Demographically, West Virginia seems like it should be a blowout in favor of Bernie Sanders. The polls, despite being all over the place, also seem to imply that West Virginia will be a blowout, with some giving Bernie a margin of victory as high as +28%(!). Recent polling has come back down to more believable margins, like +4% and +8%, which I think are much more in line with reality. Here is my estimate for what we will see tomorrow in the Mountain State:

Screen Shot 2016-05-09 at 6.50.51 PMWest Virginia is 3.4% African American, similar to states like Washington, Nebraska, and Colorado, all of which Hillary Clinton lost by margins that were quite large. But, there are other factors that point towards a <10% margin of victory. First, the relative search interest between the two candidates is leaning Hillary in West Virginia. The average relative search interest in all of the states that Bernie has won thus far is 2.23, whereas in West Virginia the relative search interest is 0.94; meaning that West Virginians are actually searching for Hillary Clinton more than they are Bernie Sanders. In fact, the lowest his three day relative search interest has ever been in a state that he ultimately won was 1.54 in Minnesota, and just last week in Indiana the relative search interest was 1.66. Secondly, Bernie’s share of Facebook Likes in West Virginia is only 75%, compared to an average of 81.2% for all of the states that he has won. I know that 6.2% doesn’t sound like much, but it really does make a big difference when you consider that the entire range of values for this measure is between 63.6%-95% (Mississippi-Vermont). West Virginia is also a relatively old state, with a median age of 41.9, compared to 37.78, which is the average median age of all the states that he has won.

Aside from this, West Virginia is a semi-closed primary, which usually helps Bernie slightly. In lieu of the Republican race being tied up at this point, it will be interesting to see if the semi-closed primary format attracts all of the independents to vote in the Democratic primary tomorrow. Also, Hillary’s comments on the coal industry have gotten quite a bit of media attention over the past week. This will probably help Bernie (even though Bernie pretty much has the same position on the issue). Also, the logged value of the relative number of campaign contributions is 0.36 in all of the states that Bernie has won so far, and in West Virginia this number is 0.405. This obviously looks positive for the Sanders campaign.

Lastly, on average, I underestimate Bernie’s vote share in semi-closed primaries by 3.27%, so I think we will end up with Hillary performing slightly worse in West Virginia than the numbers above indicate. Regardless, West Virginia is possibly the most unique state thus far with regard to the data that I look at. It’s really surprising how different pieces of data are really trying their best to pull my model very far in different directions in this state.

Thanks everyone, and happy voting to all of you West Virginians out there!


As you all know, I’m just a poor grad student. If you would like to support my work and want me to be able to afford Top Ramen (or maybe even Mellow Mushroom pizza if you all are extraordinarily generous) while I’m working on these statistics, please click this link to donate to Tyler’s Food & Rent Fund!


  1. Am I the only long-time reader disturbed by the fact he is soliciting monetary donations with his pleas as a poor grad student, failing to mention he bets -thousands- on elections at PredictIt?

    Disclosing the fact he wagers on elections should be common sense even without his begging for money, so this is getting out of hand.

    I feel like I -should- mention a red flag about his history that further compounds this problem, but I’m still hoping he comes back to reality before the election is over since he has potential to be a valuable contributor.


      • Are you arguing that spending effort recently to add a footer code to all his posts (including historical ones) that link to his paypal and beg for money is more important than 1) disclaiming to readers he has a financial interest in the outcome of elections 2) the primary source for data in his model is 538


    • What you are suggesting is a smear directed at Tyler Pedigo, without any facts or proof to support your inflammatory accusation, and I hope Tyler will delete your offensive comments.


    • I think I don’t care. You really need to get a life. Love this guy’s page and his predictions and methodology are pretty interesting. (Even if I back Clinton and he’s pro-Bernie). I’m not donating to him myself, but if ppl want to then all the best to them. #hatersgonnahate


      • @ Andrew, I don’t want to fight, I just want to ask in a friendly and non-combative manner, why do you support HRC over Bernie?

        Bernie is the stronger General Election candidate, and on nearly every issue (except guns) is far more progressive than HRC, his plans have been shown that they will help more Americans than hers (especially minorities) and he has a proven track record of integrity compared to her less than perfect record on that issue.

