A Graphical Debunking of Election Myths (TIA)

[WillE]

A Graphical Debunking of Election Myths
TruthIsAll

http://www.geocities.com/electionmodel/DebunkingElectionMyths.pdf

MYTHS
1 The recorded (official) vote is sacrosanct and equal to the True Vote.
2 Bush 48% approval is not a valid indicator that the election was stolen.
3 2004 pre-election polls did not match the exit polls.
4 2004 Election model projection assumptions were wrong.
5 Bush led the 2004 pre-election polls.

6 Exit polls are not random samples.
7 Reluctant Bush Responder (rBr) explains the 2004 exit poll discrepancies.
8 Bush won by increasing his vote share in Democratic strongholds (Urban Legend).
9 Swing vs. Red-shift: No correlation "kills the fraud argument".
10 False Recall explains the 43/37 Bush/Gore returning voter mix in the Final NEP.

11 Exit poll discrepancies were not due to voting machines/methods.
12 Assumptions used in calculating the True Vote were invalid.
13 Bush won the late voters the early exit polls missed.
14 Mid-term Generic polls are not a good predictor.
15 Hillary and Obama split the popular vote in the primaries.
16 Obama won by 9.5 million votes with a 52.9% share.

[fasttense]

1. Very interesting.

In order for the bush to have won reelection he had to bring in more than 100% of Republicon registered voters.

[WillE]

6. Not quite...read it closely

According to the Final 2004 National Exit Poll, in order for Bush to have matched his recorded vote, he needed more than 100% of living Bush 2000 voters.

Let's review the facts.

In 2000, Bush had 50.5 million recorded votes.
Very close to 2.5 million Bush voters died prior to the 2004 election.
Of the 48 million still living, approximately 46 million (96%) returned to vote in 2004.

According to the Final 2004 National Exit Poll, 43% (52.6 million) of the 122.3 million who voted in 2004 were returning Bush 2000 voters.

Therefore, there were 6.6 million phantom Bush voters (52.6-46).
Bush 2000 live voter turnout was 114% (52.6/46)

THE KEY POINT: The National Exit Poll was FORCED TO MATCH the RECORDED VOTE with an IMPOSSIBLE (52.6 million) returning Bush voters.

Therefore the official recorded vote had to be IMPOSSIBLE.

KERRY WON IN A LANDSLIDE.

SMOKING GUN!
SMOKING GUN!
SMOKING GUN!

[slay]

2. TIA always had/has good analysis of our rigged elections of 2002 and 2004

I mean really great stuff.

http://www.truthisall.net/

[On the Road]

3. Well, Unfortunately, I Would Agree with the Following Statement

Fact - Bush was the ONLY incumbent with approval below 50% to win re-election!

Then there's this:

Myth - Exit polls are not random samples; the margin of error is high. Fact - Exit pollster notes to the National Exit Poll (13047 respondents) indicate that voters were RANDOMLY SELECTED as they exited the voting booths. Fact - the pollsters STATED that the National Exit Poll MoE was 1.0%.

Whatever the pollsters claimed, they were not at the sites and have no way of verifying that the sample was random or not. Even if the individual polling station workers followed instructions precisely and selected voters at random, the response is not guaranteed to be random. The so-called MoE is based only on random variation, not systematic variation from response variables or other sources. TIA is actually counting on this fact to reach his conclusion, but only accepts one source of systematic variance, namely fraud.

It is possible to go on. TIA has been going on for almost five years now.

[WillE]

4. On the Road again...

You're right. Bush won the recorded vote - but not the True Vote.

True, TIA has been going on for five years now.
It's a good thing.

You must be on the road often.
On the other hand, maybe that's why you haven't posted randomly on ER before this.

[WillE]

5. I just went out on the road...and found you.

On the other hand, it wasn't too hard.

Anyway, here's a link.
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x386293

NH is a beautiful state.
It's just that thy can't seem to get the vote counts right.
Not in '88. Remember the GOP primary? Poppy came from nowhere to win
Not in '00. Gore "lost"; if the 3 EV weren't stolen he would have been prez.
Not in '04. Landslide denied

And surely not in the '08 primary.
Clinton won the machine counts. Obama won the hand-counts.

[Sancho]

7. Interesting, and one part of the collection of evidence...

I realize that "proving" things are difficult with post hoc parametric statistics. My view of TIA is that he is the "John Snow" of election pollsters ( http://en.wikipedia.org/wiki/John_Snow_(physician) ). Sometimes the "scientific" evidence is less compelling than the big picture. My original reason for joining DU and other groups was entirely anecdotal:

I witnessed a voting machine switching votes in a Clearwater, Fl election (for a Senate race: Betty Castor's votes would not record on the machine) and my wife and I both were asked to leave the polling area when we protested on the spot.
I saw unprotected machines stored at a local Florida library and had a battle with an election supervisor over the possible hacking of the machines.
I believe the American Statistical Association article in Chance magazine that concluded that the GOP actually lost a Senate race is correct (Christine Jennings), since it was an identical conclusion to my own analysis of the suspicious undervote.
I have witnessed missing mailed ballots and rejected mailed ballots on a number of occasions in Pinellas County, and follow up by voters is difficult after the official count is announced.
Recounts in Florida are a joke as we know - and the number issues similar to hanging chads, butterfly ballots, trashed voter machine tapes, and mail ballot mistakes happen so often that it would tie up courts for decades if the real number of complaints were followed up.
We have often seen reports of manipulation of voter registration rolls in Florida, and I have also seen illegal challenges to legitimate voters at the entrance to voting stations.
I have seen "impossible" results where tax referendums pass, Democratic local candidates win, and most voters appear to be registered Democratic voters, yet a GOP candidate will mysteriously win a federal election in some precincts (along with weird undervote or machine totals).
We have obviously had people come forward in Florida with accusations of writing software to hack the voting process.

If my admittedly personal observations are typical or repeated around the state and country (like in Oh, PA, and other reported trouble spots), you might infer that there is both polling and eyewitness evidence to suspect manipulated elections: the motive and opportunity were certainly there. Proving a legal case is pretty difficult without parallel elections, complete audits of voting and tabulating machines, or confessions of a guilty person. Also, some election supervisors and the Secretary of State (remember Harris!) are often intent on preventing fair observation of the process or gathering of evidence.

It is clear that researchers have demonstrated how machines can be hacked. If I were a person creating a formula to rig an election and had access to voting machine programming, I would set the parameters based on previously known voter registration variables and randomly switch or drop SOME votes on SOME machines and carefully STOP manipulating within a given machine or precinct if the predicted error variance was approached - thereby avoiding poll or statistical evidence that would clearly indicate manipulation. In other words, the results would almost always be within likely poll confidence intervals. If the election becomes so overwhelming that a candidate wins despite manipulation, the winner is correctly determined, but the results might appear skewed as TIA often reports as "recorded vs true" vote.

This does not prove what happened in several of the last three election cycles, but it is certainly logical to me given my own observations at local elections.