Trump Won the True Vote

Richard Charnin
Dec. 5, 2017

Look inside the book: Trump Won the True Vote

Mainstream media pundits claim that Clinton won the primary and presidential election by three million votes. It’s a myth. They fail to consider the FACT that the recorded vote is ALWAYS fraudulent.

A True Vote Model analysis indicates Trump won the popular as well as the electoral vote. The pundits always assume that the recorded vote is accurate but never consider the fraud factor. The historical statistical evidence is conclusive: every election is fraudulent.

Advertisements

 

Proof that Oswald was Doorman: Testimony of Billy Lovelady and Buell Frazier

Richard Charnin
Dec.3, 2017

Reclaiming Science:The JFK Conspiracy
JFK Blog Posts
JFK Calc Spreadsheet Database

Billy Lovelady and Buell Frazier both testified that Lovelady was standing on the STEPS at the Texas School Book Depository. But Doorman (Oswald) was standing on the TOP level.

THEREFORE LOVELADY COULD NOT BE DOORMAN! HE CANNOT STAND IN TWO PLACES AT THE SAME TIME!

Lovelady testified that he was standing on the STEPS. And Frazier said the same thing. If Lovelady was on the STEPS, then he could not be Doorman who was standing on the TOP Level. Q.E.D.
http://jfkassassination.net/russ/testimony/lovelady.htm

WARREN COMMISSION
Mr. LOVELADY – That’s on the second floor; so, I started going to the domino room where I generally went in to set down and eat and nobody was there and I happened to look on the outside and Mr. Shelley was standing outside with Miss Sarah Stanton, I believe her name is, and I said, “Well, I’ll go out there and talk with them, sit down and eat my lunch out there, set on the STEPS, so I went out there”.
Mr. BALL – You ate your lunch on the STEPS? (Ball is perplexed)
Mr. LOVELADY – YES, SIR.
Mr. BALL – Who was with you?
Mr. LOVELADY – Bill Shelley and Sarah Stanton, and right BEHIND ME me… (did Ball cut him off?)
Mr. BALL – What was that last name?
Mr. LOVELADY – Stanton.
Mr. BALL – What is the first name?
Mr. LOVELADY – Bill Shelley.
Mr. BALL – And Stanton’s first name?
Mr. LOVELADY – Miss Sarah Stanton.
Mr. BALL – Did you stay on the STEPS? (was Ball trying to get Lovelady to say he was on the TOP level?)
Mr. LOVELADY – Yes. (BNL said he stayed on the STEPS)
Mr. BALL – Were you there when the President’s motorcade went by?
Mr. LOVELADY – Right. (Once again, Lovelady does not change his location – he stayed on the STEPS).

WARREN COMMISSION
Frazier testified that Lovelady was standing two or three STEPS BELOW him. http://jfkassassination.net/russ/testimony/frazierb1.htm

Mr. BALL – We have got a picture taken the day of the parade and it shows the President’s car going by. Now, take a look at that picture. Can you see your picture any place there?
Mr. FRAZIER – No, sir; I don’t, because I was back up in this more or less black area here.
Mr. BALL – I see.
Mr. FRAZIER – Because Billy, like I say, is TWO or THREE STEPS down in FRONT of me. (why would he say this if it were not true?)
Mr. BALL – Do you recognize this fellow?
Mr. FRAZIER – That is Billy, that is Billy Lovelady.
Mr. BALL – Billy? (Ball must be concerned with Frazier’s answer. He does not want him to say Lovelady is on the STEPS, so he uses guile to twist the testimony).
Mr. FRAZIER – Right
Mr. BALL – Let’s take a marker and make an arrow down that way. That mark is Billy Lovelady?
Mr. FRAZIER – Right.
Mr. BALL – That is where you told us you were standing a moment ago.
Mr. FRAZIER – Right.
Mr. BALL – In FRONT of you to the right over to the wall?
Mr. FRAZIER – Yes. (Frazier once again says that Lovelady was standing BELOW him on the STEPS, but then UNWITTINGLY CONTRADICTS himself by pointing to Doorman who was on the TOP Level).
Mr. BALL – Is this a Commission exhibit?
We will make this a Commission Exhibit No. 369.
(The document referred to was marked Commission Exhibit No. 369 for identification.)

