A Theory on the North Carolina Funny Numbers

I have a theory about the North Carolina votes. The theory comes on the heels of the fine work done by ignatzmouse on the "Funny" numbers that were spit out from the North Carolina absentee versus final vote discrepancies. As ignatzmouse pointed out, the numbers from absentee should carry thru to the final vote and the percentages shouldn't have changed so drastically.

Following is an application of my theory of how the votes may have been altered. Basically, you take the difference in percentage from the absentee and final and use that number on the final tally to find the votes taken from Kerry and given to Bush.

As you will see, the percentages work out at the end to be the same percentages, as ignatzmouse and I both figured they should have anyway.

Now, I enjoy playing with numbers but I am not mathematician or statistician, so there could very well be large errors in my theory. Test it. Tear it apart and prove it's wrong if you can.

The following is from just one of several counties I ran this theory on. Chatham.

************************************************
CHATHAM Optical Scan:ES&S/Precinct

Final
Bush / 12712, 50.1%
Kerry / ,12679, 49.9% = y
total 25391

Absentee
Bush 5015 45.1%
Kerry 6099 54.8% = x
Total 11114
**********************
THEORY APPLIED

25391 total reported votes

6099 Absentee Kerry votes is x = 54.8% of 11114

12679 Kerry final votes is y = 49.9% of 23591

x= 54.8% Absentee Kerry votes 6099
y= 49.9% final Kerry votes 12679
c = 4.9% is difference between x and y

4.9% of 23591 total votes = 1244 stolen votes

1244 sv + 12679 = 13923 new final for Kerry

13923 % of 25391 = 54.8% = x = % of Kerry absentee

****************************
There may be another reason why it comes back around to the same percentage, but damned if I can figure it out, can you?

Should it be

total 25391

23591 total reported votes

Thanks.

I ran this theory on several other counties, it worked on them, too.
But the first time I ran it on Chatham, I had used that wrong number and it never got removed from the original unedited post.

BTW... all these numbers were taken from the NC Elections website after the counties had certified their counts a week after the election.

If the theory proves out, here's how I think they worked it. The numbers from the absentee votes were looked at and an updated source code was sent to the machines after the absentee ballots were counted.

very grateful to you and Ignatzmouse for doing these calculations. I see the implications clearly. Hoping somebody can do something with this. I know enough to know the results are statistically significant, and they may shed light on the questionable results in NC. Anyway, good work, and thanks for the updates.

How many other counties did this work on?

to see how their numbers are worked out.

The same equation was run on five different counties, all using the ES&S system, as far as I can determine. All five worked out real close to the same results. But, as I say below, it may not be a good theory until someone with higher powers than mine is able to examine it.

The numbers work out as just a plain one to one exchange, and there is no credence for the theory.

Can't quite get my head around this since I'm not a mathematician, but since I used the given number from the final, it makes the whole equation into a one to one exchange. It may be best to just let this thread sink until a mathematician can look it over and yell out.

Hey, at least I tried.

I go to bed every night going "how did they do it?" how did they do it?!

If it's because I'm just plain crazy, or all this shit has made me crazy. I know this, I never thought I'd do as much math as I've done the past few weeks.

"If we weren't all crazy, we would go insane" Jimmy Buffet

but it looks like you are simply adjusting the final vote to match the absentee percent. Change the question to this:

How many votes would I need to add to Kerry's total to make the % match his absentee %?

Well, his absentee % is 54.8% and his total % is 49.9% so I'd need to add 4.9% of the total votes, which is 1244. So add 1244 to 12769 and voila! I've made the adjustment.

Sorry, you're just arguing in a circle--it would work out with any numbers, so it doesn't prove (or even suggest) anything.

Your c may be expressed as "the percentage of the vote that one would calculate was stolen on the assumption that the breakdown of the absentee votes should be the same as the breakdown for the total."

Then, if you "correct" for c, you find that the breakdown of the absentee votes is the same as the breakdown for the total, but only because you used that assumption in defining c in the first place.

--MarkusQ