Michigan vs Ohio

I have a huge advantage over most in understanding this NYT dataset because I’ve spent a lot of time with it. It is easy to not realize that others haven’t seen other aspects of this data and the comparison can be enlightening. A couple of comments recently make the point that some of these insane pro-Biden ratios are believable to them. Basically, they make the conclusion that the cause could be just demographics.

Despite how it pains me, Ohio is a similar state to my home state of Michigan. Shared border, same size population, very similar industries, combinations of rural and city life. In Michigan, the secretary of state unilaterally sent applications for ballots to every registered voter. The one lawsuit against this practice failed because the fake judge claimed that these weren’t ballots but applications for ballots, so constitutional election law doesn’t apply and the state didn’t need to follow the law as written by the legislature. Anyone in Michigan in 2020, was able to fill any one of them out and mail them back in and an unobserved government employee would then choose whether to mail out the ballot. Conversely, those who voted in-person in Michigan were required to have identification presented, registration matched, signatures matched and addresses matched. This is done in detail by the poll workers at the time of in-person voting before you are allowed to even touch a ballot.

There are so many ways to cheat this mail in system, but the biggest one is right in front of us. Vote twice as numerous people have openly claimed to do. Send one mail-in ballot and then vote yourself in person. It’s up to the poll workers to decide if they will take the time to compare the addresses and voter names and reject one or the other. However, my wife and I personally received five applications with three of them, sent to people who have moved out of state — in other words, no way to check that we are them other than too many votes at a single address. Something no poll worker is EVER trained to look for as it is not part of election law.

I have a friend who received seven ballots for different people in Michigan. The seven had all moved out of the apartment and are very likely in different areas of the state or country. There is no harm done to nefarious folks who simply sign them sloppily and send them in because there is no proof beyond a reasonable doubt of who did it. The result is that if certain areas wish to be heavily over-represented in Michigan, the poll workers simply needed to accept the paper. These aren’t bad folks because they had no legal mechanism or data by which to reject those ballots.

That’s it — the big Michigan fraud. Not a computer glitch, not some dude hacking in, not some big pre-planned conspiracy outside of changing the rules, just accept the ballots. Very simple. Not that the other issues didn’t also happen because they also did but the big one, the giant no-conspiracy vote bias in Michigan, is right there in your face because there is no observation mechanism to counteract it.

Unfortunately, this bias in voting heavily favors apartment living city voters. Cities have a lot of people moving in and out of rentals, each time they re-register to vote (have a drivers license update), the old voter registration is not automatically updated. That is why in this and future elections but none prior to 2020, seven different ballot applications can show up at a single person’s apartment in Michigan — without any request from a voter.

It is very important to understand that in Michigan, before the Secretary of State unconstitutionally changed the voting absentee rules to automatically mail every registered person an application. Michigan already had unconditional absentee voting by voter request. We simply filled out a request form and a ballot was mailed to us. The only requirement was that we had to previously have registered to vote at some claimed address. Our previous addresses would still be registered and had been confirmed by ID as a valid Michigan voter at least one time in the past. This system could also be cheated without consequence, but not en-masse. This is important to understand because it means that there was no COVID safety NEED for an absentee voting rule change in Michigan.

Not a single prosecution or accusation has been made against a voter for multiple votes to date. Odd eh? I must be wrong……. Nobody voted twice.

Ohio, unlike Michigan, didn’t intentionally bias the ballots. They changed nothing significant in their absentee process as no changes were required for safety.

Enough yacking Id! Show me data!

Figure 1 – Ohio vote totals over ballot counting period. Data from Edison Research per NYT

Note that in the above figure, there are no blue triangles indicating ANY unusual vote percentages in Ohio. All cities and rural areas reported normally with nothing unusual showing up.

Figure 2 – Ohio vote distrbution. Log scale on x

You can see that we have a decent bell curve, probably a log type distribution but I haven’t done any distribution type testing. This point is more relevant when compared to Figure 3 below.

Figure 3 – vote distribution for Michigan.

Now in Ohio, the green and red bars are combined so the way to compare these graph is to imagine all the red bars in Figure 3 are also green. The reason for this difference is that had I split Michigan into two separate voting times per a recent previous post such that we could generally see some of the absentee count separately from the regular count.

Remember the x axis is a natural log scale so vote ratios of Biden/Trump of 3 on the x axis are 7 times more extreme than x values of 1. The likelihood of experiencing those red votes on the far right is extremely low.

Comparing Ohio’s distribution and Michigan’s, there is a distinct second smaller peak in the Michigan vote data. The Michigan graph has a distinct visual bi-modal distribution. You can visually see two separate normal distributions (mountain peaks) on top of each other. The large peak is centered perfectly on Ohio, and a second smaller peak centered to the right.

Interestingly, I did plot a fit to this smaller distribution previously, I just didn’t put it in the previous post. I have added it to Figure 4 below.

Figure 4 – Michigan data again with red distribution fit to after hours votes.

In Figure 4, the red normal distribution is a regressed (and y scaled only) fit to the red bars (votes after 3am EST). Note how well centered the red curve is over the smaller second normal distribution peak in the green bars. The red curve isn’t created using the data from the green bars, only the red but the center of the peak is very close. As most of the red bars (after 3am) represent absentee related vote counting, it is reasonable to assume that the green bars creating the second mountain peak of the smaller distribution are likely absentee vote counting done earlier in the evening. Now if we had access to the raw data, and could show absentee ballots make up this second peak, this would be statistically damning evidence in the state as these two voting groups were treated very differently as a matter of law. This is the type of distribution you would expect to see in a voting distribution if a biased group was allowed to vote multiple times.

So, all that being said. Ohio has large cities. Ohio has vote fraud too which is very slightly visible IMHO in the reporting data. That Ohio distribution (Figure 2) isn’t perfectly symmetric and normal, it has a bias in shape toward the right side of the graph. Ohio did NOT have unconstitutional rule changes to create the situation where voters are disproportionately counted multiple times based on city vs country lifestyle. Ohio also didn’t have any dramatically large unusual voting this time. This state has demographic distributions similar to Michigan but a very different election result.

Michigan’s state bird is the Robyn. A fine upstanding bird that. Perhaps we should change the bird to the dodo, as we are now the fraud capital of the United States.

Michigan does have a little competition in the multiple voting category though:

Figure 5 – Pennsylvania – same X axis scale as Ohio and Michigan above

See any similarities to Ohio or Michigan?