FRAZIER AT THE CLAY SHAW/GARRISON TRIAL 2/13/69:
http://www.jfk-online.com/fraziershaw.html
MR. ALCOCK: Can you see the spot where you were situated when the presidential motorcade came by?
FRAZIER: Yes,sir, I can.
MR. ALCOCK: Will you take this symbol and place it at that location where you were standing?
MR. ALCOCK: Mr.Frazier, do you recall who you were with during the presidential motorcade?
FRAZIER: Yes, sir, I can. When I was standing there at the TOP of the stairs, I was standing there by a heavyset lady who worked up in our office, her name is Sara, I forget her last name, but she was standing right there BESIDE me when we watched the motorcade.
MR. ALCOCK: Do you recall anyone else who may have been with you?
FRAZIER: Right down in FRONT of me at the BOTTOM OF THE STEPS my foreman Bill Shelley and BILLY LOVELADY were standing there.

1986 OSWALD MOCK TRIAL
GERRY SPENCE: “You recall that 23 years later that BNL was standing in FRONT of you. About 4 STEPS in FRONT of you. Is that correct?”
FRAZIER: Yes it is..
https://richardcharnin.wordpress.com/2014/08/07/10851/

Scientific Evidence Confirms the Testimony
Judyth Baker: Pixel Analysis
https://richardcharnin.wordpress.com/2014/07/27/jfk-judyth-baker-pixel-analysis-of-altgens6-photo-proves-oswald-is-doorman/

Larry Rivera: Photogrammetric and Overlay analysis
https://oswaldinthedoorway.blogspot.com/2015/10/the-following-paper-was-written-and.html

 

2016 Election: Introduction to my upcoming book

Richard Charnin
Nov. 13, 2017

77 Billion to One: 2016 Election Fraud
Matrix of Deceit: Forcing Pre-election and Exit Polls to Match Fraudulent Vote Counts
Proving Election Fraud: Phantom Voters, Uncounted Votes and the National Poll
LINKS TO  POSTS

Introduction

The mainstream media pundits claim that Clinton won the primary and presidential election by three million votes. It’s a myth. The pundits fail to consider the FACT that the recorded vote is ALWAYS fraudulent. A True Vote Model analysis indicates Trump won the popular as well as the electoral vote.

The pundits always assume that the recorded vote is accurate but never consider the fraud factor. The historical statistical evidence is conclusive: every election is fraudulent. The recorded vote is NEVER equal to the true vote.

The establishment-dominated media was in the tank for Hillary Clinton in the primary and general elections.

The claim that Clinton won the popular vote is quoted ad nauseam in the media, academia and by corrupt politicians. They persist in promoting the fully discredited meme of Russian “hackers” stealing the election from Clinton. But there is not one iota of proof that the Russians had anything to do with it. Included in the appendix are two memos from the Veteran Intelligence Professional for Sanity (VIPS) to Obama and Trump which prove that the Russians did not hack the vote. Election Fraud is always an inside job.

Sanders and Trump drew much larger crowds than Clinton. They won the unscientific online polls by large margins. Trump’s Republican base was solid. Clinton’s Democratic base was fractured by defecting Sanders voters.

Millions of Sanders primary voters stayed home or voted for Jill Stein or Donald Trump. Trump won Independents by a solid majority (at least 8% higher than Clinton). There was a surge of late deciders to Trump after Labor Day.

Former interim Democratic National Committee chairwoman Donna Brazile delivered a bombshell in her book “Hacked”. She claimed that the Hillary Clinton campaign seized control of the Democratic Party as far back as August 2015. Well, this was not a bombshell to researchers who have presented massive evidence that the primary was rigged from Day One.

In ‘77 Billion to One: 2016 Election Fraud’, I provided mathematical evidence that the primary was rigged for Clinton. The exit poll discrepancies were in one direction only; they showed that Sanders did consistently better in the polls than the recorded vote. It was solid proof that the primaries were rigged.

