Disclaimer:

Disclaimer: While I am a paid undergraduate researcher for Texas Tech University, the opinions expressed on this blog are my own. The statements on this blog are not endorsements of views of Texas Tech University, The Honors College of Texas Tech University, or the math department of Texas Tech University, or any other department or campus organization of Texas Tech University.

Saturday, January 30, 2016

OFFICIAL IOWA CALL GOP

Opening Notes: Since the Iowa process is weird and has multiple steps until the real delegates are chosen I think the lion's share of the delegates go to the top candidates since Rubio will probably get most of the votes of the "establishment"/moderate wing in his areas making it hard for the other more moderate candidates to get a delegate.  Carson and Paul get 2 and 1 delegates respectively because of

GOP
Delegate Ranking: (Not counting the non-bound 3 RNC members)

1. Cruz - My model actually puts the majority of the confidence interval with a Trump advantage, but the weather may be bad in parts of Iowa. My range is 25.31338-28.54439.  Looking at 2008 and 2012 I see that strong reliable conservatives normally win.  I also predict that Cruz will get the most delegates because when decisions have to be made at county conventions I think there will be some consolidation at the conventions following the caucuses.  There are essentially three wings of the party the conservative wing, the moderate wing and the Trump wing. Iowa is bound to proportional allocation but that doesn't mean it has to be exactly proportional.  So second ballots can happen with less candidates to help break ties.

Straw Poll Range -(26.31337881 - 29.54438775)
Delegates - 10 out of 27
My Center Percentage: 29.25%


2. Trump- Since there is a possibly of a snowstorm, and the fact Trump ditched the last debate I don't think he can win.  Trump is still pretty popular in the state so I think he will become a strong second.  Cruz is pulling support from more involved citizens in the republican party.  Trump has a base that is poorer, less educated than the average republican voter.

Straw Poll Range - (26.10459011 - 29.71628811)
Delegates- 8 out of 27
My Center Percentage -  28%


3.  Rubio
Rubio's draw is a mix of moderates and conservatives but I personally think he leans more moderate. He has  an endorsement from the Des Moines Register giving him the lead among more moderate "establishment" republican. He could be a wild card and get much more of the vote because of the his success in the last debate.  Although my range does not include this center 17.5 number it comes from a last minute decision to raise the number due to the last poll from the Des Moines Register and later polls.  Because of this I am assigning about half of the undecided votes the rest I think will mainly go to Ted Cruz.

Straw Poll Range - (14.81268703 - 16.29825324)
Delegates - 6
My Center Percentage -16.25%



4. Carson
Carson was doing really well in October/November but he had a few faux paus that split his vote between Trump and Cruz.  I still think he can pick up some delegates here and there.  Honestly his campaign shouldn't make it past March 1 and if it does it will be pretty much over.

Straw Poll Range - (8.889090212 -11.67925576)
Delegates - 2
My Center Percentage - 9.5%



5. Paul
Given his history and his ability to be the original  debate skipper and a strong personality and a stronger ground game I think he can get some votes with the same people who voted for his father in Iowa.

Straw Poll Range - (3.022284398 - 3.86111691)
My Center Percentage - 3.5%
Delegates - 1


6. Bush
Bush has the struggle of overcoming his family ties to politics in a year where a populist and non-experienced candidate like Trump can get a significant share of votes. Rubio's debate performance will probably hurt him. I am pulling a center outside of my calculated range because of the final poll and trend.

Straw Poll Range - (2.5-4.044822)
Delegates - 0
My Center Percentage -2.5%


7.. Christie
Christie didn't really spend as much time in Iowa as he did in New Hampshire.  Marco Rubio has the element of conservatism with moderate policies that makes him a better Iowan candidate.

Straw Poll Range -(2.491348072 - 3.378999056)
Delegates -0
My Center Percentage - 3%


8. Fiorina

She did well in the debate and her ability to attack Clinton  with being called sexist I think will help her in Iowa.  I don't think she can pull a delegate because of the process.

Delegates - 0
My Center Percentage - 2.75%
Straw Poll Range -(2.067145578 - 2.805242538)


9. Huckabee

Huckabee may have done well in Iowa in the past but the crowded field won't help him.

