Miami Herald recount: No flaws...or were there?!?

In an article published yesterday, reporters from the Miami Herald provided data from hand counts that they conducted in several northern "dixiecrat" counties in Florida. They concluded that nothing was awry, and in fact, went so far as to title the article "No flaw in Bush's state win."

Here's a link to a reprint of the story that doesn't require registration:

http://seattletimes.nwsource.com/html/nationworld/2002102809_florida28.html

Below I take a closer look at the results of their recount and come to a very different conclusion.

Here are the tallies for Union County:

Bush original: 3396
Bush hand count: 3393 (-3)
Kerry original: 1251
Kerry hand count: 1272 (+21)
Net change: Kerry +24

Here are the tallies for Lafayette County:

Bush original: 2460
Bush hand count: 2452 (-8)
Kerry original: 845
Kerry hand count: 848 (+3)
Net change: Kerry +11

It's a bit more complicated for the third county they looked at, Suwanee County, because they didn't count all of the votes, stopping at "almost 60%." But here are the original tallies and the results of the hand count:

Bush original: 11153
Kerry original: 4522
Bush hand count: 6140
Kerry hand count: 2984

In the original count, 71.2% of the votes cast for Bush or Kerry (n=15675) went to Bush. In the hand count, this drops to 67.3%. That is a significant drop. Let's translate that into numbers. If you take the percentages from the hand count and extrapolate, here's what you get:

Bush = 15675 x .673 = 10549 (loss of 604)
Kerry = 15675 x .327 = 5126 (gain of 604)
Net change: Kerry +1208

[I know it's not legit to extrapolate like this since we don't know if the 60% is a random sample. This is just for arguments sake.] Obviously a switch of 1208 votes in a county with less than 16K votes cast would be huge. Now maybe there's a very large percentage of Bush votes in that remaining 40% that they didn't count, but we can't know that because they didn't count them. Which begs the question...why did they stop counting in Suwannee County when their tabulation of 60% of the ballots deviated from the original total? And more importantly, without actually counting those remaining ballots, how can they possibly report that nothing is amiss, when the data they do have is inconsistent with the certified vote tally?

Once again, they conclude that there's "no flaw in Bush's state win" and even use this as the title of the article. But one could reasonably argue, based on their data, that Kerry may have received over 1200 additional votes from the three small counties they examined. Moreover, the votes they counted represent about 0.3% of the ballots cast for Bush and Kerry statewide. If Kerry gained votes at the same rate across the state, he picks up about 400,000 votes and wins Florida.

Thanks Miami Herald, you just revealed to us that there's a possibility that Kerry won Florida.

•  Wow!...

  ...Good catch!

  by Blue Shark on Mon Nov 29, 2004 at 11:13:29 PM PDT

  •  Ditto the accolade

    There was a huge half-page story in the Santa Rosa Pee Dee with a headline implying that all claims of vote fraud in this year's election had been conclusively debunked. I read further, and it was only about the Dixiecrat counties.

    I was grumbling to myself that they'd picked the weakest of all the vote fraud allegations, one which had been debunked within minutes by random people on Daily Kos as soon as someone mentioned it here, to promote as the example of fraud claims which they could themselves debunk to 'prove' that there was no fraud.

    I had no idea they'd actually found smoke. That Suwanee 4% difference between the 60% count and the 100% count is interesting, and might be less or more so if we knew whether they'd evenly half-recounted many precincts or left a disproportionate number of expectedly pro-Bush precincts in the uncounted 40%.

    Another interesting tidbit I heard and can't confirm is that the most heavily pro-Democratic urban areas of Florida used touch-screen machines which don't leave a paper trail. Does anyone have exit poll numbers for these areas, compared to results and the same for other areas to compare against?

    by Cecrops Tangaroa on Mon Nov 29, 2004 at 11:34:01 PM PDT

    •  Look at U of PA and other studies

      Academic Papers on 2004 Election Results - dKosopedia

      by lawnorder on Mon Nov 29, 2004 at 11:37:44 PM PDT

  •  more data

    the Suwanee number are VERY suspicious.

    Let me put it this way...there are 16 precincts, plus an "absentee/early voting" precinct.

    Now, if you add together all of the ballots cast on election day at a polling place, for Bush you get 6966, and for Kerry you get 3001.

    The Herald says they counted 6140 votes for Bush, and 2984 votes for Kerry.

    The difference is 826 votes for Bush, and 17 votes for Kerry.

    This doesn't make sense....

    Another interesting thing is that the Herald picked three counties that DON'T provide good detailed results by precinct.  Neither Lafayette nor Union have a precinct breakdown on their websites, and Suwanee does not separate out early voting and absentee ballots---and I could find no precinct based registration information on Suwannee either.

    I've been helping a researcher by converting precinct based data found on county web sites to spreadsheet form....and I've done over 25 counties, and THESE counties are exceptional in the lack of information they provide.

    by bushsux on Tue Nov 30, 2004 at 05:41:38 AM PDT

    •  except that the reporters

      wrote:

      "The Herald counted almost 60 percent of the votes in Suwannee, where nearly 64 percent of voters are registered Democrats.

      The newspaper's total: 6,140 votes for Bush and 2,984 for Kerry, which nearly matched the county's official tally."

      They knew the tally in the 60% of precincts that they counted.  They didn't tell us.

      I guess you could believe that the Miami Herald reporters just lied.

      But that seems unlikely to me.

      You really have to believe that just about everyone is in on the conspiracy - the election authorities in dozens of counties, the reporters who investigate, the democratic chairs in those counties, the duped voters, few of whom have batted an eye at the Bush totals in their counties.

      That's silly, of course.  The fact is that there is no discrepancy in having fewer Kerry votes than the registered dems casting ballots in a southern county.  You sound stranger and stranger trying to insist that this is evidence of tampering.  You have to be unaware of how election technology works and of how people vote to continue touting this theory.

      by ne plus ultra on Tue Nov 30, 2004 at 03:35:27 PM PDT

      •  oh

        my commment was in reference to the overall theory espoused in this diary, not to the specific comment I responded to.

        Sorry about that.

        by ne plus ultra on Tue Nov 30, 2004 at 03:36:17 PM PDT

•  Bullshiite from Miami Herald

  From Common Ground Common Sense

  I thought we had discounted the DixieCrat situation? Sure you can print things that CONFIRM the election....

  Do they now have a silver bullet against all future and current questions? They can reprint this little piece of history to PROVE that no fraud occurred.

  What's up in Volusia?? Where's THAT story?? Somebody get someone on the phone at the Miami Herald!!!

  ---------------

  This the same kind of BS that came out after the 2000 fiasco. Newspapers did their "independent" investigations and amazingly came to the conclusion the Bush won. Pure BS.

  ---------------

  The big problem was not in the Dixiecrat region of Northern Florida but the large Democratic counties in South Florida. These media people will do anything to discredit the fraud story to substantiate their negligence to cover it.

  -------------

  http://www.sunherald.com/mld/sunherald/new...cs/10284888.htm
  Here's a good line: So, if there is no way to recount, how do they know the discrepancies are "minor"?

  http://www.palmbeachpost.com/politics/cont...WVOTE_1105.html

  http://www.wesh.com/politics/3881646/detail.html

  http://www.votersunite.org/article.asp?id=3803

  http://www.rubberbug.com/temp/Florida2004chart.htm

  -------------

  It IS cyber BS

  Optical scan counties were not the main problem. Touch screen counties were!

  --------------------

  Miami Herald did it once before...

  Media Deliberately Distorts Miami-Dade Recount Story

  WASHINGTON - February 26, 2001- Democrats.com today exposed the deliberate distortion of the Miami-Dade recount story by the Miami Herald, USA Today, and the national news media. In addition, Democrats.com published an investigative report documenting the failure of election officials in eight counties to count 'write-in overvotes', as required by Florida law.