But just because the unadjusted exit polls were quite accurate in prior elections and the 2016 primary does not mean they reflected the true vote in the presidential election.

Six major media corporations (the National Election Pool) fund exit pollster Edison Research. The pollsters had to show that Clinton won the pre-election and unadjusted polls to lend credence to her 2.8 million recorded popular vote margin.

In 2008, 2012 and 2016 my pre-election models exactly forecast the recorded electoral votes. Trump was projected to win 306 recorded electoral votes based on adjustments made to nine final pre-election polls. It also forecast that he would have had 350 electoral votes in a fraud-free election.

Democratic Party-ID was over-weighted in the pre-election and exit polls at the expense of Independents. A post-election exit poll analysis based on the Gallup voter affiliation survey conducted the week prior to the election confirmed the forecast. But Trump did much better than the unadjusted exit polls indicated. The Gallup survey showed that Independents comprised 41% of the electorate on Election Day, with 31% Democrats and 28% Republicans.

 

Las Vegas suspicious deaths: a probability analysis

Las Vegas suspicious deaths: a probability analysis

Richard Charnin
Nov.4, 2017
Updated Nov.6, 2017

77 Billion to One: 2016 Election Fraud
Matrix of Deceit: Forcing Pre-election and Exit Polls to Match Fraudulent Vote Counts
Proving Election Fraud: Phantom Voters, Uncounted Votes and the National Poll
LINKS TO  POSTS

There have been seven suspicious deaths one month after the Las Vegas attacks .  Were they murdered? We don’t know. Therefore,  we calculate the probability over a range from 1 to 7  homicides.  http://yournewswire.com/lawyer-vegas-aldean-dead/

Given
1) N random individuals,
2) n homicides
3) R annual homicide rate
4) T years
E= N*R*T is the expected number of homicides

The probability of at least n homicides in T years is
P = 1- poisson (n-1, E, true)

Given N = 22,000 concert attendees, the probability of at least 7 homicides in one month is P = 1 in 100 billion.

These factors are used in the calculation:
N = 22,000 to 500,000  individuals (universe)
T = 1 month = .083 years
R =0.00005 (1 in 20,000)

For n = 7 homicides and
N= 22,000: P =1 in 100 billion (22,000 concert attendees)
N= 30,000: P= 1 in 11.8 billion
N= 50,000: P= 1 in 355 million
N= 100,000: P= 1 in 3.3 million
N= 200,000: P= 1 in 37 thousand
N= 500,000: P= 1 in 178

The following table displays probabilities of 1, 2, 3, 4 and 7 homicides for groups ranging from N= 22,000 to 500,000.

Probability of at least n homicides in one month in a random group of N 

n	N=22k	N=30k	N=50k	N=100k	N=200k	N= 500k
1	8.8%	11.8%	18.8%	34.1%	56.5%	87.6%
2	0.40%	0.72%	1.9%	6.6%	20.3%	61.6%
3	0.01%	0.03%	0.13%	0.89%	5.2%	34.6%
4	0.00%	0.00%	0.01%	0.09%	1.0%	15.8%
7	0.00%	0.00%	0.00%	0.00%	0.00%	0.56%
		Probability
		1 in
7	100 billion	11.8 billion	355 mil	3.3 mil	37,000	178

Orville Almon (1) was the lawyer representing the Route 91 music festival and Jason Aldean, the singer onstage when the shooting began. Almon’s death was described by local Nashville media as “seizure during sleep”.

Kymberley Suchomel (2) of Apple Valley, California, a shooting survivor who was found dead in her home hours after her husband left for work.

Danny Contreras (3) was a Las Vegas local who had been contradicting the official narrative that only one shooter fired on the Route 91 Harvest festival crowd.

Dennis (4) and Lorraine Carver (5), a Las Vegas couple who survived the  shooting, were engulfed by flames in their car just meters from their home. They were speaking out about the lies being pushed in the official narrative.

