(Huge) delegate vote anomaly in Alabama verified

[drummergirl]

Once again, isn't this to be expected? Smaller precincts should naturally be more noisy because fewer people are voting. As in my previous example, if a precinct only had 10 people voting, yes there would be more of a chance that Paul would get 0 votes. But if he gets only one vote in that same precinct, that automatically puts him at 10%. I would assume that the graphs would become less and less noisy as precinct size increases because if you look at it from a statistics standpoint, the sample sizes are larger.

I find Romney to be the really sketchy one because his peaks at the end are much higher than his peaks at the beginning which clearly shouldn't be the case (but this is nothing new).

no, he's talking about the upper level noise not the left side of the chart.

 

[RonRules]

no, he's talking about the upper level noise not the left side of the chart.

Evidently he didn't' turn his head when I told him to.

 

[drummergirl]

Isn't this to be expected? As precinct size increases, there becomes less and less of a chance that Paul will get zero votes.

For example, its far more likely that Paul would get 0 votes in a precinct with 10 people than it would be in a precinct with 100 people. So I don't quite follow.

yes, as the size increases, the odds of a zero vote are lower. And it could be nothing, but I would generally expect the level of variability to be similar for each of the 4 candidates and it's not. It does give some information about how a vote siphoning effect would work.

 

[drummergirl]

This is what I mean:

As precinct size increases, these peaks should be getting less extreme, not more extreme. When your highest peaks in the large precincts are over 10% higher in the than the highest ones in the small precincts, to me, this is what sticks out more than anything else.

If possible, I would like to know the names of the precincts I marked so that we can see whether they are demographically, any different than any of the others.

+10 rep

 

[drummergirl]

No. My point is that the distribution should be symmetrical, even for small precincts. I'd like to see 4 histograms.

Not sure how many votes are in the smallest precincts, but when you get below 20 or 30 votes in a precinct, it'll be pretty noisy.

 

[drummergirl]

You are right. I suppose I still have noise on my mind because I'm not entirely satisfied with the extra noise in the santorum line in The Man's chart of Santorum Votes Minus Adjusted Delegates. I don't like loose ends.

 

[Bohner]

Evidently he didn't' turn his head when I told him to.

Ok... I see what you mean, still doesn't make any more sense.

So you expect every 12% precinct vote that Paul receives to be cancelled out by a -2% precinct vote so that the graph looks more symmetrical? It's a non-cumulative graph so you are going to see peaks. Obviously you aren't going to see any peaks that stand out going downward because Paul is at 5%, and he can't receive a negative percentage of votes. So symmetrical or not, those peaks are ultimately canceled out... Hence the flat-line.

 

[Bohner]

yes, as the size increases, the odds of a zero vote are lower. And it could be nothing, but I would generally expect the level of variability to be similar for each of the 4 candidates and it's not. It does give some information about how a vote siphoning effect would work.

The closer a candidate is to 0% (or 100%), the less variability you will see. For example: Why would you expect Paul to have a standard deviation of 5% when he only recieved 5% of the vote? On the other hand, a 5% standard deviation for Santorum would be fairly reasonable given the amount of votes he received. So I don't understand why you would expect their level of variability to be similar.

 

[dsw]

This is what I mean:

As precinct size increases, these peaks should be getting less extreme, not more extreme. When your highest peaks in the large precincts are over 10% higher in the than the highest ones in the small precincts, to me, this is what sticks out more than anything else.

If possible, I would like to know the names of the precincts I marked so that we can see whether they are demographically, any different than any of the others.

It looks to me like the top eight precincts for Romney are all from Jefferson county. That's counting only those that have at least 100 total votes because there are some with 100% for Romney and just a handful of votes.

I looked here for the precinct locations:
http://www.evoter.com/al/jefferson-county/page-1/polling-places
And plotted those top precincts for Romney:

Check zillow.com to see where the most expensive homes are in Birmingham. A threshold of a cool million made the big pro-Romney area in Virginia Beach City light up, but in Birmingham you'll have to pick a lower threshold. Or just compare the precincts that pumped out the votes for Romney with the "wealthy neighborhoods" mapped here: http://higley1000.com/archives/29

(What made Virginia Beach City interesting was that the big affluent area that was so extremely pro-Romney was, under the precinct size ordering, all bunched up together on the right-hand side of the graph. The cumulative graph took a big jump when you hit the first of those ritzy Romney precincts, and then those precincts, and another similar cluster nearby, constituted more than half of the votes after the "crime happens here" point. Not that demographics could ever explain anything, of course.)

