Not true. A first/middle/last and DoB match is almost always a dead ringer for a match even if you're dealing with John Smiths.
EDIT: Sorry, to explain further, let's say you have an ordinary day in America. Roughly 10,700 people are born.
Nearly 50/50 male/female, and luckily we haven't moved to androgynous names only yet so most are uniquely one direction or the other. Let's give a little leeway for overlap and say a pool of 6,000 (half plus 10ish%)
So of the roughly 6,000 same-sexed + SNL-Pat-named people born on a given day, what are the odds they have the same first, middle, and last name? Obviously the pool is going to narrow given more unique names.
Even looking at a John Smith or a James Franklin or a Jane Goodman, the odds of there being another John Smith or James Franklin or Jane Goodman in the same rough geographical area (remember, PA and WI areas only here) with the same middle initial (of the 26 letters about 16-18 are commonly used) is very small.
Without getting further into the weeds with the birthday paradox (50% chance of 2 people in a room of 23 having the same birthday) and other metrics, I would be absolutely shocked if the list of 130k above contains any more than 10k that are legitimately different people. I would wager any handful you pick you're probably going to be looking at the same person every time and very rarely finding records that are different people.
It must have, because he is saying over 130k. It's one thing if you get a few hundred hits. That could be dismissed. However, if you're into the thousands with FULL name and DOB being exact, that's NOT a coincidence.
Also, if that matches and they are all registered voters, AND all voted for Biden... that's basically statistically IMPOSSIBLE to happen naturally.
In Oregon, all you need to log in to your account to request absentee ballots is first, middle, and last name, along with DOB. They use that as log in credentials, so it's not that common to have THOUSANDS of people share those same details.
Not necessarily, but of 130k matches I'd put literally all of the money in my bank account on at least 100k being the same people, or positive matches.
I just did a search across 3.6 million PA residents, grouping by first name, mi, last name, and dob and got 30K pairs. They seem to have different addresses. This dataset (unlike ones I used to work with) is supposed to be heavily processed to remove duplicates.
Only .8% but a big number because the denominator is so large. Dunno the cause, could still be bad data I suppose.
Not true. A first/middle/last and DoB match is almost always a dead ringer for a match even if you're dealing with John Smiths.
EDIT: Sorry, to explain further, let's say you have an ordinary day in America. Roughly 10,700 people are born.
Nearly 50/50 male/female, and luckily we haven't moved to androgynous names only yet so most are uniquely one direction or the other. Let's give a little leeway for overlap and say a pool of 6,000 (half plus 10ish%)
So of the roughly 6,000 same-sexed + SNL-Pat-named people born on a given day, what are the odds they have the same first, middle, and last name? Obviously the pool is going to narrow given more unique names.
Even looking at a John Smith or a James Franklin or a Jane Goodman, the odds of there being another John Smith or James Franklin or Jane Goodman in the same rough geographical area (remember, PA and WI areas only here) with the same middle initial (of the 26 letters about 16-18 are commonly used) is very small.
Without getting further into the weeds with the birthday paradox (50% chance of 2 people in a room of 23 having the same birthday) and other metrics, I would be absolutely shocked if the list of 130k above contains any more than 10k that are legitimately different people. I would wager any handful you pick you're probably going to be looking at the same person every time and very rarely finding records that are different people.
Then my datasets must have been garbage.
It must have, because he is saying over 130k. It's one thing if you get a few hundred hits. That could be dismissed. However, if you're into the thousands with FULL name and DOB being exact, that's NOT a coincidence.
Also, if that matches and they are all registered voters, AND all voted for Biden... that's basically statistically IMPOSSIBLE to happen naturally.
In Oregon, all you need to log in to your account to request absentee ballots is first, middle, and last name, along with DOB. They use that as log in credentials, so it's not that common to have THOUSANDS of people share those same details.
Not necessarily, but of 130k matches I'd put literally all of the money in my bank account on at least 100k being the same people, or positive matches.
That's being very, very generous.
I just did a search across 3.6 million PA residents, grouping by first name, mi, last name, and dob and got 30K pairs. They seem to have different addresses. This dataset (unlike ones I used to work with) is supposed to be heavily processed to remove duplicates.
Only .8% but a big number because the denominator is so large. Dunno the cause, could still be bad data I suppose.