John Beilman (6) killed himself and his disabled daughter (7) in an apparent murder-suicide. He was wanted by federal agents following the discovery of a communications device in the hotel room of the officially identified shooter Stephen Paddock.

 

Deaths of Dealey Plaza JFK Witnesses: A Probability Analysis

Richard Charnin
Oct. 30, 2017

 Reclaiming Science:The JFK Conspiracy.

JFK Blog Posts
JFK Calc Spreadsheet Database
Tables and Graphs

The 1977 House Select Committee on Assassination (HSCA) claimed that the London Sunday Times actuary’s probability calculation of 18 material witnesses deaths (13 unnatural) in the three years following the assassination (1 in 100,000 trillion) was invalid. The HSCA claimed that it was impossible to calculate the probability because the witness universe was unknown. This canard was essential to the coverup. No one did the math until I confirmed the actuary’s calculation in 2003

But there were many definable witness groups. Let’s take a look at the 600 estimated Dealey Plaza witnesses of whom 28 died suspiciously (14 were ruled unnatural: 5 homicides, 7 accidents, 2 suicides).

Assuming  600 Dealey Plaza witnesses, the probability of at least 14 ruled unnatural deaths during the period 1963-1978 is P = 1- poisson (13, 1.727, true) = 1 in 207 million.

But the nine accidents and suicides were likely homicides.
The probability of at least 14 homicides for 400, 600 and 1000 witnesses:
400: P= 1 in 750 trillion
600: P =1- poisson (13, 0.864,  true) = 1 in 3.7 trillion
1000: P = 1 in 4.8 billion

Sixteen Dealey Plaza witnesses testified at the Warren Commission, 3 were sought to testify at the Garrison trial, 3 at the Church Senate hearings and 3 at the House Select Committee on Assassinations (HSCA).

JFK witness unnatural deaths probabilities have been posted for the following groups: Warren Commission, London Times actuary, Garrison/ Shaw trial, Church Senate Hearings, HSCA, Simkin Educational Forum, JFK-related 1400+ witness “Who’s Who in the JFK Assassination”.

1 6311 Lee Harvey Oswald
2 6311 J.D. Tippit
3 6512 William Whaley
4 6606 Frank Martin
5 6608 Lee Bowers
6 6611 James Worrell
7 6701 Jack Ruby
8 6901 Charles Mentesana
9 6901 Buddy Walthers
10 7001 Merriman Smith
11 7008 Bill Decker
12 7101 Mac Wallace
13 7109 Roscoe White
14 7109 Cliff Carter
15 7309 Thomas E. Davis
16 7402 J.A. Milteer
17 7501 Allen Sweatt
18 7502 Ira (Jack) Beers
19 7505 Roger Craig
20 7509 Earl Cabell
21 7604 James Chaney
22 7608 Johnny Roselli
23 7703 Charles Nicoletti
24 7707 Ken O’Donnell
25 7801 Clint “Lummie” Lewis
26 7805 David Morales
27 7901 Billy Lovelady
28 8403 Roy Kellerman

Dealey Plaza witnesses: https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdDFSU3NVd29xWWNyekd2X1ZJYllKTnc#gid=79

https://docs.google.com/spreadsheets/d/1FmXudDf6pqisxq_mepIC6iuG47RkDskPDWzQ9L7Lykw/edit#gid=1

https://richardcharnin.wordpress.com/2013/08/30/jfk-calc-a-spreadsheetdatabase-of-mysterious-witness-deaths/

https://richardcharnin.wordpress.com/2013/04/24/jfk-assassination-a-probability-analysis-of-warren-commission-witness-unnatural-deaths/

https://richardcharnin.wordpress.com/2014/01/27/jfk-a-closer-look-at-the-convenient-deaths-of-warren-commission-witnesses/

https://richardcharnin.wordpress.com/2013/10/14/jfk-witness-deaths-graphical-proof-of-a-conspiracy/

https://richardcharnin.wordpress.com/2013/10/01/jfk-witness-deaths-7-fbi-officials-due-to-testify-at-hsca/

https://richardcharnin.wordpress.com/2013/08/18/jfk-witnesses-called-to-testify-actual-vs-expected-unnatural-deaths-1964-1977/