Romney%, total votes, county, precinct

55.45,422,Jefferson,5310 BAPTIST CH COVENANT
55.62,1548,Jefferson,4806 BROOKWOOD BAP CHR
58.76,1256,Jefferson,4804 FIRE STATION #2
59.56,225,Jefferson,5216 BHM BOTANIC GARDENS
61.97,1220,Jefferson,4609 MT BRK CITY HALL
64.02,931,Jefferson,4502 CHEROKEE BEND SCH
67.44,900,Jefferson,4608 ST. LUKES EPIS CH.
70.30,404,Jefferson,4607 MT BROOK GRAM SCH

 

[Bohner]

It looks to me like the top eight precincts for Romney are all from Jefferson county.

It LOOKS to you??? That's nice.

It's a good thing that out of the hundreds of precincts posted in that graph, that you are able to tell from a cursory glance where those exact precincts are located.

 

[RonRules]

It LOOKS to you??? That's nice.

It's a good thing that out of the hundreds of precincts posted in that graph, that you are able to tell from a cursory glance where those exact precincts are located.

Bohner, I see you're quick to not let yourself trolled by DSW. He's here to trivialize, derail and exhaust member's helpful nature to respond to reasonable questions that people have.

We have the most important fraud case in US elections, ever. We have no time to waste with people like that.

Here's what he was about to do again: (from The Man)
"Rule#4. Use a straw man.
Find or create a seeming element of your opponent's argument which you can easily knock down to make yourself look good and the opponent to look bad. Either make up an issue you may safely imply exists based on your interpretation of the opponent/opponent arguments/situation, or select the weakest aspect of the weakest charges. Amplify their significance and destroy them in a way which appears to debunk all the charges, real and fabricated alike, while actually avoiding discussion of the real issues."

If it has not been done yet, it's really time to contact the Secretary of State in Alabama and each county election clerk for answers. We have plenty of data indicating fraud.

I'm working Wisconsin mostly, who wants to take on this task in Alabama?

 

[dsw]

It LOOKS to you??? That's nice.

It's a good thing that out of the hundreds of precincts posted in that graph, that you are able to tell from a cursory glance where those exact precincts are located.

Cursory glance?

I took liberty's data set, modified an earlier program of mine to analyze it, sorted by Romney's % of the vote, filtered out the tiny precincts, and came up with the list posted at the end. I used the list of precinct locations to map the eight precincts that gave Romney the highest percentage of the vote (of precincts with >= 100 total votes). They turned out to all be within a small area. That area turned out to comprise the wealthiest neighborhoods in that county. All of this can be confirmed by starting from the data in liberty's data set.

You asked if they had anything in common. I tried to answer the question by looking at the data.

 

[affa]

Not that demographics could ever explain anything, of course.)

First off, didn't you out yourself as a troll in another thread? Oh, yea, you did.

Second, nobody is claiming demographics don't affect votes. Of course they do. A religious neighborhood in the deep South will vote differently than a hood in Brooklyn. No doubt.

Acting like we don't think demographics affect voting is just silly. However, the anomaly we're seeing is correlated directly with precinct size, and not demographics.

Not to mention, those spikes have absolutely -nothing- to do with the discussion at hand, other than someone thought they looked odd.

 

[dsw]

First off, didn't you out yourself as a troll in another thread? Oh, yea, you did.

Second, nobody is claiming demographics don't affect votes. Of course they do. A religious neighborhood in the deep South will vote differently than a hood in Brooklyn. No doubt.

Acting like we don't think demographics affect voting is just silly. However, the anomaly we're seeing is correlated directly with precinct size, and not demographics.

When did I out myself as a troll? I was called a troll, when I pointed out that The Man's data set didn't appear to agree with the data set liberty had posted, and the nice straight lines and sharp knee carefully annotated on the graphs disappeared when the same calculations were done with the data liberty had posted. Was that it, or was it something else I said?

The anomaly you're seeing is correlated with precinct size. Contrary to recent claims, precinct size can correlate with various demographic factors. However, if you believe that's impossible and when the graphs don't flatline it proves fraud, then see the graphs I posted (and links to plenty more data) for 1990's bond measures and so on that don't flatline. That one was particularly nice because it lined up so well with urban vs. rural, the thing that's never supposed to happen.

The VBC example is one where the "crime" point corresponds to hitting those large pro-Romney precincts, all clustered together, where the million dollar houses also cluster, and those affluent precincts account for more than half of the votes to the right of the "crime" point. That's demographics and precinct size, the thing that has been proven without a shadow of a doubt cannot possibly ever happen.

Not to mention, those spikes have absolutely -nothing- to do with the discussion at hand, other than someone thought they looked odd.

They looked odd to me too, that's why I looked at the data to see if there was an explanation.

 

[RonRules]

The anomaly you're seeing is correlated with precinct size.

DSW, you're a little slow on the uptake.

