Date: January 14, 2016 From: Matt Braynard To: Chuck Laudner Matt Ciepielowski Jim Merrill Ed McMullen CC: Corey Lewandowski Michael Glassner Re: All In On “Low-P High-T” Summary: Based on an internal analysis of our own modeling data and third-party research, and considering the exceptionalism of our candidate, I advise that we put one-hundred percent of our organizational effort into enfranchising the conventionally low propensity voters that support our candidate.1 Voter Math: Two Coefficients All voters can be assigned two coefficients: their propensity to participate in the Republican nominating contest (P) and their likelihood of supporting our candidate (T). XP * YT = Outcome These coefficients are in flux between now and election day, their values sliding between zero and one. A low P value is a voter who is unlikely to vote; a high P value is a voter who is exceptionally likely to vote. A low T value is someone who is not inclined to support our candidate. A high T value is someone who is. If you look at most national surveys, what you are seeing are high propensity voters (eighty percent chance of turning out) and about thirty-five percent support us: 0.8P * 0.35T 1 In Iowa, Nevada, and other caucus states, low-P includes anyone who has not previously attended a caucus. 725 Fifth Avenue · New York · New York · 10022 · www.DonaldJTrump.com Paid for by Donald J. Trump for President, Inc However, based on our modeling data on FITN states, when you look at unfiltered, low-P, Republican-inclined voters, the number looks like this: Iowa 0.1P * 0.43T New Hampshire 0.1P * 0.54T South Carolina 0.1P * 0.62T Before any additional segmenting, we are getting 43% of Iowa’s low-P, Republican-inclined voters, fifty-four percent of New Hampshire’s, and sixty-two percent of South Carolina’s. By adding a few filters to each of these universes, we can identify clusters that move our T coefficient into the seventies. On Election Day, each voter’s coefficients become either one or a zero, and we only win if both of those numbers are one. Love Our Candidate, Didn’t Vote 0P * 1T = 0 Votes Love Another Candidate, Did Vote 1P * 0T = 0 Votes Love Our Candidate, Did Vote 1P * 1T = 1 Vote Traditional primary candidates, including all of our current opponents, lock onto the same highP voter universe (“four of fours,” “three of fours,” etc.) and focus on moving T values. They do this through radio and television commercials, direct mail, persuasion calls, push polls, and lit drops. They push other candidate’s T values down and try to push theirs up, all fighting over the same heavily tilled soil. And, at the same time, our opponents ignore the low-P voters. And this makes sense for them because they are starting from a position where both coefficients are low. After all, not only are these people unlikely to vote, they have no clue who John Kasich is or why they ought to miss their kid’s soccer game on a Tuesday night for him. For the low-P crowd, it is a vicious cycle. They don’t get any voter contact love from the campaigns because they don’t vote, but they don’t vote because they don’t get any voter contact. It is a persistent state of disenfranchisement – and among these people are fervent supporters of our candidate.2 The Case for Ignoring High-P Voters 2 http://www.nytimes.com/2015/12/31/upshot/donald-trumps-strongest-supporters-a-certain-kind-ofdemocrat.html?_r=0 Page 2 of 4 Campaign ground games engage high-P voters to express focus-group tested messages to persuade undecided voters and inoculate their current supporters from being peeled away by attacks – or to go on the attack themselves. None of these are applicable to our campaign. Our candidate commands unheard amounts of earned media coverage; the message is getting out. A recent focus group by Frank Luntz found that attacks on our candidate pushed unsure voters to us and hardened the commitment of our existing supporters, 3 and our candidate can launch powerful attacks at any time on social media. Campaigns also devote effort to GOTV on their identified high-P voters, but is this really necessary if their primary qualification for engagement in the first place was that they were regular Republican primary voters? And do our own extremely motivated high-P supporters really need a knock on the door, phone calls, and dozens of direct mail pieces from us to tell them to go vote, especially when the other candidates are already stuffing their mailboxes? 4 What good is spending the effort to get a hard ID on a high-P voter going to do for us? Why not devote the entirety of that effort to increasing the other coefficient, which all of the evidence suggests makes so much more sense for us? Engaging Low-P High-T With our research, we can readily identify the Low-P voters that are likely High-T. Even better, we have them all to ourselves; the other campaigns will not give them much attention because it is not worthwhile for them to do so. Moving the P coefficient higher is more akin to evangelization than persuasion. These are people who may not know where to vote, whether or not they are eligible to participate, and what the hours are. They may have to work on the actual election day and are unaware of early voting opportunities on Saturdays. Many of them may have simply never been asked for their vote. In the media and in certain social circles, our low-P supporters have been called “stupid,” “racists,” and “bigots.” The “elites” look down on them, and there is evidence they deny their support for our candidate because they fear being ostracized.5 That’s why the engagement scripts for low-P must all open with a warm and friendly disclosure: “I’m a volunteer for Mr. Trump’s campaign for President.” This lets them know they’re in a 3 https://www.washingtonpost.com/politics/attacking-trump-just-makes-these-voters-like-himmore/2015/12/10/2de9edf8-9f44-11e5-a3c5-c77f2cc5a43c_story.html 4 In Iowa and Nevada, it’s still important to let even seasoned caucus-goers know their polling locations. Between event attendees, voters we’ve already ID’d, other support indicators and voter modeling, we can get out at solid mailer with this information to our high-P supporters. 5 http://www.deseretnews.com/article/865644555/White-males-with-secret-Trump-crushes-lying-topollsters.html Page 3 of 4 “safe space.” Then we follow with an encouraging invitation: “We’d like you join us, to help us Make America Great Again. Can we count on your vote in the upcoming primary?” We express our appreciation for them and remind them of what’s at stake if they don’t vote. We make sure they know the rules for voting, and we appraise them of early and absentee opportunities, hours and poll locations. We follow-up with mail and email. On primary and caucus day, we call them early in the day to remind them. We have our volunteers out in force at key precincts and caucusing locations operating a strike list that informs our live volunteer call centers and robocall vendors of the stragglers so we can give them further encouragement. A Radically Different Strategy for a Radically Different Candidate Republican Presidential campaigns have been run, mostly, on the same basic strategy since most states switched to primaries from the old convention tradition. And for the same basic campaigns, this makes sense. Low-P high-T voters require a little more effort, but we don’t need to engage nearly as many of them as a traditional campaign targeting high-P. By identifying and engaging a low-P high-T universe of fifteen to twenty percent of our win number in each state, we can ensure that there’s no way any opponent can overcome our lead. They will not have enough voters left to persuade. They won’t have enough gas in the tank to beat us. Our candidate is unique in the history of American politics. Our campaign defies gravity because the old rules of physics don’t apply to us. An unprecedented targeting strategy must be in sync with this unprecedented campaign. Page 4 of 4