American Politics Journal

Guest Commentary
Heraldgate Redux
By Paul Lukasiak

Wednesday, April 11, 2001 (APJP) -- Marty Merzer, who wrote the Miami Herald analysis of the undervote,responded to someone who read the piece on

>Thanks for bringing this to our attention, though we have seen others
>like it on the Web. And like most of the others, it is utter nonsense
>-- an amalgam of faulty analysis, poor logic, half-truths and outright
>Our data and our analysis clearly showed that under three of four
>standards, including all of the most likely, Bush would have prevailed
>under the conditions set for the recount by the Florida Supreme Court
>-- including the additional votes awarded to Gore by that court. All of
>our calculations are clearly expressed in the stories and charts. I
>urge you to read them for yourself as they were published by The
>Herald, not as abbreviated or misrepresented by others. They are
>available at
>Lastly, consider this: -- If we were so off-base, wouldn't the
>mainstream media - and more to the point, mainstream Democratic leaders
>- have risen to the occasion by now? Last weekend, Joe Lieberman told
>NBC that our analysis seemed accurate to him, though he was hoping that
>a ballot review by another group of newspapers would yield better news
>for him and his running mate.
>Also, consider the motivation assigned to us by the ''writer'' of the
>piece you cited. He says we merely wanted to stir controversy so we
>could sell more books. Now think about it -- what would have stirred
>more controversy -- a finding that, yep, Bush is still the winner or
>the far more explosive news that Gore was the winner?
>The Herald is a major, responsible newspaper. Our stories go where the
>facts take them.
>marty merzer

I got his response forwarded to me, and I have responded to him directly:

Dear Mr. Merzer:

Thank you for taking the time to respond to my critique of the Herald's analysis of the Florida undervote.

I must say that I do take offense at your characterization of my work. However, inasmuch as you may have read either a draft version, or a version without footnotes, I will give you the benefit of the doubt.

I would like to point out that my critique is unlike most others found on the web. I did not criticize the Herald for leaving out the overvotes. I did not criticize the Herald for publishing the "Gore Won" results the second day. On the whole, I criticized what was presented.

And unfortunately, what was presented by the Herald was pure nonsense.

It was gratifying to note, however, that you descended to ad hominem attacks and logical fallacies, rather than make any attempt to actually defend your work against the factual claims made in my analysis. When such methods are used as a defense, it usually means there is no real defense.

The flaws in your analysis are evident in your statement, "Our data and our analysis clearly showed that under three of four standards, including all of the most likely, Bush would have prevailed under the conditions set for the recount by the Florida Supreme Court."

First off, what makes you think that there would have been a single standard in all counties; that you could, in fact, predict the outcome based on a single standard? Obviously, if there had been varying standards, the entire basis for your "predictions" evaporate.

The simple fact is that you assumed two things that contradict each other in your predictions---the first is that all counties would count votes the same way, the second that counties that had already used different criteria would have their results included in the final count.

Your article makes note of four counties that "completed" their recounts, including three opscan counties. One of these counties claimed to have no undervotes (yet the BDO Seidman review found undervoted ballots). Another apparently claimed that a ballot was undervoted only if there were no votes found anywhere on the ballot, and only counted those ballots (yet the BDO Seidman review found lots of other undervoted ballots in the Presidential race.) No mention is made of the criteria used by the third county, but it is clear that this county actually evaluated ballots that were undervoted in the Presidential race only.

Your analysis incorporated the results from these three counties. Different standards were used in counting the ballots in these counties. Yet all of your predictions are based on the use of identical criteria for counting votes throughout the state. Your own article contradicts the entire basis of your predictions---that a single standard would be used for counting votes.

To me, it looks like the GOP controlled counties were doing what they could to reduce the number of votes counted. What would have happened if Duval County or other GOP controlled counties had used the same criteria as Escambia in evaluating which ballots to count, and looked only at those ballots which recorded no votes whatsoever? Your "predictions" would have been completely worthless, would they not? Under "varying standard" the only prediction that can be made is that the results of the BDO Seidman study are inconclusive with regard to what would have happened were the FSC order carried out.

The other alternative is that Judge Lewis would have imposed statewide standards. The question then becomes, what standard would he have chosen, and what impact would that have on the counties whose votes had already been counted.

As I noted in my piece, the "two-corner detached" criteria had already been ruled illegal by in two Florida Courts as a standard by which to count punch card ballots. Yet you state that the "two corner detached" standard is one of the "likely" standards. On what basis do you believe it was "likely" that Judge Lewis would have ignored the relevant findings of the lower courts on the use of the "two corner detached" standard? Why, given that the "two corner detached" standard had been ruled illegal, was that standard given any more credence than the "clean punch" standard which the Herald relegated to the status of "highly unlikely"?

If Judge Lewis had ruled that a "dimple" standard was to be used (and I note that the Herald used the negatively connotated word "loose" to describe this standard) on what basis do you assume that none of the ballots in Palm Beach, Broward, and Dade that had been reviewed by the County Canvassing Board would not be re-examined for votes? I think if you bother to read the FSC decision, you will note that the Court discussed the importance of a "statewide" remedy, and also made a point of asserting authority over the counting process. There is certainly nothing to preclude the re-examination of the Palm Beach, Broward, and Dade ballots in the FSC decision were a more inclusive standard to be used in the rest of the state.