Quick JFK Witness death Calculator:
https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdDFSU3NVd29xWWNyekd2X1ZJYllKTnc#gid=78

 

Race: 2016 National Exit Poll vs. Census

Richard Charnin
Oct. 22, 2017

77 Billion to One: 2016 Election Fraud
Matrix of Deceit: Forcing Pre-election and Exit Polls to Match Fraudulent Vote Counts
Proving Election Fraud: Phantom Voters, Uncounted Votes and the National Poll
LINKS TO  POSTS

The National Exit Poll (NEP) shows Trump winning white voters by 57-37%. Clinton had 89% of Blacks and 66% of Latinos. As always, the NEP was forced to closely match the recorded vote.

Clinton won the recorded vote by 2.8 million votes (48.3-46.2%).
Clinton won the NEP by 2.2 million votes (47.9-46.3%).
Clinton won the Census-adjusted NEP by 800,000 votes (47.4-46.8%).

Trump wins by 3.6 million votes (48.6-46.1%) after adjusting the recorded vote for illegals, disenfranchised and machine vote flips.

Compare the Census (0.3% margin of error for votes cast) to the NEP.  Which is closer to the truth?

Race Census MoE NEP
White 73.30% (0.4%) 71%
Black 12.45% (1.1%) 12%
Latino 9.22% (1.5%) 11%
Asian 3.67% (1.9%) 4%
Other  1.34% …… 2%

National Exit Poll (forced to match the recorded vote)  http://www.cnn.com/election/results/exit-polls/national/president

NEP	2016	Clinton	Trump	Other
White	71.0%	37%	57%	6%
Black	12.0%	89%	8%	3%
Latino	11.0%	66%	28%	6%
Asian	4.0%	65%	29%	8%
Other	2.0%	56%	36%	8%
Calculated	100.0%	47.93%	46.31%	5.76%
Recorded	136,216	48.25%	46.17%	5.58%

Recorded votes vs. Census votes cast

According to the 2016 Census, 137.5 million votes were cast (0.3% margin of error).
136.2 million votes were recorded.
Therefore there were 1.3 million uncounted votes
Clinton won the recorded vote by 2.8 million: 65.7-62.9 (48.3-46.2%)

 Scenario I:  Estimated Adjustments to the Recorded Vote

1- According to Greg Palast, 1 million voters were disenfranchised due to Cross-check.
2- Assume 1.3 million additional voters were disenfranchised.
3- Researchers claim there were at least 1 million illegal voters.
Matching the Census 137.5 million: There were 1.3 million uncounted votes.

Assumptions:
1) Clinton had 80% of illegal and disenfranchised voters
2) 3.2 million votes (4.8%) were flipped on voting machines from Trump to Clinton.
3) 0.6 million votes were flipped from third parties to Clinton.

Trump is a winner by 3.3 million votes (48.2-45.8%}

	Votes	Clinton	Trump	Other
Illegal	1.0	80%	15%	5%
Disenfran.	2.3	80%	15%	5%
Net Vote Flip	4.0	5%	80%	15%
Adjustment	Total	Clinton	Trump	Other	Margin
Recorded	136.22	65.72	62.89	7.61	2.83
Share		48.25%	46.17%	5.59%	2.08%
Illegal	-1.0	-0.80	-0.15	-0.05	-0.65
Disenfran.	2.3	1.84	0.35	0.12	1.50
Net Vote Flip	0.0	-3.80	3.20	0.60	-7.00
AdjVote	137.52	62.96	66.28	8.27	3.33
Share		45.78%	48.20%	6.02%	2.42%

NEP Race Demographic: Census and share of Whites matched to the adjusted  vote

Race	Census Pct	Clinton	Trump	Other
White	73.30%	34.7%	58.9%	6.4%
Black	12.45%	89%	8%	3%
Latino	9.22%	66%	28%	6%
Asian	3.67%	65%	27%	8%
Other	1.36%	56%	36%	8%
Adj Share	100.0%	45.78%	48.20%	6.02%
Votes	137.52	62.96	66.28	8.27
Recorded	100.0%	48.25%	46.17%	5.59%

