2016 Primary Exit Polls Party-ID: forcing a match to the recorded vote
Richard Charnin
July 22, 2016
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
Democratic Primaries spread sheet
From TDMS Research: Democratic 2016 primaries
This analysis seeks to determine an approximate True Vote in the 2016 Democratic primaries.
Clinton led the average of the adjusted exit polls (which were forced to match the recorded vote) by 56.4-43.6%. But there has been a significant increase in Independents tn National Party-ID statistics since 2014.
2014 National Party-ID: Dem 40.5 – Rep 35.2 – Ind 24.2%
2-party mix: Dem 62.6 – Ind 37.4%
Note the 19% increase in Independents since 2014:
2016 National Party-ID: Dem 32 -Rep 25 -Ind 43%
2-party mix: Ind 57.3- Dem 42.7%.
Gallup poll Party preference trend: http://www.gallup.com/poll/15370/party-affiliation.aspx
In the exit polled primaries, Sanders had a 51.3% True Vote and a 43.6% Recorded Vote.
The calculation of the True Vote differs from the Recorded Vote (adjusted Exit poll) in just one aspect: Independent and Democratic weights are based on the actual 2016 Party-ID mix in which independents comprise 57.3% and Democrats just 42.7% of the two-party mix. Sanders vote shares are unchanged.
In other words the True Vote assumes that no voters were disenfranchised and/or removed from the rolls.
The 2016 Party-ID mix for each state is estimated by applying the equivalent proportional change from the 2014 National Party-ID to the 2016 National Party ID.
Let’s compare the the 2016 Party-ID effect on vote shares to the adjusted exit poll Party-ID weights in three primaries: NY, MA, WI.
Candidate vote shares are held constant. The only difference between the adjusted exit poll (which was forced to match the recorded vote) and the True Vote is the Party-ID mix. Sanders does significantly better when the 2016 Party-ID percentages are used.
View the exit polls here:
http://www.cnn.com/election/primaries/polls/NY/Dem
View the spreadsheet calculations here: https://docs.google.com/spreadsheets/d/1sGxtIofohrj3POpwq-85Id2_fYKgvgoWbPZacZw0XlY/edit#gid=610570359
NY Primary Exit poll matched to the recorded vote
| Party ID | Pct | Clinton | Sanders | |
| Dem | 83% | 62% | 38% | |
| Ind | 14% | 28% | 72% | |
| Vote | 97% | 55.4% | 41.6% | |
| 2-party | 100% | 57.1% | 42.9% |
NY True Vote
| 2016 | PartyID | Clinton | Sanders |
| Dem | 47% | 62% | 38% |
| Ind | 53% | 28% | 72% |
| 2-party | 44.0% | 56.0% |
MA Primary Exit poll matched to the recorded vote
| Party ID | Pct | Clinton | Sanders | |
| Dem | 65% | 60% | 40% | |
| Ind | 33% | 33% | 66% | |
| Vote | 98% | 49.9% | 47.8% | |
| 2-party | 100% | 51.1% | 48.9% |
MA True Vote
| 2016 | PartyID | Clinton | Sanders |
| Dem | 30% | 60% | 40% |
| Ind | 70% | 33% | 66% |
| 2-party | 41.2 | 58.8% |
WI Primary Exit poll matched to the recorded vote
| PartyID | Clinton | Sanders | |
| Dem | 71% | 50% | 50% |
| Ind | 27% | 28% | 72% |
| Vote | 43.1% | 54.9% | |
| 2-party | 43.9% | 56.1% |
WI True Vote
| 2016 | PartyID | Clinton | Sanders |
| Dem | 48% | 50% | 50% |
| Ind | 52% | 28% | 72% |
| 2-party | 38.5% | 61.5% |
| Sanders | True Vote (2016 Party-ID) | Exit Poll (Recorded) |
| AL | 30.6% | 19.8% |
| AR | 39.8% | 31.0% |
| CT | 62.3% | 47.4% |
| FL | 44.1% | 34.1% |
| GA | 36.6% | 28.3% |
| IA | 60.1% | 49.9% |
| IL | 55.9% | 49.1% |
| IN | 61.4% | 52.4% |
| MA | 58.1% | 49.3% |
| MD | 40.0% | 34.5% |
| MI | 56.8% | 50.8% |
| MO | 56.9% | 49.9% |
| MS | 24.3% | 16.6% |
| NC | 48.1% | 42.8% |
| NH | 67.3% | 61.4% |
| NV | 58.5% | 47.3% |
| NY | 56.0% | 42.1% |
| OH | 51.9% | 43.1% |
| OK | 57.1% | 55.5% |
| PA | 54.0% | 43.9% |
| SC | 38.2% | 26.1% |
| TN | 42.4% | 32.9% |
| TX | 40.9% | 33.7% |
| VA | 45.0% | 35.4% |
| VT | 87.2% | 86.3% |
| WI | 61.5% | 56.7% |
| WV | 53.0% | 51.4% |
Richard Charnin's Blog 
Don Deppeller
July 22, 2016 at 8:20 pm
Well done as usual, though I’m not clear on whether the percentage of independents remains uniform across all the states in your analysis or not, especially considering how it changes in time from 2014 to 2016.
jeffreyaackerman
July 23, 2016 at 7:17 am
NY true vote is broken up by party-ID as either Dem. or Ind. I thought that independents were not allowed to vote in the dem. primary.
Richard Charnin
July 23, 2016 at 8:10 pm
The True Vote assumes all who would have voted but were disenfranchised or shut out or whose whose votes were flipped.
Check the NY exit poll Party-ID: http://www.cnn.com/election/primaries/polls/NY/Dem
Adriana A.
July 25, 2016 at 4:09 am
I think it depends on state. Here in Arizona, if you registered as an Independent, you could not vote in the primary. But is that true for all states?
Richard Charnin
July 25, 2016 at 7:49 pm
No, there are open states where Independents can vote.
jeffreyaackerman
July 26, 2016 at 6:30 am
In N.Y. state only democrats could vote. I heard there was a March 25 deadline to register as a democrat. I made sure to register before the March 25 as a democrat, but the rule here has for many years been that when you register, you are not registering for the upcoming election, but rather for the future election cycles. So I was not able to vote for Bernie.
rockethead
July 27, 2016 at 8:30 pm
This is interesting, but why wouldn’t the Party ID for people at the Democratic primary skew towards a higher percentage of Democrats?
Is it because there has been a strong shift toward Independent identification, but no similar change in exit polling?
Is it because the Sanders turnout should be heavily skewed toward Independent voters?
Richard Charnin
July 28, 2016 at 4:18 pm
Vote shares and Party-ID mix of Dems and Repubs in the exit polls are adjusted to match the recorded vote to within 0.5%.
But the mix does not reflect the actual composition of Democrats and Independents.
Using 2016 party-preference mix based on the Gallup survey, Bernie does significantly better.
We can conclude that a combination of closed and rigged (strip and flip) primaries cost Bernie the nomination.
danisms
August 22, 2016 at 1:30 pm
Richard could there be a virus attached, since I’ve posted this article I cannot share it in my Facebook Bernie Sanders groups. Has anyone else experienced anything weird?
Richard Charnin
August 22, 2016 at 1:43 pm
I have not heard of any.