  Democrats.com co-founder Bob Fertik published an article entitled, 'Bush Propaganda Machine Distorts Miami-Dade Recount Story,' which described the media's distortion as a continuation of a propaganda campaign by the Bush campaign that began on Election Day, when George W. Bush unexpectedly found himself trailing in Florida. Fertik's article is available here:
  http://democrats.com/view.cfm?id=1604

  'Today's misleading headlines by the Miami Herald and USA Today are just another attempt to legitimize Bush's stolen presidency,' said Bob Fertik, co-founder of Democrats.com.

  http://www.commondreams.org/news2001/0226-13.htm
  

  by lawnorder on Mon Nov 29, 2004 at 11:35:29 PM PDT

•  Saw your post on DU just now

  Great find!!  This could be the greatest news of the week!  60% of the vote counted will net you a MOE of less that 1% as a sample of the total vote in that county.  In other words, it is HIGHLY unlikely that that percentage will change.

  We must get this to every news organization!

  And put up a tip jar!

  by JohnnyCougar on Mon Nov 29, 2004 at 11:36:18 PM PDT

  •  If 60% gets you a MOE of less than 1% as a sample,

    then why did so many people pooh-pooh the analysis done by "markusd" (or was it "bruh1"?) a week or so ago showing the significant deviation in North Carolina in election-day tallies -- for presidential and U.S. senate races ONLY -- from BOTH exit poll data AND combined early voting/absentee ballot figures (which I believe were consistent with each other), even though a higher proportion of registered Republicans voted early or absentee than voted on election day, and even though the combined early/absentee vote was a full 30% of the total vote count?  I'm a theologian, not a statistician, but surely 30% of the total vote for a state is somewhat representative, especially given that the absentee voters were more weighted to Bush than were early voters?

    by GreekGirl on Tue Nov 30, 2004 at 01:13:59 AM PDT

    •  Mmmmm, theologian...

      You just made my top ten list of people I'd most like to spend an afternoon talking with over coffee.

      by tompaine2004 on Tue Nov 30, 2004 at 04:17:21 AM PDT

•  Great catch!!!

  Oh the irony!!!!

  MH finds +400,000 votes for Kerry possibility, after trying to parrot Carl Rove...

  

  by lawnorder on Mon Nov 29, 2004 at 11:48:32 PM PDT

•  Hung Jury

  I think it's great that we have this checks and balances mentallity here at DKos concerning information pertaining to voting irregularities or potential fraud.

  But after reading this diary, you really need too pull your nose away from the words and look at the entire page.

  There is just way too many things pointing (stronly or weakly) towards shenanigai in this election.

  As much as I appreciate everyone's effots not to allow us to run with a story that can be debunked or put into question by the "smother" side, the occasional mention to the growing mountain of circumstancial evidence pointing towards foul play would at least get us a "Hung Jury" in most courts.

  All I'm saying is, step back from the tree in front of you, before the burning forest around you engulfs you.

  by The 1n Only Leoni on Mon Nov 29, 2004 at 11:48:42 PM PDT

•  Hmmmm.

  by warbly on Mon Nov 29, 2004 at 11:51:45 PM PDT

•  sinister thought of the day

  what if the writers knew what was going on, but wrote the story like this to get it printed? eg. they knew the story would get killed if they actually found anything.

  they have some incredible faith in you bloggers.

  by Furious Hal on Tue Nov 30, 2004 at 12:16:25 AM PDT

  •  They should have their articles...

    ...reviewed by dKos before submitting them to their editors.  And I say that only partly tongue in cheek.  

    by tompaine2004 on Tue Nov 30, 2004 at 04:20:46 AM PDT

  •  No, this is what's sinister

    You make a good point. It could easily be a way of reversing the bait-and-switch we've seen for the last four years. Headline favourable to Repubs and body that should lead to obvious conclusions.

    Key word here is should. Sinister thought is that the Herald allowed it to be printed fully aware of this cognitive dissonance simply because they decided most readers didn't have the cognition or filters that would allow them to see the truth underlying the headline.

    by thingamabob on Tue Nov 30, 2004 at 06:58:40 AM PDT

    •  kinda ot but...

      that's exactly what i thought about colin powell's speech when he said we couldn't accept the ukraine election results.

      i felt like, "if this man has a good bone left in his body...he's writing a roadmap here."

      i know, i know. i'll be getting a tinfoil jumpsuit and helmet under my x-mas tree for sure.

      by cais27 on Tue Nov 30, 2004 at 07:27:58 AM PDT

      •  gogo

        sign language interpreters.

        by Furious Hal on Tue Nov 30, 2004 at 02:06:34 PM PDT

•  Not probable: Suwanee County

  Please excuse the ramblings of a tired statistician. Think of an election as flipping a coin, but not with 50-50 odds. Instead, once we count the votes, we know the odds of a randomly choosen ballot in Union County being for Kerry. It is p = 1251/(1251+3396) = 0.2692, according to the orginal count. I assume a binomial distribution.
  Now I can ask about the probability of the distribution Kerry=1272 and Bush=3393 varying from the above distribution based solely on math of the binomial distribution. Turns out the handcount data has p between 0.260 and 0.286 95% of the time. In fine agreement with the original count.
  As for Lafayette County, the original count has p = 0.256. The handcount has p between 0.242 and 0.272 at 95% C.L.
  So, let's turn to the Suwanee County data. The reported handcount gives p = 0.2885. Now, we assume the null hypothesis of this p, when I have 9124 samples with 2984 of them being Kerry ballots. When I run this test, I find poor agreement: p between 0.317 and 0.337 at 95% C.L. At this level, the models representing the orginal and handcounts are 3.9 standard deviations apart.
  I will think about this more tomorrow (later today); but there might be something here.

  by Nicholas Phillips on Tue Nov 30, 2004 at 12:21:02 AM PDT

  •  Thanks for that additional analysis.

    And to this person's eyes, nobody in there right minds would look at those results from the 60% sample, say "all is good," and quit counting. At the very least, that pattern with the 60% sample would compel you to keep counting because the tallies are, in fact, way out of line.

    Yet our good friends from the Miami Herald not only say everything is fine in these three counties, contrary to the data they themselves collected, they even have the gall to say everything is fine in the whole state!

    It's really, really shoddy reporting, and they need to be taken to task on it.

    by pointsoflight on Tue Nov 30, 2004 at 12:28:49 AM PDT

    •  It was pretty stupid

      It was pretty stupid that they stopped conting at 60%. Doing so was about as useful, perhaps even less useful, than not counting any of the county at all.

      by Chris Bowers on Tue Nov 30, 2004 at 12:32:13 AM PDT

      •  Yes, it was pretty stupid

        It was pretty stupid that they stopped conting at 60%. Doing so was about as useful, perhaps even less useful, than not counting any of the county at all.

        They only needed to count 6,500 more and they would have had an argument.

        by fly on Tue Nov 30, 2004 at 07:24:38 AM PDT

        •  I wonder...

          ...why the writer included partial results from this county, when the picture would have been clear had the story only reported the two counties with negligible discrepancies?
          

          by KTinOhio on Tue Nov 30, 2004 at 02:03:28 PM PDT

      •  Hmmmm

        Did they say which precints were counted? Does that county have more than one type of voting machine? That might explain why there is a partial count. But without a satisfactory explaination they should have just dropped this county from the story. It just gives rise to more questions.
        

        by ineedalife on Tue Nov 30, 2004 at 11:33:38 AM PDT

    •  They stopped counting

      when it became clear that they were in a problem area.
       The first 2 counties listed above had Kerry gains of 0.5% and 0.3% (of the original total), while the extrapolation of the 3rd indicates that they would have found a 7.7% gain for Kerry had they continued counting. I don't know how the scan ballots would be sorted so that 60% was NOT a random sample. I think they have presented a smoking gun, and called it an alibi!

      by Cache on Tue Nov 30, 2004 at 11:59:13 AM PDT

  •  Goodness! 3.9 standard deviations

    I have no idea what that means, and your post made my head hurt, but good work.