The charts flat line with a CUMULATIVE X-Axis.

Except when Romney's around.

For a second time now:
http://lmgtfy.com/?q=cumulative

 

[The Man]

This is what I mean:

As precinct size increases, these peaks should be getting less extreme, not more extreme. When your highest peaks in the large precincts are over 10% higher in the than the highest ones in the small precincts, to me, this is what sticks out more than anything else.

If possible, I would like to know the names of the precincts I marked so that we can see whether they are demographically, any different than any of the others.

Here you go. ALL were in Jefferson County except for Montgomery precinct 101. Honestly, this is simply astounding to me. Included are all 4 candidates' totals in these same precincts so you can gauge the tendency of that particular precinct ov over/ under voting delegates. Note that the first 4 columns are vote totals while the next 4 are the delegate#2 totals. Geeezzz:
note: the third from the right precinct was NOT included here (not purposefully) and, instead, the precinct "BHM Botanic Gardens" was added, which is the Romney spike at vote total 85,489.

The elephant in the room is the number of votes that Romney picks up in relation to his delegates. Now look- there was HUGE voter error happening all over Alabama in regards to delegates- it's a FACT. BUT you look for trends in each precinct as to WHAT the voters were instructed to do- how clear was it to each voter that he/she was to vote for only his/ her candidate's delegate AND for all positions at that. Precincts are manned by human beings whose communication skills/ motives vary widely. When you see a candidate with a 300+ vote delegate/candidate gain in a single precinct(s) while the other candidates' comparisons are reasonable or exhibit losses? I want some of you, including you trolls, to give me a good creative explanation here.

 

[dsw]

DSW, you're a little slow on the uptake.

The charts flat line with a CUMULATIVE X-Axis.

Except when Romney's around.

For a second time now:
http://lmgtfy.com/?q=cumulative

I was responding to affa's comment: "However, the anomaly we're seeing is correlated directly with precinct size, and not demographics." That's why I said, in agreement, "The anomaly you're seeing is correlated with precinct size." Why is that hard to understand?

Except when Romney's around.

And 1996 votes on bond measures, etc. etc.

 

[RonRules]

Here you go. ALL were in Jefferson County except for Montgomery precinct 101. Honestly, this is simply astounding to me.

Actually I don't have a problem with that. Romney did great in 8 rich counties. I am mostly fine with that. I just need to see how well Romney does in similarly SIZED precincts. I would expect him to do better in very rich counties like Jeffeson, probably 30-40% better with no problem.

The problem is the precinct size relationship, not the income demographics. Chart ANY demographics cumulatively, except when Romney (or establishment Republicans like McCain, Bush, Dole) are around and all candidates will flat-line.

I can pull up a chart (that I didn't post yet) that plots the Ron Paul to Romney Ratio in Madison Wisconsin. There's one precinct where Ron Paul does 600% better than Romney. It's in the University area. Again, that's perfectly fine and does not indicate fraud.

Any substantial deviation from a flat-line for a large county, with lots of precincts > 250 votes IS a problem.

 

[The Man]

When did I out myself as a troll? I was called a troll, when I pointed out that The Man's data set didn't appear to agree with the data set liberty had posted, and the nice straight lines and sharp knee carefully annotated on the graphs disappeared when the same calculations were done with the data liberty had posted. Was that it, or was it something else I said?

How do you "pass off" the FACT that in relation to ALL of Romney's delegate curves, his new vote receiving % increases dramatically (even MORE dramatically, not as sharply, when arranged in correct ascending order as evidenced by the "before" and "after" slopes) iN the 300- 400K cumulative total range? As Affa eloquently stated, Romney accomplishes the impossible demographically- outpacing his OWN demographics.

 

[drummergirl]

And 1996 votes on bond measures, etc. etc.

As I continue to poke around (causing my mother to worry over my safety), I'm finding more and more pieces pointing towards the conclusion that this exploit has been used at times since at least the 1980s (The historical data in Louisiana, the plea deal by the LA state election administrator in 1999, various information relayed to me about elections in LA from poll watchers, etc.)

We've been operating on the premise that what we are seeing is new. However, it is actually more likely that the new thing is our "fraud detection" technique.

Now that I've looked at hundreds (are we at thousands yet?) of these charts, I can tell (and I know many of you can too) at a glance most of the time if there is flipping or not. (Once in a while there is a chart with slight slopes and I'm not sure without looking at confidence intervals)

Also, before anyone goes spouting off that all our elections have been tainted since the invention of the central tabulator, I think it needs to be pointed out that the VAST MAJORITY of historical data we've looked at is non-flipping. So you can put the tin foil back in the kitchen drawer. It's more like suddenly being able to review security video for the past 20 years at a local store and finding out there were other incidents of shoplifting.