        So I am curious, honestly and without rancor, what is the reason you support her over him?


  2. Hi Tyler,

    Curious to know how you feel about the results considering the 10%+ that was allocated to other candidates. On Bernie’s predictions, you really couldn’t have been closer, but I imagine your model was only looking at the two front runners. Do you think your model would have pulled that 10+% from only Hillary or both candidates instead? Just curious to know your thoughts. Excellent work regardless.



    • It’s not like a candidate just dropped out and they no longer had their first choice and couldn’t decide. It was expected WV would have the highest protest vote in the nation, and it did not surprise any other “election data experts.”

      Thinking the model was partially accurate because by coincidence it had a similar percentage for Bernie in an apples to oranges comparison and it shows you need to escape the bernie bubble! You need to be comparing the percentage gaps since it is a 2 person race. The rest is noise.

      If we remove the protest votes:
      Bernie: 58.9%
      Hillary: 41.1%

      Tyler’s predicted gap: 3.6%
      Actual Gap: 17.8%

      Just accept it: his model predicted a tossup with a gap less than his models margin of error, but it was a blowout.

      If you want models like this to improve and be reasonable predictors in the future, you must accept that some states have variables his social media data is not accounting for. Same thing happened in Florida, and in that case it was a combination of misunderstanding social media activity of seniors, hispanics, and one of the highest rates of non-citizens in the country.

      Try to figure out what went wrong instead of doing mental gymnastics to suggest the model was amazingly accurate for Bernie.


  3. Gotta say, you did an excellent job predicting Bernie’s vote share. As for Hillary’s vote share, that can be adequately explained by 12.5% of voters going for Paul T. Farrell, Jr. (almost 9% of the vote), Judd, O’Malley, and De La Fuente Guerra, IMO. West Virginia just doesn’t like the establishment.

    So…any ideas how Kentucky is looking like?


    • If you want Tyler’s work to improve over time, stoooopppppppppp. He needs to realize this was a YUGE miss.

      Tyler has explained on numerous occasions his model is based on an election with only 2 options. He could include the third variable since it is predictable based on both this primary season and the past (hint: Appalachia). He decided it was too much work for something that only mattered in like 3-4 states. (In Missouri, it would have changed the outcome if the whole protest vote went to Bernie!)

      It’s not a big deal at all but then I keep seeing comments like this, and Tyler is so desperate for praise he might just convince himself y’all make sense.

      If you understood how models like his work, the results of WV are plugged into his model to try to improve accuracy for the next state that votes.

      If you look at the results the way his model should (based on how he has coded other variables), and remove the extraneous data, this is the result

      Bernie: 58.9%
      Hillary: 41.1%

      Tyler’s predicted gap: 3.6%
      Actual Gap: 17.8%


      • NateSilversStalker: Wow. All that could be true, or you could just be a complete douche.

        You’ve got a grad student posting his predictions based on his model and receiving nothing for it other than some valdiation of his personal interest and time spent. He’s not getting paid by a campaign, or anyone else for that matter.

        Reading through your posts here, it’s apparent you’ve got some bone to pick with Tyler. Why not come clean about whatever it is, rather than whine about your perception that he missed, that you don’t agree with other’s complimenting or thanking him, or his inclusion of a donation link. If you truly don’t have an issue with him, then you might want to do a bit of introspection since your attacks are making you look worse than your intended target.


      • GE polls don’t count. The nominee is chosen by ACTUAL votes and pledged delegates…both of which are overwhelmingly in favor of HRC. She’ll beta Trump as well, as the map is not favorable and he is a very bad candidate.


    • He’s still running to get his message out and to help progressives figure out which districts they can mount primary challenges in 2018.


      • NateSilversStalker,

        You don’t think that how well Bernie does in a given district is a good indicator that a progressive could successfully campaign there? You think the 85% of 18-29 year olds that voted for Bernie are just going to disappear and that American politics aren’t going to shift left in the future?

        LOL! But then, I can’t expect too much out of someone that idolizes the guy who swore up and down that Trump could never get the Republican nomination.