    This election was rigged. I know it and I don't need no stinkin proof, but since others seem to require some, I'm glad you math enabled folks are on the case.

    by warbly on Tue Nov 30, 2004 at 12:34:19 AM PDT

    •  standard deviation

      You've probably heard the expression "bell-shaped curve."   Let's say you plot a graph of the deviation from average of some parameter.  If the object being studied is random then 2/3 of all the values (SAT scores, IQs, height, weight, whatever) will fall within plus or minus one standard deviation of the average.

      95% will be within plus or minus 2 standard deviations, and 99% will be within plus or minus 3 standard deviations.   So when someone says that something is 3.9 standard deviations from the average, it means there was a very low probability of its happening by random chance.  

      This doesn't, for example, necessarily mean that the election was rigged, but it means there is a high liklihood of some sort of anomaly having occurred (rigging being one obvious possibility) and suggests that further inquiry is needed.

      BTW, in the above discussion I've glossed over some of the subtle statistical points to try to get a simple explanation for non-technical people, but if anyone thinks I've oversimplified too much please correct/followup.
      

      by quake on Tue Nov 30, 2004 at 02:48:41 AM PDT

      •  3.9 standard deviations

        to put it simply.  if the average male population is about 5'10 inches and you see someone who's 6'10" you know that person is tall.  If you see a bunch of people who are 5'10" but a doctor tells them ALL they are 6'10" you've got a measurement problem.

        In this case we don't know if the individual we're looking at is really tall or if the measurements are off. (but if i could find a way to place a bet, you can guess where i'd put my money.)

        That's why we like to have more than one case.

        by dr colque on Tue Nov 30, 2004 at 02:50:32 PM PDT

  •  See my comment down thread

    on why this was probably not a random sample

    by daunte on Tue Nov 30, 2004 at 12:53:45 AM PDT

  •  The one flaw in extrapolating these numbers...

    ...is that, under the null hypothesis of the total being accurate, the last 40% of the votes are not dependent on the first 60% of the votes.

    Take a deck of cards: 25% of the cards are spades.  I deal myself 31 cards (almost 60%) and, by chance, I get 11 spades (11/31 = 35.5%).  If I extrapolate to the 21 remaining cards, I would estimate that the deck has 0.355*21 = 7.45 more spades, for a total of 17 or 18 spades in the deck. Which is nonsense.  I know that there are exactly two spades remaining in the deck because there were only 13 spades to begin with.

    Now, the real question is: given the expected total we began with (if there are no shenanigai), what are the odds that the hand recounted SUBsample could have been produced?

    Probability analysis of these counties is asking exactly that -- given that we expect 28.8% Kerry votes, if we pick 9,124 of the 15,675 total votes, what are the odds that we will get 32.6%?  95% of the time, we expect the total for that large a sample to be between 2548 (27.9%) and 2717 (29.7%).  Getting 2984 votes is way outside that...

    

    To put it another way: the odds of getting this many "spades" out of the deck are less than 10^-15, or less than one in a quadrillion (thousand trillion). Conclusion: our initial hypothesis (28.8% Kerry) must be rejected.

    HOWEVER, I must emphasize that this number is HEAVILY dependent on random sampling. As other posters have said, we don't know if there were very pro-bush precincts excluded -- and counting the rest of the deck will tell us if there are 2 or 7 spades left.

    Only one thing to do: COUNT THE VOTES!

    (I did these calculations and graphs in R, an open-source stats package.  Its very hard to use, and I am far from an expert, but heres the code I entered.)
    ####
    expected = 4522/(11153+4522)
    result = 2984
    total = 2984 + 6140

    # c(2300:3000) says do every integer from 2300 to 3000
    plot(c(2300:3000), dbinom(c(2300:3000), total, expected, log=FALSE))

    # lower 95% interval
    qbinom(0.025, total, expected, log.p=FALSE)

    # upper 95% interval
    qbinom(0.975, total, expected, log.p=FALSE)
    #####

    by bilge on Tue Nov 30, 2004 at 09:08:21 AM PDT

•  Not seeing this

  It's a bit more complicated for the third county they looked at, Suwanee County, because they didn't count all of the votes, stopping at "almost 60%." But here are the original tallies and the results of the hand count:

  Bush original: 11153
  Kerry original: 4522
  Bush hand count: 6140
  Kerry hand count: 2984

  In the original count, 71.2% of the votes cast for Bush or Kerry (n=15675) went to Bush. In the hand count, this drops to 67.3%. That is a significant drop.

  The two counties they hand counted in their entirety showed no real difference. In the third county, they only did a partial count.

  Extrapolating from the 60% count in one county to show a difference in that county assumes that the 60% of votes they tallied in that county is exactly representative of the county as a whole. It is a huge assumption, since the 60% almost certainly was not tallied evenly across precincts. On what basis are we to assume that the 60% of votes they counted in that county are exactly representative of the county as a whole? They could easily have skipped some of the more heavily Bush or Kerry precincts. I'm sure we all know by now that Bush and Kerry supporters tend to live in clumps. Considering this, a 5% difference at this stage of the counting seems pretty reasonable to me.

  The number one key in producing an accurate poll is making certain that you are drawing your data from a representative sample population. We have no way of knowing that the 60% they tallied is, in fact, a representative sample population of the entire county. In fact, since they certainly did not draw evenly from every precinct to reach their 60% total, we can be pretty certain that this is not a representative sample. Thus, extrapolating from that 60% does not demonstrate much of anything, except that they should have tallied the entire county or not bothered to tally any of it at all.  

  We are fond of calling John Kerry a coward on this matter, but his top field man, Michael Whouley, was the guy who stopped Gore from publicly conceeding in 2000, and is the best field man in the business. Perhaps I am putting too much faith in him, but it is very hard for me to imagine that he would have missed some county, or even precinct, based irregularity that we would pick up on based on a 60% count.
  

  by Chris Bowers on Tue Nov 30, 2004 at 12:30:49 AM PDT

  •  I completely understand all of that.

    My main point is that there conclusion does not follow from the data.

    The (incomplete) data they have is not in line with the official numbers, so how can one possibly conclude that there is "no flaw in Bush's win in state"?

    At the very least, their hand counts compel further counting. It certainly doesn't compel the type of conclusion they make.

    by pointsoflight on Tue Nov 30, 2004 at 12:42:07 AM PDT

    •  I'm all for calling media blunders

      Which this clearly is, and which you are correct in pointing out. Their information about the third county demonstrates nothing at all, excepts that they are lazy and have weak analytical skills.

      by Chris Bowers on Tue Nov 30, 2004 at 12:52:27 AM PDT

      •  Motives.

        Yes, and since they clearly had political motives, they need to be called on this. They're trying to act like great citizens who are investigating for the greater good of the state, when in fact, they're nothing but tools.

        by pointsoflight on Tue Nov 30, 2004 at 01:57:00 AM PDT

      •  weak analytical skills

        I see weak analytical skills. Two different parts of the brain are used for math and writing. Writers seldom are statisticians and stats people are seldom good story writers.  Someone who is good with stats and writing is usually working for a corp. in sales or marketing.
        I think the author of this diary should email the writer of the article and point out the mistakes. Any dKOS statistician in FL able to travel with a writer from the Miami Herald?

        Drink shade grown coffee!

        by mad ramblings of a sane woman on Tue Nov 30, 2004 at 05:42:40 AM PDT

  •  Yep.