Sensitivity Analysis: Adjusted NEP

		Illegal to Clinton
Vote Flip	70.0%	80.0%	90.0%
to Clinton		Trump
3.0%	48.2%	48.3%	48.3%
5.0%	48.1%	48.20%	48.3%
7.0%	48.1%	48.1%	48.2%
		Clinton
3.0%	45.8%	45.7%	45.7%
5.0%	45.9%	45.78%	45.7%
7.0%	45.9%	45.8%	45.8%
		Trump Vote
3.0%	66.26	66.36	66.46
5.0%	66.18	66.28	66.38
7.0%	66.10	66.20	66.30
		Margin
3.0%	3.28	3.48	3.68
5.0%	3.12	3.33	3.52
7.0%	2.96	3.17	3.36

Sensitivity Analysis: Trump shares of whites and blacks

Census			% of Whites
	56.9%	57.9%	58.86%	59.86%	60.86%
% of Blacks			Trump
10%	46.98%	47.72%	48.45%	49.18%	49.92%
9%	46.86%	47.59%	48.33%	49.06%	49.79%
8%	46.73%	47.47%	48.20%	48.93%	49.67%
7%	46.61%	47.34%	48.08%	48.81%	49.54%
6%	46.49%	47.22%	47.95%	48.68%	49.42%
			Clinton
10%	47.00%	46.27%	45.53%	44.80%	44.07%
9%	47.12%	46.39%	45.66%	44.93%	44.19%
8%	47.25%	46.52%	45.78%	45.05%	44.32%
7%	47.37%	46.64%	45.91%	45.17%	44.44%
6%	47.50%	46.77%	46.03%	45.30%	44.57%
		Share	Margin
10%	-0.02%	1.45%	2.92%	4.38%	5.85%
9%	-0.27%	1.20%	2.67%	4.13%	5.60%
8%	-0.51%	0.95%	2.42%	3.88%	5.35%
7%	-0.76%	0.70%	2.17%	3.63%	5.10%
6%	-1.01%	0.45%	1.92%	3.39%	4.85%
		Vote	Margin
10%	-0.02	1.99	4.01	6.03	8.04
9%	-0.36	1.65	3.67	5.68	7.70
8%	-0.71	1.31	3.33	5.34	7.36
7%	-1.05	0.97	2.98	5.00	7.01
6%	-1.39	0.62	2.64	4.66	6.67

Scenario II:  Adjustments to the Recorded Vote

Base case assumptions: Illegals, disenfranchised voters and machine vote flips

1) Illegals: 80% of 1 million for Clinton
2) Uncounted: 80% of 7 million disenfranchised and cross-checked for Clinton
3) Voting machines: 4.1 million (net) Trump votes  and 0.5 million third-party votes flipped to Clinton

 Adjustments to the Recorded Vote

Assumption	Votes	to Clinton	Trump	Other
Illegal	1.0	80%	15%	5%
Disinfran.	7.0	80%	15%	5%
Net Vote Flip	5.0	8%	82%	10%
	Total	Clinton	Trump	Other	Margin
Recorded	136.22	65.72	62.89	7.61	2.83
		48.25%	46.17%	5.59%	2.08%
Illegal	-1.0	-0.80	-0.15	-0.05
Disenfran.	7.0	5.60	1.05	0.35
Net Vote Flip	0.0	-4.60	4.10	0.50
True Vote	142.22	65.92	67.89	8.41	1.97
Share		46.35%	47.74%	5.91%	1.39%