    • James Williams To: Everyone,
      I hope you know that the Political Revolution started on 02/01/2016 in IOWA and this means no more talk, it’s time for action.
      If you are too afraid to stand up for your RIGHTS then you cannot complain of what’s to come, for what you are fighting for is not just for you but everyone that come after you, your children and grandchildren. It’s NOW or never, we have waited for the people that we elect to get us out of this BIG MONEY control, the wait is OVER the elected officials are in too deep with BIG MONEY.
      We have to do this or BIG MONEY will run your life forever, we had to do this in the Nineteen Thirty’s, then in 1980s Big money started to run amuck and kept telling us to CUT our TAXES and WE will CREATE JOBS, it’s been OVER THIRTY years, WHERE are the JOBS? Oh wait there in CHINA and other Countries without regulations. Big Money milked us dry and collapsed in 2008 and then WE had to bail them out, what the HELL. They get a bail out and we have to suffer, it’s time to get MAD, It was OUR TAX MONEY that WE worked for to help US and they got IT. Look they are NOT paying their fair share of taxes to benefit US? I don’t think so. What would have happened if instead they bailed us out? We would still be IN OUR HOMES driving OUR CARS PAYING OUR BILLS, instead we have to live on the streets and live in POVERTY. If you know of a homeless person take them with you to CAUCUS get them registered and let then help out as well.
      We are at WAR with a CORRUPTED SYSTEM, for those of you that think the Bernie Sanders POLITICAL REVOLUTION is just a theme, you are WRONG, it’s what we are going to do. We don’t need a gun to get our way, WE will use our VOTE and our bodies as our weapon of choice. See you don’t have to KILL to get your point across, numbers of people can do the same without KILLING anyone. This is not to say that while we are at it some of us may go to jail, but I bet when it’s all said and done, some people that thought they were above the law due to having MONEY, will find themselves behind bars.
      The POLITICAL REVOLUTION of the United States of America Starts TODAY
      This is NOT a game or a gimmick this is a WAR on BIG MONEY and taking back our government. If you stand in our way we will move you, if you try and STOP US we will grow LARGER. Join us or get out of the way, a STORM is coming in the likes that this NATION has never seen and hopefully will never have to happen again.
      This is NOT just about the United States of America for this is affecting the entire GLOBE, our COUNTRY is not the only one that is FIGHTING BIG MONEY. The FIGHT STARTS HERE AND IT STARTS NOW!!!!!
      This is a WAR on POLITICS, it’s NOT just a President we are electing, there are a lot of Congress seat that will be taken (the do NOTHING congress) as well as we have a large number of Supreme Court Judges that are out as well. Thing can be changed to get it done, if not we know where the WHITE HOUSE is and were CONGRESS sits. If we have something NOT getting done that we want done, we have to march onto the window of CONGRESS and tell them ourselves. People in this country have to STOP being lazy and yell from the comfort of their living room, but hey we have plenty of us LIVING ON THE STREETS that wouldn’t mind marching. The STORM is coming and people better wake up, you join us or get out of the way. For we will bring our huddled masses.


    • Are you concerned that Bernie keeps winning while Hillary keeps losing? Hillary should quit, her unfavorability with voters is sky high and she cannot beat Trump, only Bernie can. Besides, the FBI indictment looms in the horizon. This ain’t over until California votes.

      Liked by 1 person

    • Not going to happen. Hillary and her Foundation is under an FBI investigation. She should not even be running in this contest. She and the Dem Party are corrupt to the core. #Dropouthillary.

      Should she get the nomination, she will lose the general. If she does, the onus is on the Dem Party, Hillary, DNC, her surrogates and her supporters for the election of Trump.


      • the onus would be squarely on people like you. and for bernie for staying in when every other recent dem candidate in modern history dropped out for party unity based on MATH.

        fortunately, you are just a very vocal minority, and it was never expected that you would be needed in November. Plus, the vast majority of bernie’s primary voters have repeatedly stated in 100 freakin polls that they will vote Hillary in November. And that is what they are saying now–when many know that Bernie’s sole remaining argument is to use the poll results to sway supers, something that will obviously cause some future Hillary voters to lie to the pollster in hopes they make Hillary look bad.

        I’m sorry if you feel left out as a minority of the minority faction that supports the losing candidate. Must be frustrating to know most of your fellow voters ignore your bernieorbust crap because they arent as privileged as you or they voted in 2000 when Nader gave us Bush.