    Extrapolating the 60% is, well, a useless thing to do since you don't know which precincts had been counted. Which invalidates the entire analysis.
    

    by Frank on Tue Nov 30, 2004 at 12:42:33 AM PDT

    •  no it doesn't invalidate the analysis

      the poster concluded that there is no justification for the "No Flaws" headline, and in his diary he said that possibly that uncounted 40% could be heavy with Bushies.

      The point is that the no flaws in Florida headline is bunk, based on 2.6 counties.

      Since they left them uncounted, extrapolating them is all he could do. To just accept the shoddy reporting from the Herald, is a useless thing to do, especially given their track record regarding the 2000 recount.

      by warbly on Tue Nov 30, 2004 at 12:51:12 AM PDT

      •  Nah.

        Extrapolating is a useless thing to do in this case. Now, the reporting wasn't very good either; the county that was counted for 60% shouldn't have been mentioned. But, countering this with an extrapolation that doesn't make sense isn't useful either.
        

        by Frank on Tue Nov 30, 2004 at 12:55:13 AM PDT

        •  Not useless at all

          it provides one possible result of a complete count, based on a 60% sample.

          Not a smoking gun, but not useless, if only because it invalidates the Herald's 'move along, nothing to see here' reporting.
          

          by warbly on Tue Nov 30, 2004 at 01:00:17 AM PDT

        •  The point of the extrapolation...

          ...is just to drive home the shoddy reporting. I don't at all suggest that those extrapolated numbers are what you'd find. The intent is to point out that it's ridiculous for them to claim there is "no flaw in the state" based on the data they have in hand, which trended to Kerry to a large enough extent that on a statewide basis it would overturn the election.

          Another way to think about it is that my ending statement (said tongue in cheek) that Kerry may have won Florida holds just as much water, based on their hand counts, as their conclusion that there was no flaw.

          by pointsoflight on Tue Nov 30, 2004 at 01:28:44 AM PDT

          •  argh

            Just because they didn't count every precinct in the state doesn't mean there was shoddy reporting.  They counted 60% of the precincts in the county and found no problems at all.  

            by Damon on Tue Nov 30, 2004 at 06:11:18 AM PDT

            •  unqualified

              It is shoddy reporting because they don't qualify the basis of their conclusions. They don't say why the recount was limited. They don't say if they are comparing it on a precinct basis, and although they give the tally of votes they counted, they don't give the numbers they are comparing them to. That is sloppy, and it makes their study an unqualified success.

              by white blitz on Tue Nov 30, 2004 at 06:32:37 AM PDT

      •  imho

        i think the shoddy headline was so the story would get picked up, otherwise it would get buried by the SCLM.

        by Furious Hal on Tue Nov 30, 2004 at 12:58:46 AM PDT

  •  even further

    I think its likely that if you were starting a recount and looking for problems you would start with the precincts that had the highest percentages of democratic registration to try and find a problem.  Thats the flaw in this analysis- its looking at the problem as if the only way to find a problem is to look at the whole county when in reality- looking at every precinct in its entirety is what really matters.  If they had only counted 60% of every precinct then this analysis would hold some weight, but assuming that every precinct was fully counted and showed no major problems then there just isn't anything at all to this analysis.

    by Damon on Tue Nov 30, 2004 at 05:42:23 AM PDT

    •  Why is a "Moderate Republican" who

      didn't vote for Kerry spending so much time on dkos trying to debunk allegations of election irregularities?

      0

      Mambo

      by Mambo on Tue Nov 30, 2004 at 06:22:01 AM PDT

      •  get a thicker skin

        I will stand by my status as a trusted user to back up that I do not disrupt the conversation.  Giving me a zero for that post that is clearly bringing value to the thread shows just how rediculous you are.  I would give you a zero in retaliation, but that would just lower me to your level and I have no intention of doing that.

        by Damon on Tue Nov 30, 2004 at 06:32:21 AM PDT

  •  but seeing this..

    The two counties they hand counted in their entirety showed no real difference.

    I've been coming to an opinion that these 'vote irregularities' are too systematic. Perhaps not enough to be responsible for Bush's victory but still significant. A long line here, a provisional challenge there, ... adding up to a margin that would be enough to swing a very close election or pad the lead in a merely close one.  The Miami Herald was looking for big swings suggested by the dem reg vs Bush vote mismatch, not something more subtle.

    Note that the no real difference is actually ~0.45% (net-change per total). I really don't know whether that is small or not. Note also that it is systematic - both hand counts went in Kerry's favor (though 2 is clearly not a sufficient sample). Note also that Bush "lost"  votes in both cases. That seems odd.

    This did catch my eye because I remember hearing an election official here in WA saying before our recount that such recounts typically add to the leader. That is, random irregularities during the initial count will cost the one who is ahead more often than the other way around.

    I don't think these specific results are sufficient to claim anything except for a need to look harder.

     

    by pvjeff on Tue Nov 30, 2004 at 10:45:53 AM PDT

•  From the article:

  The Herald counted almost 60 percent of the votes in Suwannee, where nearly 64 percent of voters are registered Democrats.

  The newspaper's total: 6,140 votes for Bush and 2,984 for Kerry, which nearly matched the county's official tally.

  Thanks for looking up the actual count and doing the comparison. I thought it was awfully suspicious when I read that the count stopped at almost 60% and they couldn't be bothered to tell me the official count. BTW, do we know if all of the votes in Suwannee had a paper trail that could be counted? In other words, did they only count that many because that's all that could be counted or because it's all they chose to count?

  And despite the small numbers in the other two counties and their falling well withing the 95% confidence interval, it is getting remarkably repetitive that at least 90% (or it 100%) of the hand counts we've seen invariably result in a reduction in Bush's count and an increase in Kerry's count. Increases in both is reasonable, what the machine couldn't read. It should not even be possible to be a decrease unless it's human error on the part of the hand counters (easily possible at those small numbers). But it's getting awfully darn repetitive that every last ballot that conveniently was not counted the first time is a vote for Kerry. And they are just small enough numbers to not be suspicious. At least not until it's a pattern in every frelling precinct!

  by NotInMyName on Tue Nov 30, 2004 at 12:47:09 AM PDT

  •  Yes, what is the statistical likelihood of this?!

    "[I]t is getting remarkably repetitive that at least 90% (or [is] it 100%) of the hand counts we've seen invariably result in a reduction in Bush's count and an increase in Kerry's count"

    Hasn't anyone done a statistical analysis on the likelihood that virtually all the anomalies so far detected have been in Bush's favor, both in terms of overcounts for Bush and undercounts for Kerry?  We have examples of this in several different states now, with very different demographics (e.g., Ohio, Indiana, New Hampshire, Florida).

    As I mentioned earlier, I'm not a statistician, but I find it hard to believe that this lopsided weighting of the anomalies to favor one candidate can be attributable either to random chance or to particular human and/or machine errors always working in one direction.

    Alternatively, are we only hearing about anomalies which have benefited Bush?  Does anyone have information on the opposite (I've only heard of a couple of instances of Republicans downticket benefiting from recounts, e.g., in one of the New Hampshire counties, but not at either the presidential or U.S. senate level)?

    by GreekGirl on Tue Nov 30, 2004 at 01:24:46 AM PDT

    •  there was a count

      error in bush's favor in utah, @8k votes.

      by Furious Hal on Tue Nov 30, 2004 at 02:09:30 AM PDT

      •  Do you have a link for this?

        n/t

        by MyName on Tue Nov 30, 2004 at 09:32:59 AM PDT

        •  probably

          i would do the same thing as you: google it.

          by Furious Hal on Tue Nov 30, 2004 at 02:03:58 PM PDT

    •  remember the bubble

      on Kos you just don't hear about the stories where a glitch caused Kerry to get more votes that were later discovered and corrected.  It just doesn't fit the script so people here don't report it- and it certainly never makes it to the front page. I know there was at least one country in florida that had an 8000 vote glitch in Kerry's favor, but that just isn't a story you will ever see here.

      by Damon on Tue Nov 30, 2004 at 05:44:21 AM PDT

      •  wow!

        Really?  Because I've been looking at non-bubble blogs and haven't seen it...I'd love to see a link, or any other information.  I'm totally serious, because this is something I too have found troubling.  The imbalance in the effect of glitches has been mentioned numerous times here, and any evidence to the contrary would be greatly appreciated.  Which county?

        by Chicago Lulu on Tue Nov 30, 2004 at 07:14:40 AM PDT

        •  quick link

          here is the first google hit on kerry extra votes florida.  I don't know what the affiliation of the site is but they link to the original article (I would link to that but it requires registration)  It was Orange county by the way.

          http://www.atsnn.com/story/97250.html

          original news source:

          www.orlandosentinel.com
          The cause of the error, Orange officials said Thursday, was a software program that could not tabulate more than 32,767 votes in a single precinct. On election night, officials anticipated the problem and adjusted for it, deputy election official Lonn Fluke said Thursday.

          But the next day, workers failed to account for the glitch while posting precinct results online. When absentee-ballot totals exceeded the limit in one precinct, the software caused additional votes to be subtracted from Bush's total.

          A similar discrepancy affected vote totals posted online for the U.S. Senate race between Republican Mel Martinez and Democrat Betty Castor. But neither online counting problem made it into the real totals sent to Tallahassee, election officials insist.

          "The election results we certified to the state are correct," Fluke said. The presidential and U.S. Senate absentee results posted online were "garbage."

          by Damon on Tue Nov 30, 2004 at 07:24:32 AM PDT

          •  Interesting

            Thank you.  I hadn't seen that one, probably because it was caught before the official results went up.   There were definitely "bi-partisan"? glitches, and the counting backwards one would have affected both parties randomly.  For example, that was caught in Broward county as well, and overturned a ballot initiative, I think in the other direction.  Another example of equally distributed problems would be the loss of votes,  what happened in Carteret County, NC.  One would expect overall for each candidate to be affected about the same, over a large sample.   These are examples of generic poor programming- which must be fixed, nonetheless.  I'm surprised that the national media isn't covering these issues- they seem fairly non- partisan and straightforward.

            What most of us are talking about here are problems like the one in Indiana, where straight party Dem votes were switched to Libertarian.  That's not something that happens by accident, or because of "memory size".  Those problems are disproportionately tilted in Bush's favor.    Absent a  generic programming problem- which hasn't come to light as of now- what caused these anomalies?  That's not an unreasonable question to be asking.

            As to the link you provided, a small point.  You do realize that the 2nd story on that site is not a glitch in favor of Kerry, but the opposite, right?  The retally gave Kerry 22000 extra votes- that is, when the problem was caught, not before.

            by Chicago Lulu on Tue Nov 30, 2004 at 08:02:31 AM PDT

          •  Um, a problem with that analysis...

            In this case the 32,767-vote-limit glitch hurt Bush. One of the rare instances. But, this 32,767-vote-limit glitch has come up several times in this election.

            Democratic voters in more populated areas are more likely to run into this glitch.

            So this glitch hurts Kerry, not Bush.

            by bejammin075 on Tue Nov 30, 2004 at 10:02:18 AM PDT

•  incomplete

  Without getting into the broader issues of conspiracy mongering and evidence interpretation, I don't find your analysis particularly meaningful because that 4% isn't that big a difference, and regardless, would only be suspect if the 60% counted were chosen as a random sample - i.e., each vote was equally likely to be included in the Herald's recount.

  This is highly doubtful because the votes are probably stored by precinct or stacked by the time of day people voted, or something like that, which means the sample was skewed in some way (and since they "stopped at 60%" they apparently were not counting with the intention of randomly sampling voters).

  by daunte on Tue Nov 30, 2004 at 12:50:45 AM PDT

  •  Which begs the question:

    Why the devil did they all of a sudden stop counting in that one county?  If 60% of the county was finished, and you had that counting discrepancy, why would you all of a sudden stop?

    That makes absolutely no sense.  Either they were just plain scared of what they were finding, or they were covering something up.  Which one of these options do you think is the case?  Because nobody would stop just for the hell of it at that point.

    by JohnnyCougar on Tue Nov 30, 2004 at 01:18:11 AM PDT

    •  Or

      maybe they were lazy. Or maybe they had a deadline. Jeez, if the Herald can access them I'm sure someone else can. It's not a cloak and dagger thing.

      by daunte on Tue Nov 30, 2004 at 01:50:49 AM PDT

  •  My main point

    ...which apparently hasn't come across, is that their data does not support their conclusion that there is "no flaw."

    I'm not claiming that Kerry actually won Suwannee County by 1200 votes. We don't know that, because they didn't count the damn votes. But in the same way that I can not definitively conclude that Kerry got 1200 more votes in Suwannee, they can not conclude that there's no meaningful difference in Suwannee.

    From another angle, the reporters are basically using this incomplete data to say that there's nothing wrong with the vote in Florida and voting activists should move on. In direct contradiction to that, I would say that their data compels the need for more counting.

    Follow me?

    by pointsoflight on Tue Nov 30, 2004 at 01:36:07 AM PDT

    •  no flaw

      If there was no flaw in 100% of the precincts in 2 of the counties, and no flaw in any of the precincts they counted in Suwanee isn't that enough for you?  If they counted the precincts in Suwanee that had the most democratic registrations and still found no problems there really isn't any value in counting the rest of the county.  At some point they can see there is no issue and they aren't going to pay for further research.  The problem you are still making is that you are looking at 60% of the votes as if you needed to count the whole county at once to verify that the numbers were coming out correctly.  In reality, that would be done on a precinct by precinct basis.

      by Damon on Tue Nov 30, 2004 at 06:09:58 AM PDT

      •  Fine.

        Why, then, don't they give the official totals from the precincts they counted for comparison? They did side by side comparison in the other counties, but here which just get the vague, unsubstantiated claim that the counts "nearly matched the county's official tally."

        Again, they don't provide data to support their conclusion.

        by pointsoflight on Tue Nov 30, 2004 at 07:25:20 AM PDT

        •  they were looking for a story

          You need to understand that the whole point is that they were looking at the data to find a story.  Once they counted enough precincts to see that there wasn't any story there, there was no need for them to carry out the academic exercise of counting the rest of them.  Doing that count cost money and time and once its a dead end they have absolutely no reason to keep going.  I agree that they should have posted their data on the website and if you ask them for the precinct level data they should give it to you, but I really just don't think they messed up by not following through with the rest of the analysis once they determined that there wasn't a problem.

          by Damon on Tue Nov 30, 2004 at 07:39:50 AM PDT

          •  Trying to find a story?

            By looking at the least suspicious counties in the state--ones that statisticians already showed weren't the problem areas? They cherry picked the counties most likely to show minimal differences from the original counts! If they wanted a story, why not count ballots in the precincts where statisticians tell us there are huge anomalies? Why not count ballots in the precincts where blackboxvoting.org has already uncovered suspicious activity and vote totals that don't match?

            The Miami Herald was not looking for a story, they were looking to shut down criticisms of the election. Their choice of counties shows this. Their methods show this. Their premature conclusions show this. The rhetoric they include in their article shows this. And their history (the publication of several articles in 2000 that were also shown to be very misleading) also shows this.

            by pointsoflight on Tue Nov 30, 2004 at 08:38:19 AM PDT

      •  Another thing...

        ...you offer an unsubstantiated explanation based on differences across precincts. Have you even looked at the precinct level data?

        I have.

        by pointsoflight on Tue Nov 30, 2004 at 08:45:48 AM PDT

        •  Show us

          Then show us ONE precinct that had a major shift in bush's favor.  If you have looked at the precinct data then show us where the problem is.  I don't believe for a second that you have looked at the precinct data and found any problem whatsoever.

          by Damon on Tue Nov 30, 2004 at 09:12:01 AM PDT

          •  What hypocrisy.

            You can offer an explanation based on differences across precincts without provided ANY support for that argument, but if someone else has a different explanation, you insist that they prove it?

            by pointsoflight on Tue Nov 30, 2004 at 10:06:08 AM PDT

          •  Miami Herald doesn't give the

            recount data by precinct.

            by white blitz on Tue Nov 30, 2004 at 10:20:51 AM PDT

            •  the data

              He is the one claiming he looked at the data on the precinct level.  Its not true but he is the one claiming it.

              by Damon on Tue Nov 30, 2004 at 11:07:16 AM PDT

              •  I beleive

                he claimed to have seen precinct level data of the official polls, not the hand count.

                by white blitz on Tue Nov 30, 2004 at 11:18:08 AM PDT

              •  Whatever.

                Now resorting now to calling people a liar?

                I have looked at the precinct data and have even crunched the numbers with that data. But you're the one making claims about the differences across precincts, so why aren't you crunching the numbers? You're quick to be critical of others ideas, yet you don't support your own. How hypocritical is that?

                What I know for a fact is that there's only one way that Bush's total in the hand count can "catch up" to the official total, and that's if the Miami Herald chose to count those counties with the smallest Bush margin, and deliberately did not count the 10 or so counties with the largest Bush margin. That's the only way.

                Now maybe that's what happened, but that alone is odd and open to suspician. Why would you only do a recount in precincts where Kerry faired a bit better, and ignore those precincts where Bush won by the biggest margins?

                And regardless of all of these details, my main point still holds true. Their claim that there is "no flaw in Bush's state win" is still ridiculous and is obviously driven by political motives. Yeah, lets only hand count in those counties that are known to be dixiecrat counties, and that statisticians have already said aren't a big problem. And lets cherry pick the precincts where Bush faired least well. And lets include a bunch of rhetoric and snide remarks from partisan election officials. And lets generalize from 0.3% of the vote in three small, dixecrat counties to the entire state. And lets ignore the counties that statisticians have pointed out to be a problem, like Miami-Dade and Broward. And lets ignore the counties where very suspicious activity has already been uncovered, a vote tallies do not match, like Volusia County.

                Yeah, their report is right on the money. It shows conclusively that there's no flaw in the whole state. Right. And I have some land to sell you.

                by pointsoflight on Tue Nov 30, 2004 at 11:45:32 AM PDT

                •  The formula

                  And lets include a bunch of rhetoric and snide remarks from partisan election officials.

                  It was the highly partisan language from the SOS's that concerned me the most. Echoing the phrase "the liberal senator from Massachusetts" seemed entirely inappropriate to me.

                  by white blitz on Tue Nov 30, 2004 at 12:09:01 PM PDT

                •  BTW

                  Since you have seen the precinct numbers, could you share with us some info about the distribution? It would be helpful to know what % Bush had in his strongest and weakest precincts.

                  by white blitz on Tue Nov 30, 2004 at 12:45:27 PM PDT

                  •  Here's the data...

                    http://www.suwanneevotes.com/new_page_3.htm

                    Kerry won one precinct, Bush got 65-80% in all other precincts.

                    by pointsoflight on Tue Nov 30, 2004 at 05:58:24 PM PDT

      •  They have you brainwashed don't they?

        Why would they complete counting the counties that match, and quit counting the county that didn't???  That makes absolutely no sense at all.  The fact that there was that much of a discrepency should have made national news, and, at the very least, they should have FINISHED THE COUNT!!!!

        One possibility - maybe they did finish the count, found it way off, and claimed to not have finished the count, so as to discredit the data....

        And, as the diary says, the data does not support the headline.

        As a citizen of the U.S., you should be just as concerned as anyone if votes aren't being counted correctly.

        by MyName on Tue Nov 30, 2004 at 09:11:15 AM PDT

        •  argh

          The unfinished county DID match.  All that matters is whether every individual precinct matches, not whether the 60% of the precincts were representative of the entire county.  If there was a problem in any of the precincts, that is the news.  Not that every precinct they looked at showed no problems at all.

          by Damon on Tue Nov 30, 2004 at 09:13:56 AM PDT

          •  ????? asdf

            No they didn't!

            by MyName on Tue Nov 30, 2004 at 09:34:17 AM PDT

          •  I got a 100 onn my ACT in math:

            Bush original: 11153
            Kerry original: 4522
            Bush hand count: 6140
            Kerry hand count: 2984

            11153 + 4522 = 15675 (Original)
            Bush 11154/15675 = 71.151%
            Kerry 4522 / 15675 = 28.848%

            6140 + 2984 = 9124 (Recount)
            Bush 6140 / 9124 = 67.295%
            Kerry 2984 / 9124 = 32.704%

            An almost 4% difference!
            And 60% is enough of a representative sample...

            by MyName on Tue Nov 30, 2004 at 09:44:52 AM PDT

            •  apples and oranges

              When they say they matched they are talking about on the precinct level.  They didn't bother to post the matching data on the precinct level because there were no anomalies.  All you have shows is that they counted the more democratic precincts first.  If they counted all the precincts in the other 2 counties and found no problems and there were no problems in any of the precincts that they did count in the 3rd, why on earth should they keep looking?  They are the ones paying the bill for the study, once there is no story why should they keep paying?  For anyone not conviced there is a problem right from the start- its obvious that this is how you would go though and do the recount.  

              by Damon on Tue Nov 30, 2004 at 11:11:50 AM PDT

          •  Read the diary

            The diary shows the offical results give Bush 4% more than the hand count. I looked up the official county results and noticed that Kerry gained votes in the handcount. I figured that the total differance in the vote margin was greater than 7%. That is NOT a match.

            by white blitz on Tue Nov 30, 2004 at 09:47:49 AM PDT

          •  Union and Lafayette

            Had changes in vote margin of 1/4% to 1/3%. That's a little bigger than I would expect for optical scanning, but not obscene. A change in margin of 7% is huge.

            In our county, the scanners are at the polls and are set to reject ballots that it can't read, up to 3 times. I was under the impression that this setting was supposed to be in place state wide in 2004 for all the FL counties which used paper ballots.

            by white blitz on Tue Nov 30, 2004 at 09:58:20 AM PDT

          •  Show me anything in their article....

            that shows the unfinished county matched. Anything at all.

            You can't. They give the side-by-side comparisons for the two counties they counted in full. They do NOT give the official vote tally for the precincts they counted in the unfinished county. So how can you say the hand counts matched? Matched what?

            Show me the money.

            by pointsoflight on Tue Nov 30, 2004 at 10:03:58 AM PDT

            •  proof

              You are the one claiming to have looked at this data on the precinct level.  I think you are full of it.   Its obvious that what they would need to match is on the precinct level to everyone else.  You are looking for anomalies where they don't exist so you aren't going to look for the blatantly obvious answer. But once again, you say you have looked at this data on the precinct level- show us one situation where a precinct in this county was counted differently than the hand recount.  But this diary is garbage because no such precinct exists which is why they stopped counting.

              by Damon on Tue Nov 30, 2004 at 11:14:49 AM PDT

  •  4% would swing the state

    If there were a 4% discrepancy across Florida. So the question of whether these numbers actually match on a precinct-by-precinct basis become very important -- yet for some reason they didn't provide that information.

    by karenvanhoek on Tue Nov 30, 2004 at 01:51:54 PM PDT

  •  inconsistent logic

    You can only draw conclusions from the 60% (which the reporters did) by assuming it is a meaningful sample of total votes cast.

    But if you assume the 60% is representative of total votes cast then there is definitely a problem.

    by sophiebrown on Tue Nov 30, 2004 at 01:56:06 PM PDT

  •  You do realize that polls and exit polls use

    a miniscule sample, don't you? The final weighted exit polls are always dead on, and their sample is 13,000 of 120 million voters, and you're going to complain about a 60% sample?

    by upstate NY on Tue Nov 30, 2004 at 02:21:24 PM PDT

    •  You are confusing

      the difference between a 'sample' and a 'random sample.' Please re-read my comment.

      Yes, I can measure a population with a small random sample, but if I select the sample based on geography or some other non-random factor, then I am likely not measuring accurately. For example, if I take a survey of a million people in New York, California, and Illinois, I will have a very large sample and find that Kerry 'won.'

      by daunte on Tue Nov 30, 2004 at 03:26:50 PM PDT

•  Excellent analysis

  and if the pattern you've identified continued statewide, then Bush has a problem.

  by southlib on Tue Nov 30, 2004 at 01:21:07 AM PDT

•  Wake These Morons UP!

  We need to write to every paper that carried this nonsense.

  Here's what I wrote to the Herald. I hope you join me.

  To: [e-mail] [e-mail] [e-mail] [e-mail] [e-mail]
  Subject: Herald's Fuzzy Math

  Do you people own a calculator?

  You falsely claim your count of Suwanee County ballots "nearly matched the county's official tally."

  This is simply false.

  Your count discovered that only 67.3% of ballots were for Bush and that 32.7% were for Kerry. This differs from the "offishyl" result by several percentage points.

  Extrapolate your results to the entire county and you would have "found" (had you looked) a swing from Bush to Kerry of nearly 8% or approximately 1200 votes.

  And that's after examining only 00.12% of the votes in the state.

  Hello!?! Is anybody home!?!

  by Dusty on Tue Nov 30, 2004 at 01:29:40 AM PDT

•  Media sign language..?

  ... I'm begining to wonder if we are not seeing some media in the US acting somewhat like the sign-language reporter in Ukraine.

  If you haven't read or seen the story, this brave women was supposed to sign the election results in the bottom right corner of the screen during news reports - as the evening wore on, she started signin "This is all a lie ... you may not see me again, but I can't continue with this deception.." etc.

  Now, our media is no way that brave, and our election problems are less easy to characterize that blatant ballot-box stuffing.  But stories like the Miami Herald piece seem to me to be telegraphing "Hey, look ... something's wrong here!" but trying to do so under the radar, hoping smart readers will pick it up.

  Some of the Ukraine reporting on the MSM cable news channels gives me the same impression - there is a not-so-subtle tone of irony and sarcasm coming from some of he reporters when talking about election polls showing the vote is unreliable, etc.

  I've worked for a large state gov't ... and I know what it's like to be 'muzzled' when speaking in public... and I know how you can try to get your message across in other ways.

  by Timoteo on Tue Nov 30, 2004 at 04:25:46 AM PDT

  •  Signing trouble ?

    Cool story! In Brazil one of the newspapers used to put warnings to protesters on the weather report: "Today: Dark, cloudy, oppressive, stay indoors or you will be struck by it's fury".

    Ukraine journalist signs silent protest - A Bravenet.com Headline News Service

    by lawnorder on Wed Dec 01, 2004 at 11:20:33 AM PDT

•  I think you missed the point

  They don't have to count all data on the county level- all they have to do is count all data on the precinct level and see if they matched.  If they got through 60% of the precincts in Suwanee county and had no problems at all, after going through every precinct of the other 2 counties and finding nothing it seems to me to be a pretty logical conclusion that there isn't any need to contintue.  Its not like this was a legal recount- they were just investigating the story.  Its certainly possible that they didn't even do a random recount of the precincts- just the precincts that had the highest percentage of registered democrats.  

  by Damon on Tue Nov 30, 2004 at 04:44:19 AM PDT

  •  Zero!

    Its just rediculous that anyone would be so afraid of an opposing point of view that they would give this post a zero.  Learn to argue with actual words and not ratings you hope would be invisible.  Nothing in this post deserves to be censored- get a thicker skin!

    by Damon on Tue Nov 30, 2004 at 06:34:39 AM PDT

  •  ratings abuse

    Mambo- What on earth did I do in this thread that would cause you to go through and rate every one of my posts a zero?  What did I do that was the least bit inflammatory, derogatory, or disruptive?  Its really sad that you don't trust your words to say that I am wrong and have relied on the ratings system to hopefully erase the posts simply because you don't agree with them.  

    by Damon on Tue Nov 30, 2004 at 07:57:11 AM PDT

•  PLEASE everyone WRITE the Herald!

  Don't just talk about this on here, write them!

  Point out their error, and how their numbers in fact point to the exact opposite conclusion of what the article and headline said.  

  If they get enough letters, they will be forced to print one of them at least and perhaps revisit the subject.  

  Quick caveat: As I read the article, they could have a fallback position on Suwanee County.  Perhaps what they are saying is that the percentage of 67.3% matches almost exactly that originally recorded in the 60% of the precincts they recounted, not the percentage of Bush vote overall.  I don't think this is a particularly good defense of a poorly researched story that was obviously a cheap and quick attempt to knock down "Florida fraud" allegations, BUT I want to point out the Herald may very well say this.  So be prepared to respond to that possible counterclaim.

  It is good that the media will get to recount these votes (eventually), but how do we know the ballots were secured?  What assurances do we have that ballot box-stuffing could not take place?  Even with a paper trail, it seems to me that some sort of voter ID number registered BOTH at the polling place AND on each ballot cast is the only way to eliminate this potential fraud.  Hope to see it one day (in the distant future).

  by Rithmck on Tue Nov 30, 2004 at 05:11:08 AM PDT

  •  WRITE THE HERALD

    Amen. Write letters to the editor.

    by mcta0001 on Tue Nov 30, 2004 at 07:08:32 AM PDT

  •  I did

    Here's the letter:

    Re your Nov. 28 article, "No flaw found in Bush's state win": I was struck by the fact that neither of your reporters found anything odd about the votes for President Bush that appeared in the official count but somehow disappeared when they counted the ballots by hand.  According to your article, in Union county, President Bush's handcounted total went down 3 votes; in Lafayette County his totals went down 8 votes.  In both counties, Senator Kerry's totals went up, as one would expect since human eyes can determine voter intent even when an optical scanner cannot.  So the question remains, how did these supposedly properly-functioning scanners find more votes for the President than your reporters could find counting the ballots by hand?  I'd love to hear the explanation.

    by katerina on Tue Nov 30, 2004 at 10:02:39 AM PDT

•  Send this info. to Olbermann!

  Has anyone sent this to Keith Olbermann yet???
  Maybe if he gets enough emails, it'll make the show tonight?
  Let's do it.
  Sharon

  by raleighsharon on Tue Nov 30, 2004 at 05:36:19 AM PDT

•  Thanks for the catch

  I especially want to thank you for the non-reg. link. I put the difference at 7.4%. This is the difference in vote margin between official and semi-counted results. Remember the percentage of each candidate is changed.

  For the official results to hold, Bush would need 75% of the remaining ballots and Kerry would need 23%. The partial results have Bush at 67% and Kerry at 32%.

  It does make you wonder why they stopped counting. It seems like another paper willing to do a full recount could really slam the Herald.

  by white blitz on Tue Nov 30, 2004 at 05:40:36 AM PDT

  •  why they stopped counting

    If they looked at the precincts with the highest democratic registration, and found absolutely no problems in those after finding no problems at all after counting all the precincts in the other 2 counties why wouldn't they stop?  Thats still the flaw- you don't have to count all the votes in a county to verify that they were counted accurately, just all the votes in any precinct to make sure they match.

    by Damon on Tue Nov 30, 2004 at 05:47:09 AM PDT

    •  You find nothing odd...

      about the fact that for the two counties they counted in full, they gave the exact numbers of the hand count and official tally for side-by-side comparison, yet with Suwannee, they don't give us a side-by-side comparison? You think it's sufficient that whereas they give exact numbers for Union and Lafayette, they merely say the numbers "nearly matched" in Suwannee?