My Center Percentage - 2.25%
Delegates - 0
Straw Poll Range -(2.097198201 - 2.733258864)


10. Kasich

Iowa isn't the place for Kasich, I do think he will do much better in New Hampshire which has a more moderate climate.

My Center Percentage - 2%
Delegates - 0
Straw Poll Range -(1.779842263 - 2.446807669)


11. Santorum

He did do well in another cycle but I think he is having trouble and losing support. His visit to the Trump Rally may be hard to get past.

Delegates - 0
My Center Percentage - 0.9%
Straw Poll Range - (0.798901757 - 1.250383059)


12. Gilmore

His support is pretty low and he only made one debate and is struggling to make ballots.  I don't think he will do well.

Delegates - 0
My Center Percentage - 0.1
Straw Poll Range - (-0.046439701 - 0.295066167)



Also I would like to remind people that even if Trump wins he is far from the majority of the party.  Also the  weather and turnout will probably call this election.


OFFICIAL Iowa Call Democrats

Martin O'Malley does not in my opinion have enough support to get a delegate due to a 15% threshold.  I am reassigning his  share, estimated to be about 5% percent with 3% to Sanders and 2% to Clinton, based on the average profile of a O'Malley voter and the profiles of a vote of the other two candidates. I have not been watching the democrat side of the process as much but I am releasing an official call.

Ranking: 44  Delegates at stake.
1. Clinton -
Although Bernie Sanders is putting up a very close race, I do not think he can win Iowa, however I think he can gain over 40% of the vote.  The fact that Clinton's emails have been recently withheld due to classification issues I think will swing the vote.
Delegates - 24
Percentage Range - 53.28385-56.35697%
My Center Percentage : 54%
2. Sanders -
Sanders recent rise has increased his popularity.  He does have a chance of winning but his primary draw of younger voters and in light of the possible winter weather I don't know if he can get the turnout he needs to win.  However I think Sander is a force to be reckoned with and can win the election if he can appeal to minority voters.
Delegates -20
Percentage Range - 43.00412-47.50853%
My Center  Percentage
3. O'Malley-
Due to the Democratic policy of only counting candidates with a 15% support at caucus I predict that O'Malley will not pass this process in most if not all caucus locations giving him 0 delegates. The polls place him at about  I predict this supporters will split  between Clinton and Sanders, with a larger number going to Sanders based on the demographics of O'Malley supporters. It is possible in a few locations O'Malley will make the 15% mark but he will probably not have enough support to get a single delegate.
Delegates - 0

Guide to Election Calls

Just a clarification right before I begin calling Iowa. First, let's talk about my methodology.  For Iowa, New Hampshire, South Carolina, and Nevada I will use a t-interval to generate a range. The other states will use a Bayesian approach using the results of the prior states and poll data to rank candidates based on probability of winning.  Then I will also generate a range of percentage points. For both sets of races I will follow similar calling procedures. Based on that range I will pick a number based on recent events and polls that are regarded as more reliable that is in my opinion the most likely percentage on election day (My Center Percentage).  I will then assign delegates based on the rules of that state for delegate allocation allowing for possible combinations of candidates if applicable.  My primary goal is predicting delegates vs. percentages because delegates ultimately decide the race.  Closer to July I will reorganize delegates if a possibly of a brokered convention still exists.

Main Debate Thoughts

After watching the debate and the aftermath I have some conclusions.
1.  Ted Cruz probably lost some support  because he was the main  target of the other candidates.
2.  Donald Trump  probably lost some support to because ditching the debate might anger the people who historically show up to caucus: dedicated republicans.
3.  Fiorina may not last much past New Hampshire but had a strong performance.
4. Rubio will do pretty well do to recent endorsements and a good debate performance.

Thursday, January 28, 2016

Undercard Debate review

I worked on homework during the undercard debate so I didn't live tweet it but the main stage I will.
Overall I think Carly Fiorina showed her strength. she most likely won't make it past March 1, but she is a very powerful voice for any candidate to have. I don't quite understand what Jim Gilmore was doing there because he isn't very popular.  My predictions for the undercard candidates future is as follows: (Not always backed by solid math)
Carly Fiorina will probably go up a little but not enough to receive any delegates.
Jim Gilmore might break 1% (maybe 2%)  in Iowa on Caucus Day in the first ballot votes.
Mike Huckabee and Rick Santorum's stunt on going to the Trump Rally might actually help them and take votes away from Ted Cruz.

