Cheating Allegations...The Truth about Bush 2000

[Monty]

All I'm saying is that (ignoring the Buchanan-Palm Beach situation) more people went to the voting booth on November 7, 2000 to vote for Al Gore than for George W. Bush. 

[J. J.]

All I'm saying is that (ignoring the Buchanan-Palm Beach situation) more people went to the voting booth on November 7, 2000 to vote for Al Gore than for George W. Bush. 

Well, duh, Gore won a plurality of the popular vote.

[Monty]

I meant in Florida.

[The Vorlon]

Do you mean this study, Vorlon?

"The study's key result: When the consortium tried to simulate a recount of all uncounted ballots statewide using six different standards for what constituted a vote, under each scenario <b> they found enough new votes to have narrowly given the Florida election--and by extension the presidency--to Al Gore." </b>

http://www.fair.org/extra/0201/fla-recount.html

Basically what I said originally...

Yes there are counting scenarios where if you bend the rules one way Gore wins, and other scenarios if you bend the rules the other way Bush wins by a larger margin than he did.

Obviously out of 6 million votes even tiny changes can reverse a 537 vote margin

There are a vast number of counting standards that are is some sense reasonable.

For example, the voter has a chance to look at their punchcard ballot before they turn it in, and a person who cares about their vote certainly can take 15 seconds to check that all punches are clean and that the "chad" is fully detatched.

To ask a voter to do this is not "unreasonable" in any way, and if we define a valid ballot as one that had been properly punched so that a machine can read it, Bush won.

On the other hand, if we take another point of view and try to devine voter intent, ballots more creatively marked get included, and Gore would likely have won.

Neither of these standards is either definitively right, or difinitively wrong, they are just different standards.  So which standard do you use ?

Federal  law, quite correctly, requires that the rules for voting cannot change after the vote has taken place (This law was part of the Voting Rights act and was intended to prevent the states in the Deep South from retroactively changing rules to disenfranchise Blacks)

Under the established rules and counting standards that existed on election Day, Bush - probably - won.

If the Democrats running Palm Beach had designed a ballot that folks understood, Gore might have had 10,000 more votes and been President.

If the Networks had not called Florida while the panhandle was still voting, Bush would have had 10,000 more votes and their never would have been this controversy...

To quote Yoggi Berri

"If things were different, they wouldn't be the same

To me, however, the KEY finding of the NORC ballot report "recount" was this:

Although trained to produce accurate, impartial reports, the NORC investigators are human and prone to human judgment and error. In particular, NORC discovered that male investigators were more likely to record marks on ballots than women. NORC also found a slight but statistically significant relationship between candidate marks and the investigators' party affiliation.

Most importantly, there is no guarantee that the judgments of the NORC investigators would have matched those of local election boards had the recount been permitted to proceed under any scenario."

So if Men counted Palm beach Gore wins, but if women counted it Bush wins?

If Dems count Gore wins, but if GOPers count Bush wins?

Human beings have bias, it's just the way it is.

537 votes out of 6 million is 0.00009

There is no human being on the planet that is 99.99991% fair and impartial, - and remember these ballots were being counted and supervised by partisan elected Dems and GOPers. 

If trained impartial investigatorts have a "statistically signifigant" bias, what is the bias of partisan elected officials?

I am sure it is more than 0.00009.

If we assume the Dems counting the ballots were 99.99990 percent fair versus 99.99991 fair, it would have been enough for even this tiny bias to flip the presidency.

The only solution IMHO, is to take the humans totally out of the parts of the count that require any judgement.

I am in Canada and we just had a mayoral election that was a good model for how to do things IMHO.

1) - You get an optical scan ballot.
2) - After you have marked it, the clerk runs it through the machine to scan it.
3) - If it scans you are done, if it does not scan you get to correct or replace your ballot if you want to.
4) The only ballots that count are the ones the machine can read - if the machine can't read it, it's a spoiled ballot. - You can write in 2 inch high magic marker "I vote for Smith!" and it still doesn't count - it scans, or it's doesn't.

Done - finished - no arguments & no human "judgement" to argue about.

[Monty]

Human beings have bias, it's just the way it is.

Machines may not have "bias", but they do have error rates.  It's not always the voter's fault if the machine fails to read their ballot.  Surely, your copier, printer, etc. has not always functioned properly, to put it mildly.

[Monty]

Human beings have bias, it's just the way it is.

Machines may not have "bias", but they do have error rates.  It's not always the voter's fault if the machine fails to read their ballot.  Surely, your copier, printer, etc. has not always functioned properly, to put it mildly.


But your plan sounds pretty good.

[Schmitz in 1972]

On November 4, 2000: that brutal night that took a total turn.

CBS reported Florida for Gore.
ABC reported Florida for Gore.
NBC reported Florida for Gore.
MSNBC reported Florida for Gore.
CNN reported Florida for Gore.
Fox News Channel reported Florida for Bush.
CNN admits they made a mistake; call it for Bush.
MSNBC and NBC admit they've made a mistake; call it for Bush.
CBS admit they've made a mistake; call it for Bush.
ABC admit they've made a mistake; call it for Bush.

That's not a very accurate description of what really happened (and it was Nov. 7, not 4). A timeline that gives a more accurate picture is this:

7:00 Florida polls close in competitive eastern time zone

7:50 Networks (including Fox) call Florida for Gore around based on an infintesimally small ammount of data

8:00 polls close in conservative central time zone, as more precincts trickle in Bush begins to build a lead

9:00 Bush's lead continues to grow, but news station continue to ignore it and insist that Gore has won Florida

10:00 Networks re-classify Florida as too close too call with Bush's lead around 150,000 votes

2:20 Networks (led by Fox as Democrats like to point out) call Florida for Bush even though his lead has shrunk greatly

3:40 Networks once again classify Florida as too close too call, Bush still leading by a hair

5:00 Florida finishes counting with Bush ahead by 1784 votes, post-election struggle begins

[The Vorlon]

Machines may not have "bias", but they do have error rates.  It's not always the voter's fault if the machine fails to read their ballot.  Surely, your copier, printer, etc. has not always functioned properly, to put it mildly.

There is a vastly important difference between error and bias.

Let's say we have vote counting machines that systemically shreaded 10% of the ballots into tiny pieces.

This is actually a fair count, in the sense that no candidate gets an advantage.  Provided the machines shreaded the same % of ballots in all counties, the outcome of the election is (within the limits of probability) unchanged.

By contrast, here is a key finding of the ballot review the media did of Florida 2000:

Although trained to produce accurate, impartial reports, the NORC investigators are human and prone to human judgment and error. In particular, NORC discovered that male investigators were more likely to record marks on ballots than women. NORC also found a slight but statistically significant relationship between candidate marks and the investigators' party affiliation.

This is a bias or systemic error.  529 votes out of 6 million is something like  a factor of 0.000091.

Here even a TINY bias could tip the balance one way or the other.

A machine, even if tragically flawed to the point that it missed even 10% of the ballots would be better than the humans because it would have no systemic bias.  The 10% of votes it shreads have identical probabilities of being for either candidate, so unlike the humans it may be flawed, but is still fair.

[freedomburns]

Machines may not have "bias", but they do have error rates.  It's not always the voter's fault if the machine fails to read their ballot.  Surely, your copier, printer, etc. has not always functioned properly, to put it mildly.

There is a vastly important difference between error and bias.

Let's say we have vote counting machines that systemically shreaded 10% of the ballots into tiny pieces.

This is actually a fair count, in the sense that no candidate gets an advantage.  Provided the machines shreaded the same % of ballots in all counties, the outcome of the election is (within the limits of probability) unchanged.

By contrast, here is a key finding of the ballot review the media did of Florida 2000:

Although trained to produce accurate, impartial reports, the NORC investigators are human and prone to human judgment and error. In particular, NORC discovered that male investigators were more likely to record marks on ballots than women. NORC also found a slight but statistically significant relationship between candidate marks and the investigators' party affiliation.

This is a bias or systemic error.  529 votes out of 6 million is something like  a factor of 0.000091.

Here even a TINY bias could tip the balance one way or the other.

A machine, even if tragically flawed to the point that it missed even 10% of the ballots would be better than the humans because it would have no systemic bias.  The 10% of votes it shreads have identical probabilities of being for either candidate, so unlike the humans it may be flawed, but is still fair.

Neither of the above options are acceptable in my opinion.  The only acceptable answer is a Constitutionally granted right to vote for electors, and a Federally mandated, nation-wide system of tabulation that records every vote with miniscule fallibility.  We have the technology, we have the means, the only thing holding this idea back is that one party thinks it will benefit from it, and one thinks it won’t

As Americans, I wish we gave voting a higher priority.  I think that most other countries are better at this (Democracy) than we are. 

fb

[Inverted Things]

The most idiotic thing Gore did was not ask for a state-wide recount. Cheery-picking your best counties to recount is not a good way to convince people that your just interested in having the votes counted.

[??????????]

The most idiotic thing Gore did was not ask for a state-wide recount. Cheery-picking your best counties to recount is not a good way to convince people that your just interested in having the votes counted.

It's also unconstitutional.

[zorkpolitics]

You must be logged in to read this quote.

Neither of the above options are acceptable in my opinion.  The only acceptable answer is a Constitutionally granted right to vote for electors, and a Federally mandated, nation-wide system of tabulation that records every vote with miniscule fallibility.  We have the technology, we have the means, the only thing holding this idea back is that one party thinks it will benefit from it, and one thinks it won’t

As Americans, I wish we gave voting a higher priority.  I think that most other countries are better at this (Democracy) than we are. 

fb

You must be logged in to read this quote.

I challenge you to define a system of tabulation that records every vote with miniscule fallibility.    (I assume you mean a system that eliminates overvotes and undervotes, as well as one which accurately counts every cast vote.)  For example, the voting system with the highest voter error rate is absentee ballots.  Voters consistently fail to follow directions at between 2-5% of the time (they don’t fill in the bubbles, they fill in too  many bubbles, they circle names, they X out bubbles, they fail to sign the ballot).  The only conceivable way to solve these errors  is to forbid absentee voting.  Of course this  won’t happen.   
If you truly want to maximize voter participation, all voting could be done over the internet so people could  logon anywhere in the world and get their home precinct ballot to vote.  (No lines, no provisional ballots, and error checking could be built in).  But let’s face it, if the conspiracy theorists on the left are afraid of electronic stand alone voting, they’ll go insane over Internet voting.