ views. The poll also shows both Blunt and Kander is becoming more unfavorable (compared to August), but Kander’s net favorability rating is still at +19, 1 point down from August’s +20, whereas Blunt’s net favorability has become negative.


This race might just be a replay of Arizona’s Senate race of 2012, but the R nominee for president is much more unfavorable among ordinary Americans than Mitt Romney is. If Trump gets trumped by Clinton in the 3rd debate, he would be starting to drag down downballot Republicans, making 2016, otherwise a Republican golden opportunity to expand their majority in both houses of the Congress, a 2006-style Democratic wave.

Rating: Tossup


Indiana Senate–Bye? Bayh?

Incumbent profile:

Dan Coats (R)

First elected in 2010, 1 term, retiring


Challenger profile:

Evan Bayh (D)

Former U.S. Senator, 2 terms; Former Governor, 2 terms


Todd Young (R)

U.S. Representative, 3 terms


Pundits have been very, very careful on Bayh’s chances of returning to the Senate. Most pundits rate this race at most Tilts Democratic. The race started with the release of three internal polls showing Bayh up nearly 20 points. I was initially puzzled–Bayh won his first four statewide races by 60-40 margins, his poll numbers are good, and his name recognition is almost universal in the state. Soon, I found the reason that pundits were so careful regarding Bayh. Todd Young is much stronger than Bayh’s previous statewide opponents–he is a U.S. Representative with a net positive approval rating, not lightweight opponents who Bayh faced during his two previous Senate campaigns. If Bayh is running for governor, he would win the race easily by distancing himself with national Democratic figures, but he is running for a federal office, and he hasn’t lived in Indiana for quite a long time. Since his entrance to the race, Bayh has found himself in numerous controversies–residence, job during six years, why back at this time, etc. Todd Young’s campaign also aired different attack ads on Bayh surrounding these controversies, enough for him to almost close a 20 point gap. For now, Bayh is still favored to win (the poll showing him +1 is strongly Republican-skewed), but Young is well within reach for his new job.

Illinois Senate–more competitive than you think

This is my first installment covering the Senate race. One installment would come per week.

Many pundits have Illinois and Wisconsin Senate locked for the Democrats. While recent polling has suggested Feingold well on path to victory, Illinois’ Senate race has received only minimal coverage. That’s normal because Illinois is not a swing state and polling just for the sake of the Senate race is not very cost effective.


Incumbent profile:

Mark Kirk (R)

First elected in 2010, 1 term.


Challenger profile:

Tammy Duckworth (D)

U.S. Representative (2 terms)


Mark Kirk has suffered from low approval ratings, non progressive records, and corporate-friendly stances. In a state as progressive as Illinois, Kirk has definitely been targeted by DSCC for defeat. The challenger Tammy Duckworth also has a solid record–Purple Heart Army Lieutenant Colonel, Amputee, Assistant Secretary of Veterans Affairs, and Congresswoman for two terms. But Kirk is nowhere near weak–he’s a former Navy commander and ten-year representative from the most Democratic district represented by a Republican in the nation (the district went Obama in a landslide twice), and he defeated the state’s treasurer six years ago.


Kirk generated much publicity by being the first sitting Republican Senator to support Merrick Garland’s nomination and has met Garland personally. This gave him a significant advantage ahead of Ron Johnson (Wisconsin Senator target for defeat by DSCC), who refused to support Garland’s nomination and to meet him. Also Kirk’s record is much more moderate than Johnson’s.


Recent pollings also suggest Kirk is stronger than what we perceive. In an internal poll, Duckworth only leads Kirk by 7 points. Duckworth’s lead is even smaller in public polls–a Loras College Poll has Duckworth leading by a sheer five points, whereas in the same poll Clinton leads by 14. An Emerson poll also shows Duckworth leading only by 2, but Emerson only samples on landlines so it skews heavily in favor of the Republicans. Compared to Feingold being up by almost 15 points, this polling result raises an alarm for Democrats so optimistic about defeating Kirk.


But Duckworth starts with a major advantage–she has a war chest of almost 6 million dollars. This is significantly higher than Kirk’s 3 million. Keep in mind that Kirk has almost been abandoned by the GOP and the Republican Senatorial committee, so I am not very optimistic about his chances of defeating Duckworth, especially when Duckworth’s ads bombard Illinois. Also, her campaign seems pretty smart too.


In conclusion, I believe Duckworth starts with a major advantage (mainly due to her warchest campaign funding), but Kirk still has his chances.

Rating: Strongly Leans D

Update (10/18 2016): I am moving this to D Favored mainly because of Loras’ massive fail at Wisconsin and a subsequent poll showing Duckworth up 15 points.