I know this is being posted like 45 minutes from showtime but my debate predictions are:

Ted Cruz becomes the new Donald Trump in terms of attacks and bold statements at the debate.
All the the candidates walk a fine line between pointing out that Trump betrayed the party and pretending he doesn't exist.

My Iowa prediction should come out before 7PM CST on Saturday but probably Saturday morning.

Wednesday, January 13, 2016

How Bernie Sanders can be the Democratic Nominee

I am not saying that Bernie has it in the bag but what I am saying is that he does have a chance to win. Obama won South Carolina, Iowa, and New Hampshire in both elections.  I am pretty far from  a Sanders supporter but I can see how he can pull it off.  As of today, I am pretty sure he will win New Hampshire and he can win Iowa and South Carolina if he continues to rise. But Bernie Sanders is very popular with middle and lower class voters who don't always vote.  I think if he can win those three states he can gain momentum and win the nomination.   However, in most polls Sanders can't win very  significantly versus most GOP candidates  except Trump. Clinton has a much closer race.  If I had to call the entire election today I would pick Sanders and Cruz, with Cruz winning.

Wednesday, January 6, 2016

The Donald Trump Effect and a Brokered Convention

The thing that makes this race interesting is no one has gotten a 50.01% majority in any reputable poll in any state or a national poll.  But the thing is a lot of the early states use proportional delegate assignment or allow the candidates to group together to get a majority.  And in the end a candidate needs 50.01 to win the GOP nomination.  But here is the thing what if no one has a  majority?  Well then the candidates that have either dropped out or have not enough delegates to have a shot at winning will give their delegates in a brokered convention to candidates who did better than them that are at the top.  There  are really three groups of candidates: conservative, moderate, and Trump. Trump gets his own candidate because no candidate will give their delegates to him.  He is too dangerous for the party.  The Trump group has well Trump.  The conservative group has Cruz, Carson, Paul, Santorum, Huckabee,    The moderate group has Rubio, Christie, Bush, Kasich, Fiorina. And there is also Jim Gilmore who is technically running  but has very few supporters and isn't on any ballot right now so he doesn't really count.  Candidates like Carson and Kaisch may not make it much past Iowa and New Hampshire but they have a chance to have a delegate or two each because of proportional allocation.  If there is no majority candidate which I  believe is very likely at this point, then Kaisch will give his delegates to the leading moderate candidate (probably Rubio) and Carson will probably give this delegates to the leading conservative candidate (probably Cruz).  In the end the nomination will go to the winner of the moderate and conservative wing.  Remember in the national poll where Trump has the highest percentage the other candidates combined have the majority.

Feel free use the Contact form and comments, but overtly biased and inflammatory remarks will be removed.



Source : gop.org and elections.huffingtonpost.com/pollster
I will have a full biblography for the real paper and presentation.

Welcome

My name  is Brittany Alexander.  I am a math major with a minor in economics at Texas Tech University.   This is the official blog of my predictions and comments for the 2016 election.  I am entering my second semester of college and I decided to start a undergraduate research project on presidential political statistics.  I will present my predictions at the Texas Tech Undergraduate Research Conference. I am a republican and I plan on voting for Marco Rubio.  I will avoid bias at all costs. This is were I set down my rules.

1.  I will not edit any posts, but I may change my predictions  in additional posts.
2.  All posts will be permalinked.
3.  All posts will be shared to my twitter @RaceTo2016Stats .
4.  I will live tweet election results on March 1, March 14, and November 8 and every primary night  I can.
5.  I will focus on the GOP nomination, because it has very interesting dynamics and will take most of my free time.  If I can I will try to get a prediction for each democratic primary too.
6.  I will read other predictions, but I will use my own math. (Combination of Bayesian and t-intervals)
7. For states without adequate polling I will use the states that already voted and/or have numerous polls, and previous elections to predict.
8. I will possibly take statistical liberty to exclude polls that I do not believe are valid.
9.  I will provide at least a weekly update.
10. All predictions for the case of tracking will be made final at 9 PM CST before the election.


My goal is to predict the republican nominee and possibly the democratic nominee.  I want a 80%  correct delegate assignment and a 90% correct prediction of leader in state polls.