		Illegal to Clinton
Vote Flip	75.0%	80.0%	85.0%
to Clinton		Trump Vote
6.0%	67.94	67.99	68.04
8.0%	67.84	67.89	67.94
10.0%	67.74	67.79	67.84
Flip		Trump
6.0%	47.77%	47.81%	47.84%
8.0%	47.70%	47.74%	47.77%
10.0%	47.63%	47.67%	47.70%
Flip		Clinton
6.0%	46.32%	46.28%	46.25%
8.0%	46.39%	46.35%	46.32%
10.0%	46.46%	46.42%	46.39%
Flip		Margin
6.0%	2.07	2.17	2.27
8.0%	1.87	1.97	2.07
10.0%	1.67	1.77	1.87

 

 

 

Sensitivity of 2016 Electoral and Popular Vote to Registered Voter Turnout

Sensitivity of 2016 Electoral and Popular Vote to Registered Voter Turnout

Richard Charnin
Oct.4, 2017

77 Billion to One: 2016 Election Fraud
Matrix of Deceit: Forcing Pre-election and Exit Polls to Match Fraudulent Vote Counts
Proving Election Fraud: Phantom Voters, Uncounted Votes and the National Poll
LINKS TO  POSTS

Eight voter turnout scenarios:
Trump wins the base case (86% Dem, 91% Rep) by 328-210 EV and 1.15 million votes.

Trump needs 89% Dem and 88% Rep turnout to match his 306 EV.
Clinton needs an implausible 92% Dem, 85% Rep turnout to tie Trump at 269 EV.
Clinton needs 93% Dem, 87% Rep turnout to win by 298-240 EV and 3.12 million.
Clinton needs 92.5% Dem, 84.5% Rep turnout to match her 2.8 million margin.

Trump vote margins are conservative since the calculations are based on state exit poll vote shares forced to match the recorded vote.

Sensitivity analysis (assume constant 87% Independent voter turnout)

Turnout Trump Votes (000) Vote Shares
Dem Rep EV Trump Clinton Margin Trump Clinton
0.85 0.92 332 64,647 62,885 1,762 47.5 46.2
0.86 0.91 328 64,347 63,195 1,152 47.2 46.4
0.87 0.90 321 64,047 63,505 542 47.0 46.6
0.88 0.89 315 63,747 63,815 -68 46.8 46.9
0.89 0.88 305 63,447 64,125 -678 46.6 47.1
0.90 0.87 289 63,147 64,435 -1,288 46.4 47.3
0.91 0.86 289 62,847 64,745 -1,899 46.1 47.5
0.92 0.85 269 62,546 65,055 -2,509 45.9 47.8
0.93 0.84 240 62,246 65,365 -3,119 45.7 48.0
0.94 0.83 240 61,946 65,675 -3,729 45.5 48.2

Given these facts:
– Census 2016 registered voter turnout of 87%.
– Gallup national voter affiliation (Party-ID) on Election Day:
(41% Independents, 31% Democrats and 28% Republicans)
– 28 exit poll states: vote shares forced to match recorded vote.
– 23 non-exit poll states recorded vote shares .

Assumptions: Bernie Sanders defectors…
– 5% of registered Democrats stayed home
– 4% voted for Jill Stein and 1% for Trump.
Then….
1. Adjusted Voter Turnout: 78.6% Dem, 91.6% Rep, 91.6% Ind
2. Adjusted Gallup Party-ID: 29.5% Dem, 29.1% Rep, 41.4% Ind
3. Gallup Party-ID calculated for each of the 28 exit polled states
4. Trump wins by 48.13-45.33% (3.81 million votes) with 332-206 EV

Since the 28 state exit polls were forced to match the recorded vote and likely inflated for Clinton (as were the 23 states not polled), Trump probably did better than indicated. 

…………..Clinton Trump
28 states 45.67% 47.67% Exit polls (Gallup/NEP-adjusted)
Votes…… 50,664 52,776

23 states 43.71% 50.40% Not exit polled. Assume recorded vote.
Votes…… 11,079 12,777

51 states 45.33% 48.13% (conservative- Trump may have done better)
Votes…… 61,744 65,554

https://docs.google.com/spreadsheets/d/1R9Y3ae2uyW8SUxVUnnOt9ZyvheAxa0fAhesAw_nhciM/edit#gid=857963642