Vote rigging by machine allocation in Franklin County, Ohio

[jmknapp]

[On the Road]

1. That is Great!

Very clear statistical evidence. Some of this is sure to be the urban vs rural polling places, but it's still unequal protection.

Do you have the raw data by any chance? Is it tied in to the actual vote? That would make an interesting further analysis -- you could show how fewer machines derpressed turnout.

[jmknapp]

4. Good idea... here it is

[jhgatiss]

13. Now....

Can you make it three dimensional with the Kerry vote as another variable? ;)

[jmknapp]

16. Not 3D, but color-coded

[Critical Thinker]

17. Excellent, this is the graph

...that most clearly presents your case, IMO. Great work !!!

[John Q. Citizen]

2. Very cool. The 9 hour poll tax in graphic format!

[ParanoidPat]

8. Well put.....

.....9 hour poll tax indeed! :evilfrown:

[Critical Thinker]

3. Very interesting analysis, but

You may have left out the single most important plot:
what I don't see is an analysis of total ballots recorded per machine vs. Kerry % in precinct

If voter turnout per machine were actually "flat", then it would show that the BoE people had done a good job of providing a "level field" throughout the county.

[jmknapp]

5. Not really

Basically the machines were operating at their maximum capacity in most precincts. So the throughput would be expected to be flat across precincts in the overloaded case. Your analysis doesn't take into account the people deterred by the long lines, walking away never to return.

[esvhicl]

6. Ohio testimony kpfk right now!

http:///www.kpfk.org

Listen to testimony.

[NewYorkerfromMass]

7. only shows a slight bias

Sorry but it does not appear outrageous enough.
The ture test is to simply draw the line at the 50% mark. Now that DOES show more machines in anti-Kerry precincts.

[jmknapp]

9. The way queuing theory works is

that near the maximum capacity of a server, a smallish increase in traffic can cause the queues to balloon.

[Critical Thinker]

11. Plot #3 is the best

OK, I just now saw the plot you posted in Reply 4. IMO, that plot does present good evidence that inadequate machine distribution correlates to surpressed turnout. Need to figure some way to tie the info in plot 3 with the info in the first 2 plots...

[jmknapp]

12. Here's the spreadhseet

...if you want to try your own analysis:

http://www.copperas.com/votingmachines/fcresults.zip (zipped excel file)

[On the Road]

10. It May not be Outrageous

but it's plenty large enough to swing the election. Smaller precincts averaged about 60% turnout, larger ones were under 50%.

The thing that's especially great about this is it's all Franklin County. The comparison is much more apt than if Cleveland and Toledo were shown next to some farming communities.

[NewYorkerfromMass]

14. well then similar tables need to be drawn up statewide

to get a fuller picture. The anecodtal evidence is overwhelming that the Dem precincts got screwed.
We need tangible proof.

[jmknapp]

15. Here's a fuller picture of Franklin County

larger version: http://www.copperas.com/votingmachines

[John Q. Citizen]

18. That's a 20% + swing

[jmknapp]

19. Here's a web page summarizing the results

Thanks for the suggestions, DUers...

http://www.copperas.com/machinery