Big D vs Twitter
- Thread starter Mors
- Start date
- Latest activity Latest activity:
Of course. The fact he is an academic leaves a slight nagging doubt that he's a fucking moron and my instinct is that I should wish death upon him. Aside from that though.
Yeah this guy is great. I watched his video explaining Benford's law and why it shouldn't and can't be used to analyse election data.
The most important part is that the original video you posted is complete garbage, and they constructed analysis to prove their own biased view. It was a waste of time.
Data that doesn't have a wide spread in the numbers of digits. You can use it for types of election data where the spread is wide enough, but you need a court order to get he raw data first.
His first video got to the point quickly. This video seems like I'll have to watch the whole thing and pay attention, skimming through I don't see what he's saying. Was his only issue that the same pattern appears in Biden's numbers that it does Trump's? Or is there some other math thing he mentions somewhere?
Why? I skimmed through the garbage video, got to the main graph, and provided the timestamp of it so as not to waste yours or my time. I can't find the main point in this guys video using the same skimming. If you know it it'd save me 20 minutes
Basically that they ignored the disparity between data sets (in size) and therefore ended up using a formula that inverted the graph (for Biden). Or in other words they either didn't do rigorous testing of their theory before publication or (my words not his) set out to deliberately mislead.
I think it's the later simply because since any mathematician worth his salt would have done the testing and thrown it out for peer review before making spurious claims.
So you now agree that Benford's Law can't be used to find manipulation or fraud in election data?
Try from 11.30
So on the Shiva graph up the y-axis he has the % of votes lost from straight party voters. Instead of that being a percentage of the voters, he subtracted one percentage from another and gave the answer as the percentage. I can't even begin to understand how fucking dumb you need to be to do this.
No, I agree it can't be used on data that is all of the same order of magnitude.
Yep as I said above. Two data sets were used, of differing sizes and from two towns. Matt Parker showed the data they used earlier in the video.
Actually in that Matt Parker video we referred to earlier he clearly states it can't be used for elections ... period. It was designed for accountants to find discrepancies in accounts for further investigation (though again it's not indicative per se of fraud).
Then he's wrong or he was specifically referring to the count data, not the masses of other data gathered during an election.
Not at all. He states categorically : Benford's Law is problematic AT BEST when used to detect election fraud. Literally in the first 2 mins. He later explains exactly why it shouldn't be used in relation to election data.
His explanation is that there isn't a wide range of numbers in election data, so that constrains the digits that can arise, which negates the law. His definition of election data that he has in mind is the votes, in which case it is true. My definition of election data is all the data.
The way I'd explain the Benfords law video, is that it shows with that first data set, that it is manipulated data, which is correct. But it's manipulated as each voting precinct size had been adjusted to be of similar size, as that's what we want to do with voting areas, not because there was fraud.
Dantes is wondering about the other data. Like how many hotdogs they ate during the count at Wade County, and at what intervals. Polling data, and why it was so suspiciously correct. Why the turnout was higher than ever and whether this could be linked to a majority of Americans thinking maybe this time I should vote, otherwise we're halfway to fascism. That kinda thing.
I like the way you're thinking about the data. Don't like the way you're assuming the motivations of the people who determined the precinct boundaries were anything other than how will this keep me onboard the gravy train.
Fake news. I simply don't have a youtube channel which compensates me for the time taken to get hold of the source data and recheck someones work for a mistake so unbelievably stupid it would not even cross the mind of a gcse student.
I think you should probably just assume all the news about voter fraud is at best, immaterial, and at worst, totally fake and damaging, at this stage.
There's way too many graphs left to go before it comes anywhere even close that point. The democrats are also spending way too much of their money to join lawsuits in order to block the data which can be used to plot those graphs from being handed over, for the strange reason that everyone is so busy counting or checking votes, that you simply can't have them spare the minute it takes to stick a pen drive into a computer, copy paste some files, put it in an envelope, and post it to the president.
From what I understand it was the Republicans that changed many of the boundaries after the 2016 election so you may have a point. Biden should have won by a lot more.
He'd only destroy it, assuming it came from Russia and would therefore incriminate him ... again.
Lol at the 100+ hours Dantes has put into this thread
I think you missed the point(although you're talking about something else valid. It's not the actual boundaries and who's in each one, for the pure maths discussion in this case. It's that the boundaries are in anyway judged be population. If boundaries were purely done on geographical size i.e. every 10 km2, then Benfords law would apply, as population sizes would be randomly distributed amongst them. But they're not, they're designed to have roughly the same size of population.
Similar threads