Next we come to the issue of the Palm Beach Standard. Under the methodology used by BDO Seidman, it is virtually impossible to evaluate the data presented by BDO Seidman within the standard articulated by the Palm Beach County Canvassing Board. That standard was a pattern of dimples on the ballot. BDO Seidman's methodology allowed only for the counters to indicate whether there was at least one other dimple on the ballot. If there was at least on other dimple, a column was marked "Y", if no other dimples, the column was marked "N". Needless to say, one other dimple does not constitute a "pattern" of dimples.

Nevertheless, you labeled as the "Palm Beach Standard" any ballot that had at least one other dimple. The extent to which this is an absolute misrepresentation of the facts---amounting to a flat out lie---can be shown by looking at the actual results. There is no way to know exactly how many undervotes were counted as votes by the Palm Beach County Canvassing Board, because they evaluated all of their ballots. The best source for this information is probably the Palm Beach Post, which stated that "about a 1000" undervotes were counted as votes for Bush or Gore by the PBC board. That is a rate of 15% of the marked ballots for Bush or Gore found by BDO Seidman in its review. Yet, using your version of the "Palm Beach Standard", your data shows that that the recovery rate for all punch card ballots in the state would have been about 45%.

In other words, you are claiming that the Palm Beach County Canvassing Board missed at least one out of three of the votes that met their standard. You are doing so in spite of the fact that they created and articulated the standard. You are doing so in spite of the fact that the data you are using is wholly inadequate to judge whether or not a ballot met the standard articulated by the Palm Beach County Canvassing Board.

And you based your prediction of a "Bush victory" on this complete misrepresentation of the standard used by the PBC board. The extent of your intellectual dishonesty in this matter is astonishing. Calling the "one other dimple" standard the Palm Beach Standard is a lie. The data does not support it, and cannot support it. But it was only through this lie that you could effectively exclude the Palm Beach ballots from your calculations, because, as noted above, it is more than likely that the PBC ballots would have been re-opened had a "one other dimple" rule been established statewide.

I stated that the reason that you misrepresented the results in order to sell books was accurate. I am fully aware that the reason the Herald engaged in the examination of the ballots was to find out what would have happened has the FSC decision been carried out. Had you told the truth in your analysis-that the results were inconclusive, because no one knows exactly what would have happened and the data does not conclusively support a win by either side---would, in fact, have sold far fewer books. The *truth* is that nobody knows what would have happened next, whether counties would be allowed to use wholly different standards in counting votes, whether overvotes would eventually be included in the final totals, etc.

The *likely* result, given the FSC's consistent tendency to count all the votes, is that if the drama had been allowed to play itself out in Florida, that Al Gore would have won. There is no question, based on your data, that more people went to the polls to vote for Gore, and that if all the votes were counted, that Gore would have won. There is little question, even at this point, that the overvote analysis being conducted by NORC will show that Gore would have won if all the legal votes had been counted in Florida. There is no question, based on the analysis of overvotes done by Dan Keating for the Washington Post, that tens of thousands more people went to the polls to vote for Gore. The only question is whether or not all the votes would have been counted.

As to your claim that if your results were so inaccurate, why was nobody complaining about them, the fact that the Herald is not reporting the complaints does not mean they don't exist. The fact that Joe Lieberman has not bothered to analyze your work does not mean that your work is of any value. And the fact that the "pack" tendency of the media to simply parrot other reporters rather than look at the facts themselves does not mean that the first reports were accurate. I won't bother to detail exactly how often and to what extent this happens---suffice it to say that the accusations regarding financial irregularities with Jesse Jackson's PUSH were front page news….the finding that there were no financial irregularities went unreported or was completely buried when the report was released.

As to your claim that the Herald is a responsible newspaper, and that you stories go where the facts take them, that is pure hogwash. The Herald decides which facts to pursue, and which facts to ignore. It decides to make a major effort to find dead people who voted, while ignoring the corruption of state officials in this election. Has the Herald done ONE story on the failure of Katherine Harris to fulfill her statutory duties to ensure that the election laws of the State of Florida are implemented consistently throughout the state? Has the Herald done ONE investigative piece on felons list, and how a statute that demanded an accurate list was deliberately misinterpreted by state officials to allow inaccurate lists? Have you even read the "felon's list" statute? Do you know anything about the legislative history of the statute?

Please don't try and pull off the "Herald is responsible" nonsense on someone who has been following this story. The Herald flat out lied about not releasing partial results when it released its incredibly biased, misleading and inaccurately reported "results" of the Dade County ballots. The Herald has no right to any claim of "integrity" after that episode.

I don't expect a response, because I'm pretty sure that you cannot provide an adequate one, and like most reporters will ignore contrary facts rather than admit that you are completely wrong. Just keep in mind that the reason the press has so little respect and credibility in this nation is that people like you are doing the "reporting."

Paul Lukasiak

SpacerAmerican Politics Journal

Copyright © 2001, 2000, 1999, 1998, 1997, 1996, American Politics Journal Publications, Inc.
All rights reserved. Read our privacy policy. Contact us.
ISSN No. 1523-1690