You are viewing an obsolete version of the DU website which is no longer supported by the Administrators. Visit The New DU.
Democratic Underground Latest Greatest Lobby Journals Search Options Help Login

Reply #61: There is much to refute here, but it's very easy to do. Just watch. [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
Home » Discuss » Topic Forums » Election Reform Donate to DU
WillE Donating Member (150 posts) Send PM | Profile | Ignore Mon Aug-24-09 12:09 PM
Response to Reply #60
61. There is much to refute here, but it's very easy to do. Just watch.
Edited on Mon Aug-24-09 12:21 PM by WillE
Well, I'll be honest. Your argument here is an unsupported appeal to personal incredulity, of the same sort as creationists' claims that it is "impossible" for eyes to develop through evolutionary processes. (In your second post you add, or revert to, argument by adjective: "massive" defection. Obviously, the more Gore votes there were to begin with, the more "massive" a defection is possible; there really isn't much more one can say about that.)

Creationist? Incredulity? Thats what YOU are famous for. You keep setting the standard.You are now reduced to resorting to pure jargon that is meant to fog, not illuminate.

Remember False Recall?
You were hoisted on your own petard. NES respondents correctly recalled their past vote.

Remember Swing vs. Rred-shift?
You were hoisted on your own petard. There was a correlation.

In real life, we don't have enough information to infer confidently what percentage of Gore voters 'must have' defected to Bush in these 15 counties -- and we certainly don't have enough information to declare any particular percentage "impossible." The existence of four late pre-election polls, each one giving Kerry a lead between 15 and 18 points, is a pretty good indication that Kerry probably didn't win the state by 30-plus points, even if we assume some bias in the LV models. It's mildly interesting to see you screen out that inconvenient evidence, but it isn't very interesting to argue with you about it.

So you trust pre-election LV polls more than exits?

You say there is not enough information to infer what percentage of Gore voters must have defected? Only if you dont do the analysis.

We can make a very sound estimate based on the Kerry exit poll shares of New and returning Nader voters. So there is not enough information to evaluate the Election Day NY 2000/2004 county recorded votes in conjunction with 500,000 late recorded paper ballots and the unadjusted WPE, GEO and Composite exit poll data? Thats what a REAL analyst would do.

You have plenty of information. You just avoid using it. Why not? As you said, its just arithmetic.Screen out inconvenient evidence? So Rovian of you to accuse me of what you yourself are guilty of. You screen out the NY exit poll timeline at Call 1 (Kerry 64%), Call 2 (Gore 65%) and Call 3 (Kerry 64%). These polls were taken immediately AFTER the votes were cast.

YOU screen out the 1988-2004 Bush vs Clinton average WPE anomaly.

YOU screen out the fact that unlike your pre-election polls,in the exit polls there are ZERO undecided voters - they already voted; there are ZERO likely voters (LV) they already voted.

Must I remind you again that UNDECIDED voters broke for Kerry everywhere, not just in NY?

Must I remind you that Gore won NY by 25% - without the help of the 4% NADER/other voters who defected to Kerry by 3-1?

Must I remind you that Kerry won NEW voters nationwide with 57-60%?

Must I remind tou that since he did 10% better in NY, we can reasonably infer that he had 65-70% of new voters in the state?

Why must you keep hanging on those unallocated NY pre-election LV polls and ignore all else?

You are a fan of cherry-picked pre-election polls but abhor the exits. Why is that? Is that all you have?

I don't think this requires any explanation at all. The WPEs reported by E/M are -11.4 in 2004; -3.3 in 2000; +2.1 in 1996; -4.6 in 1992; -7.2 in 1988. We don't have standard errors for any of these, but we know the Best Geo standard error in 2004 was 3.7. We can infer, of course, that most of these WPEs are statistically indistinguishable from each other, although clearly they aren't all equal. Your retrospective partitioning is a variant of the Texas sharpshooter fallacy. Even if one believes that the exit polls are infallible beyond sampling error, it's wishful to infer that the vote count was more accurate in 2000 than in 1992

You avoid the REAL issue here: the average of the 3 Bush WPEs vs. that of the 2 Clinton. Only YOU can infer that they are indistinguishable a)because you dont look at them hard enough or b) it's more likely that you can't explain it. The difference between average 8% Bush MoE and the average 0.6% Clinton MoE is certainly distinguishable to any FAIR observer. Are you a FAIR open-minded observer? Thats a rhetorical question.

Your observation could at least be interesting if you had answers to my straightforward and pertinent questions in post 52. Do you? I am not at all sure, but I suppose some DUers might be interested in your explanation of how Bush Sr.'s role as vice president empowered him or his minions to alter the lever machine counts, and why this capacity apparently didn't extend to, say, Pennsylvania. As a bonus, I note that the WPE in Texas in 2000 was -0.4. I will leave it as an exercise for you to deduce my question about that.

You know the old saying: the apple never falls too far from the tree.
Or: Like father, like son.
Or the latest version: Like Bush 1, Bush 2

We know who was running the show behind the scene throughout the Reagan years. Bush was Reagans Cheney. What makes you think he did not have the foresight to set things up prior to the campaign? Which by the way was less than stellar. Remember Willie Horton?

So typical of you to cherry-pick ONE exit poll (Texas 2000).Nationally, the exit polls matched up fairly well in 2000. Just like in NY. Clinton was the incumbent. So what in heavens name is your point? You are just confirming what I said before.

Ill say it again:
The average NY WPE in the two Clinton elections (1996 and 2000) was 0.6.
The average NY WPE in the three Bush elections (1988,1992, 2004) was 8.0.

I provided all these numbers. You call them "creationist".
1. 5 NY wpes (8.0 average for Bush in 3 elections; 0.6 average for Clinton in 2)
2. The NY exit poll timeline (64.65,64% for Kerry)
3. 15 top NY counties went 64-31% for Gore; he won the RECORDED vote by 60-35% over Bush (that doesnt include 200k uncounted NY votes in 2000 the vast majority to Gore)
4. 19 levers stuck on Bush; none on Kerry
5. 66% of 500,000 late paper ballot votes for Kerry6. 70-80% of Nader voters for Kerry.
7. 67-70% of 1 million new NY voters for Kerry.
8. The strong .57 correlation between NY county size and Bush % vote change from 2000 (Urban Legend)
9. The corresponding weak .07 Kerry correlation.

And what did you give? Just one little old 2000 Texas exit poll that only confirms the fact that the exit polls were accurate nationally as well as in NY. Clinton was the incumbent, NOT Bush.

That was s classic cherry-pick for the ages.

Printer Friendly | Permalink | Reply | Top

Home » Discuss » Topic Forums » Election Reform Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002
Software has been extensively modified by the DU administrators

Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC