The Significance of a Corrupted Recount of the Cleveland 2004 Presidential Vote [View All]

[Time for change]

It was well known in the days prior to the 2004 Presidential election that a Kerry/Edwards victory was almost certain if they carried either Ohio or Florida. Central to Kerry’s chances of carrying Ohio was a good turnout in Cuyahoga County, and especially in Cleveland, the most heavily Democratic city in the state.

Therefore, Cleveland was very heavily targeted by Democrats, with a massive voter registration drive and an intense voter turnout effort on Election Day. And these efforts appeared to be highly successful by Election Day, with 230 thousand new voters registered in Cuyahoga County in 2004, the success of the voter turnout effort evidenced by the presence of impressively long voting lines throughout Cleveland, and the Ohio exit poll showing a comfortable Kerry lead throughout the day. Even CNN’s right wing hack, Robert Novak, acknowledged that it would be an uphill climb for Bush.

But disappointing returns throughout the evening meant that by late evening, despite the exit poll continuing to show a comfortable Kerry lead, the hope for a Kerry victory depended on Cuyahoga County, and especially Cleveland. But this remaining hope soon faded, as it became clear that the voter turnout from Cleveland was in fact miserably low, and by noon the next day John Kerry conceded the election.


What went wrong in Cleveland?

The Democrats were not the only major Party that heavily targeted Cleveland. The Republicans also appear to have targeted it – but in a different manner. The 230 thousand new voters registered in Cuyahoga County did not show up in the final official voter registration figures because of massive, apparently illegal purging of the voter roles, which was apparently targeted at Democrats, as discussed in detail in this thread.

But that wouldn’t explain the low “voter turnout”, which is calculated as the percent of registered voters who vote. Why, in the face of widespread expectations of exceptionally high voter turnout in Cleveland, would official voter turnout be so low?

The voting lines in Cleveland were exceptionally long, both according to observers in Cleveland and according to official reports. Analysis of reports to the national Election Incident Reporting System (EIRS) showed that not only the highest number of reports of long lines, but also the highest rate of long lines per registered voter, came from Cuyahoga County. Furthermore, the great majority of these reports came from Cleveland, which contained less than a third of the registered voters in Cuyahoga County, and where voters heavily favored John Kerry.

Yet despite the long lines all over Cleveland, official voter turnout was not only not recorded as high, but it was quite low compared to elsewhere in Ohio. According to the Cuyahoga County Board of Elections website, the voter turnout in Cleveland was only 53.14%, compared to 73.41% elsewhere in Cuyahoga County and about 70% in the rest of Ohio.

What would explain the combination of exceptionally long voting lines and low turnout in Cleveland? In some counties where voters voted electronically, the explanation was too few voting machines for too many voters. That was the explanation in Franklin County, for example. But Cuyahoga County used punch card voting, not electronic voting in 2004. According to data used to produce the Democratic National Committee report on the 2004 Ohio Presidential election, only counties that used electronic voting were characterized by long lines caused by too few voting machines. Counties that used either optical scan voting or punch card voting did not experience that problem.

So, without the excuse of too few voting machines as a plausible explanation for the long voting lines in Cleveland, the most plausible remaining explanation is an exceptionally high turnout. This explanation is consistent with the massive efforts that went into obtaining a high voter turnout in Cleveland, as well as observations on Election Day.

And yet, official voter turnout in Cleveland on Election Day was exceptionally low, rather than exceptionally high.

What would explain a very high real turnout of voters in Cleveland, in the presence of a very low official turnout? One possibility comes to mind: Votes from Cleveland precincts could have been deleted by the Cuyahoga County central tabulator after being reported there.


What might the effects have been of an artificially low vote count in Cleveland?

Suppose that the voter turnout in Cleveland (as opposed to the official count) was as high as in the rest of Cuyahoga County. That would have meant that more than an additional 20% of Cleveland’s registered voters would have voted, resulting in an additional 65,563 votes. Cleveland voted very heavily for Kerry – 81%, compared to 15.4% for Bush. Assuming that same rate for the deleted votes, that would have meant an addition of 43,009 net votes for the Kerry/Edwards ticket.

Nobody knows what the real loss of votes for the Kerry/Edwards ticket might have been in Cleveland. Maybe the voter turnout in Cleveland really was low. If so, I don’t know what the explanation would be for the long voting lines all over the city, but perhaps there is an explanation that nobody has thought of or publicized. Or, maybe the official turnout results on the Cuyahoga County BOE website are wrong. Maybe those results didn’t include absentee ballots, as at least one person has suggested.


How would we ever know?

One good way to find out if votes were deleted by the Cuyahoga County tabulator would be to compare the individual precinct totals, as reported by precincts prior to being sent to the Cuyahoga County central tabulator (pre-tabulator results), with the official results reported after the central tabulator added up the votes in all the precincts (post-tabulator results).

I tried numerous times to obtain the pre-tabulator results from Michael Vu, and he promised them to me several times, but he never delivered on his promises. Consequently, I collaborated on this issue with a computer science professional, Ron (last name withheld), who works for Ray Beckerman’s Ohio Project. Ron’s initial audit of 15 precincts identified an apparent vote undercount of 163 votes that resulted in a net loss to the Kerry/Edwards ticket of 140 votes. Ron tried to proceed with a more thorough audit of the Cuyahoga County vote, but he ran into numerous technical problems, and I doubt very much whether he was ever able to complete it.

There was also, as we all know, a “recount” of a 3% “random sample” of the Cuyahoga County votes. However, we now know that that recount appears to have been rigged, and three elections workers are facing criminal charges for that.

Additional evidence of a corrupted recount comes from the observations of the Green Party observer at the recount, who noted:

Anomalies were found. Almost all of the witnesses that I spoke with felt that the ballots were not in random order, that they had been previously sorted. There would be long runs of votes for only one candidate and then long runs for another, which seemed statistically improbable to most. From what they were able to get through, witnesses found that signature counts were very much different from the official recorded number of ballots

The significance of the corrupted recount

If my assumptions and calculations are correct – if the Kerry/Edwards ticket did lose 43 thousand votes in Cleveland because the Cuyahoga County central tabulator deleted Cleveland votes, that would not by itself have changed the results of the Presidential election in Ohio, which John Kerry “lost” by 118 thousand votes. However, this analysis may have underestimated the number of deleted votes, for example if it underestimated the voter turnout in Cleveland or if vote deletion was selectively targeted at the most Democratic Cleveland precincts. Additional evidence (See Cuyahoga County results) for that possibility comes from a study that shows that in Cuyahoga County there was an inverse relationship between voter turnout and percentage of the vote for John Kerry, county wide.

Also, there is a great deal of additional evidence of election fraud in Ohio. In Cuyahoga County alone there appears to have been massive illegal purging of registered voters, resulting in a net loss of additional tens of thousands of net Kerry/Edwards votes in Ohio. John Conyers’ report, “Preserving Democracy – What Went Wrong in Ohio”, provides a great deal of additional solid evidence of numerous “irregularities” occurring in the 2004 Ohio Presidential election. And there is much much more.

The prosecutors who are prosecuting the Cuyahoga County election officials for rigging the Cuyahoga County recount have implied that the sole reason for that crime may have been to avoid the excess work of having to recount all the votes in the county. That explanation does not seem very plausible to me. Election workers get paid for their work. If they didn’t have to hand count the presidential votes they certainly would have been given other work to do. Would three people commit a felony simply to avoid a little excess work?

Far more important than the prosecution of those three individual election officials is figuring out what happened, so that measures may be taken to prevent a repeat occurrence – in Ohio or elsewhere. Perhaps the trial will shed light on this. But just as or more important than the trial would be a complete hand recount of the Cuyahoga County vote. I don’t know if the records are still available, but if they are there is no reason not to proceed with a full recount – which should have been done a long time ago.