  4. Could this

    First, the relative search interest between the two candidates is leaning Hillary in West Virginia. The average relative search interest in all of the states that Bernie has won thus far is 2.23, whereas in West Virginia the relative search interest is 0.94; meaning that West Virginians are actually searching for Hillary Clinton more than they are Bernie Sanders.

    simply be due to this?

    Hillary’s comments on the coal industry have gotten quite a bit of media attention over the past week.

    Liked by 1 person

  5. Some quick comparison between WV and Indiana voter demographics before I give my prediction on who will win and by how much. According to, Indiana population in 2014 was made up of 80.3% white, 6.6% Latino, and 9.6% AA and West Virginia population is made up of 92.5% white, 1.5% latino and 3.6% AA. The exit poll for Indiana had voters by race broken down to 72% white, 19% AA and 6% latino. There is no exit poll for West Virginia in 2008 because it held a caucus then. If we use the same numbers from Indiana, AAs will account for about 5.9% of the votes, Latinos would account for about 1.65% of the votes and whites would account for about 92% of the votes in West Virginia. With that said Sanders won the white vote in Indiana 59-41 and lost the AA vote 26 – 74 but in the case for West Virginia I do not believe AAs will have such a huge turnout as they did in Indiana which accounted for about ~14 points of the 47.5 points Clinton won.

    So my prediction is that Sanders will win because of the demographics in WV and I believe he will win around 59-60% if not slightly higher. I am predicting this percentage on a few things. 1) There is very little percentage of minorities in WV. I am not saying that sanders doesn’t do well with minorities but because they are such a small number they will have very little effect on the outcome. 2) Sanders won whites with a college degree 55-45 and whites without a degree 65-35 in Indiana. In West Virginia there are slightly more people who either do not have a college degree or finished high school than there are with a college degree or higher compared to Indiana. So not only should sanders again when these voters in West Virginia but by a slightly larger margins especially the voters with no degree. 3) Sanders won all income groups $30k or higher by roughly 54-46 while losing the $30k and under group by 47-53 in Indiana. Clinton shot herself in the foot by talking bad about coal a few months ago and it caught up to her while she was campaigning in West Virginia and she of course flipped-flopped on her stance. Something tells me WV people do not believe clinton at all and so Sanders should win all income groups including the 30k and under by slightly large margins than he did in Indiana, maybe 56/57 – 44/43 if not higher.

    Tomorrow is going to be a bad night for Clinton and I only wished WV had more than 29 pledged delegates to offer. The question is how bad will it be and how much will MSM downplay or spin it in favor for Clinton. My guess they will continue to talk about how it is almost impossible for Sanders to catch up to Clinton but I this takes away from the fact that even if this is true, Sanders supporters are not given up without a fight and as Sanders continues to win states going into June 7th I believe this will bring more people to the poll to vote for him thus given him a lot of fire power going into the convention.

    P.S. I see very little difference between an open primary and a semi-closed primary especially if look at past open primaries, like indiana, you had very little Republican cross-over, 2% in this case. Sure this might change in future states that have open primaries/caucuses now that Trump is pretty much the GOP nominee and thus you will have some Republicans coming over and either voting for Sanders or Clinton to vote against Trump.


    • Sorry it seems WV didn’t have a caucus like I thought but a primary and after looking over the exit polls for WV for 2008 my prediction/numbers still stand after seeing that 96% of the voters were white and 3% were AA which are better numbers than I came up with. Also even though 66% of voters in WV in 2008 were either moderate or conservative, polling for WV between Clinton and Sanders shows Sanders beating Clinton with Moderates and Conservatives but losing surprisingly with Liberals. I think like Indiana, Sanders will win liberals but a 10+ margin but only lose moderates/conservatives by 5 points if not tie them.


  6. Hey there Tyler ~ I’ve appreciated your work on this primary season. You’re very talented & have a bright future ahead. I donated at least enough for a small bit of groceries. I hope more donate to get your through! Peace!

    Liked by 1 person

  7. dude your predictions literally move entire markets on PI. thanks for the forecast!

    good luck on the job front. anyone would be lucky to have you


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s