      What does nearly matched mean? Nowhere do they say they counted X number of precincts and the totals in those precincts matched the official precinct tallies. You infer that, but nowhere do they say that. And if precinct after precinct matched, wouldn't they make it point to spell that out? We have no idea what they mean by nearly matched. We don't know how big of a difference qualifies as nearly matched, and we don't know what numbers they are comparing. They could even be looking at the 67% vs. 71% numbers that I gave and consider that a near match, when in fact that is a significant deviation.

      by pointsoflight on Tue Nov 30, 2004 at 10:17:48 AM PDT

      •  it was obvious

        To anyone who went into this with an open mind, this was completely obvious.  It was borne out by the person below who went ahead and asked the reporter for the numbers (as I suggested you should do).  Its the difference between looking into something looking for something so hard you will find it even when it is not there.  If you just think about the most obvious explanations first then check those out you will have a ton more credibility.

        by Damon on Tue Nov 30, 2004 at 03:24:40 PM PDT

•  anyway-

  I bet if you wrote the author they would give you the data they collected on the precinct level for the county you are interested in.  Then you can go through the precincts to see whether there was any anomaly and whether the remaining precincts would make up the difference in votes.  

  by Damon on Tue Nov 30, 2004 at 05:54:32 AM PDT

•  Good diary.

  Enough anomalies in enough states have been collected to be offered as "proof" for some, in the sense that this election, because of shoddy security, could have been tampered with.

  Also, no safeproof auditing, folks. No auditing.

  Our votes are privatized, and most with no verifiable paper trail,  and this is red flag enough for me.

  It is time to organize around this issue, and get something done...yesterday.

  by scorpiorising on Tue Nov 30, 2004 at 06:02:53 AM PDT

•  Did Bush lose popular vote nationwide?

  You know, the real conspiracy theorists are putting together all of these discrepancies, adding up some numbers and starting to wonder if that difference in the popular vote is real. I know 3.5 million seems like a lot, but take 200,000 here, and 400,000 there and you certainly cut into it.

  The truth is likely that we will never know with any certainty what the real intent of the voters was. But I would be willing to say that if numbers are significantly in error (note: not necessarily fraudulent) then the question of whether they can be extrapolated from Florida, Ohio and New Mexico starts to be at least interesting.

  We know Rove desperately wanted to win the pop. vote as well as the electoral college. What better strategy could there be than to seriously pad solid red states that no one would have the least interest in challenging?

  Sad note: regardless of the outcome of any of this, there is now a significant proportion of the voting public (and a highly well-informed segment, too) that distrusts the fundamentals of American democracy.

  by thingamabob on Tue Nov 30, 2004 at 07:06:23 AM PDT

•  My letter to Meg Laughlin of the Herald

  Dear Meg Laughlin,
  
  I am writing you concerning your article from Sunday, Nov 28: "No flaw is found in Bush's state win". I have a question regarding your counting of votes in Suwannee County. In your article you state you counted 9124 votes, which represent 58% of the votes cast and find 67% went for Pres. Bush. This is to be compared with the actual tally of 15675 votes, 71.1% for Bush.
  
  Now, if I simulate counting 58% of the votes 50000 times, I never get a ratio of Bush to Kerry votes comparable to what you report. This would on the face of it suggest your results are inconsistant with the "unoffical" reported results. What my analysis fails to consider is the precinct by precinct variations in the votes. There are enough precincts where Kerry's margin is high enough, if I could consider this information, to show your partial tally is consistant with the county totals.
  
  Thus, if I might trouble you, would you consider providing me the precinct breakdown of your counts. I would enjoy to be able to convince myself that your claims are valid.
  
  I look forward to your reply,
  Nicholas Phillips, PhD

  by Nicholas Phillips on Tue Nov 30, 2004 at 07:54:27 AM PDT

  •  Her reply...

    We counted in precincts 1, 2,5,7,12,15 and early and absentee ballots.

    Bush original: 6138
    Kerry original: 2975
    Bush hand count: 6140
    Kerry hand count: 2984
    So no problems here, move on to Ohio.

    by Nicholas Phillips on Tue Nov 30, 2004 at 02:05:48 PM PDT

    •  Yes, I got same nums from the 2 reporters

      For the 6 precincts they counted, plus absentee (whole county), plus early votes (whole county), the handcount numbers match up well to the county's own count.

      2 new votes for Bush, 9 new for Kerry. The reporter saw some spoiled ballots for Kerry, in poor precincts, where the intent was clear but the voter's mark was not the way a machine would read it, so only a handcount would pick it up.

      In the paper edition, a graphic was included with the comparison numbers easily viewable for Suwannee, but those comparison numbers weren't in the text, and the graphic's not on the web pages.

      So the count looks ok, so far. The reporters did the handcount Monday, Tuesday and Wednesday until the elections supervisor had to go home and get ready for Thanksgiving.

      Recommendation: People who say that numbers are awry should FIRST run it by someone who is learned in STATISTICS, at least a little bit.  Sometimes the stats person would set someone straight before you make claims that fall apart. (I'm thinking about the person who claimed the optical scan counties in Fla. showed more deviations than the touchscreen counties.  A statistician would have pointed out that the touchscreen locations were associated, not just with a different D/R voting pattern, but also with a different demographic (more urban, densely populated, large counties, and southern tip.) The population difference would enter into and partly explain the pattern-difference -- and could cancel out the machine difference spotted on the surface.

      I may post this additional Miami Herald data as a short diary so that people see it and don't run away with unfounded suspicions about Suwannee County.  I hope duplicating this as a Diary post for that reason isn't bad Kos etiquette.

      by joan reports on Tue Nov 30, 2004 at 02:36:23 PM PDT

      •  I am learned in statistics

        part of my day job. Please feel free to contact me if you want some analysis done. I keep an eye out for places where I can apply my powers for good, e.g. this diary.
        
        [e-mail]
        

        by Nicholas Phillips on Tue Nov 30, 2004 at 03:01:34 PM PDT

        •  No slight intended, and

          not to you.

          I was only trying to point out to the original diary author that numbers that look anomalous should be passed by a "numbers person" before we draw conclusions. Otherwise, we can raise a fuss and have egg on our face and get people excited over nothing.

          I'll keep you in mind when I have numbers questions.  I have some stats/econometrics background but I'm not a full-fledged statistician by any stretch.   Thanks Nich.

          by joan reports on Tue Nov 30, 2004 at 03:35:21 PM PDT

    •  Not so fast

      Recount results of Bush (+2) and Kerry (+9) are very similar to full recount gains in the other two counties. This is not resolved sufficiently to close the book and move on to Ohio.

      by newsouth on Tue Nov 30, 2004 at 02:37:59 PM PDT

    •  In 2000 these precincts ...

      ... went Bush 61%, Gore 36%, other 3%. Precinct 1 was the only Suwannee precinct to prefer Gore over Bush.

      The non-hand counted precincts voted Bush 67%, Gore 30%, other 3%. Now there are several factors to compensate for -- about 2500 more votes in 2004, early voting from all precincts included in the hand count, and there are likely more absentee votes. But a small pro-Bush discrepancy doesn't seem out of line with the 2000 results.

      A full count would be a lot more convincing.

      by black dog barking on Tue Nov 30, 2004 at 04:55:35 PM PDT

•  More disappearing votes

  This is a significant analysis in several ways, but the one that caught my eye is that once again we have disappearing votes in a optiscan county!  Notice that in both counties where full count was made, Bush actually lost votes.  

  How does this happen?

  I commented several times (