Nearly 150,000 attempts to hack SC voter registration system on Election Day: report

Nearly 150,000 attempts to hack SC voter registration system on Election Day: report
Nearly 150,000 attempts to hack SC voter registration system on Election Day: report

In SC 2008 According to the polls, Obama would have 43% McCain would have 53% (+10%). Final results: 45% VS 54% (+9%). 1% difference with the polls.

2012 According to polls, Romney would win with 46% against Obama his 40% (+6%), but he won the election in South Carolina with 44.1% VS 55.6% (10.5%). 4% difference with the polls.

2016 According to the polls Trump would win with 42% VS 38% (+4%) But he won 55% VS 41% (+14%). Polls were 10% off.

It is extremely fishy that the largest attack on voter systems, happens to be the same year that almost none of the the polls came close to predicting the final results.

I'm also curious to know if the registrations were at all tampered with. Seems like a lot of people suddenly found out they weren't registered or had their voting location changed on them. By some accounts Trump won by roughly 4,500 votes in key areas. Some states were hammered with 50k+ registration mishaps.

Trump said he and Russian President Vladimir Putin have discussed ways to combat cybersecurity threats.

Must be some newfangled usage of "combat" that means "facilitate" or "aid."

The electro college is why our election is so unbelievably hackable.

Gore vs bush came down to about 10k votes in 1 state.

Clinton vs trump was 70k votes in 3-4 states.

When you have only about 5-6 states that actually matter ( swing states ) and you only need a 1-3% vote swing to change the election result its hackable.

In fact you don't even need to hack anything to get a voter swing that small. Strategic closing of polling stations, strategic polling hours, fake news, fake scandals, and this is before throwing in hacking voter rolls to change minor things like birthdays or addresses.

Elections of 100 million plus voters are decided by .01% of votes. It's not rocket science to come up with shady ways to change such a small percent both legal and illegal.

Does 150,000 attempts just mean that there was a brute force attack that was trying to guess a password/find a port number? The number is kind of useless without additional information.

Tldr:

There were multiple hack attempts on the SC voter registration systems from late June to early August. No evidence of access / data theft at that time.

This was similar to the pattern in other states, with hacks or attempts from around June onwards and dying off for unknown reasons in early August.

SC hired consultants to fix the vulnerabilities. As of November 8, only one relatively minor vulnerability was left.

Russian hackers made 150,000 attempts on Election Day alone to attack this vulnerability in the registration system.

Two points to make:

This tallies closely with the Steele database's contention that, as at early August, Russia withdrew active support for Trump as it believed he would lose the election.

The uptick in attacks on Election Day was likely a Russian attempt to sow chaos in registration systems, particularly in Clinton supporting areas, which would assist either a Trump win or an argument that Clinton "rigged" the vote.

most polls were actually in the margin of error tho...

not saying there wasn't any foul play, but its actually a misconception that the polls were wrong

http://www.gallup.com/opinion/polling-matters/198155/national-polling-accurately-nails-popular-vo...

This is what gets me. It is widely reported after the event that the Russian attack on the election took the form of three fronts. Attack Clinton herself, spread propaganda and fake news, and hack electoral machines and voter registers.

We know they did the first two fronts, they did very well on these. We know they actually got in to and had access to electoral systems all over the country. Are we expected to believe, after all that effort, they did nothing at all after they got inside.

https://medium.com/@SIIPCampaigns/22-million-eligible-democratic-votes-were-eliminated-from-the-2016-election-was-russia-involved-3afc42eaf31

I was informed that my registration was invalid after I voted via mail about 3 weeks after the election, and that my ballot would not be counted. How in the fuck did they know what ballot was mine in a "secret" vote system? That was the question I posed to my county clerk and she could not answer. Many of the votes thrown out were mail in ballots.

Of course Russia wasn't involved. They're Russians, they wouldn't know which Americans to kick out of the voter rolls. There's no way for them to just randomly find a bunch of high quality targeted advertising data laying around completely unprotected.

And it's a huge indictment of our voting system that a nation of 300 million people has its president chosen by a few thousand people in six states.

yes, but when I went to clarify in january(the first time I could since the election) what was wrong with my registration, they could find nothing wrong with it or my wifes(who's ballot was also discarded). When asked directly then if their election computer systems had been audited for being compromised, she offered no comment, and handed me a complaint form that would likely be round filed when received. our ballots were not the only ones tossed "errantly". This election was very likely rigged to go in trumps favor, and more and more voter suppression and evidence is beginning to support that theory.

I'd rather have it exposed, forcing the courts and lawmakers to address the issues in the transparency and trustworthiness of our voting systems.

I'm not really sure if this counts as evidence of hacking or voter suppression but I live in SC, and my brother who was able to vote in local mandates earlier in the year and was registered to vote found that he had been unregistered on election day. Again not saying it's evidence but with this information it's definitely fishy

See - there was no interference as these were only attacks - no votes were affected, no cause for concern.

I'm also curious to know if the registrations were at all tampered with.

Thing is, if you're gonna steal an election this is where you target. Actual vote counts are likely to be fairly well hardened, because that's the first thing people think of. Only problem is, you need to know who's potentially going to vote Democrat, rather than altering their votes after the fact.

So you find a way to get yourself a bunch of high-quality targeting data that just happened to be sitting around completely unprotected, and fuck with the registrations of likely Democrats in a few key states.

Bam presto, your desired candidate wins by a margin of less than 80,000.

The ballots are usually delivered in a two-ply system. The outer-ply is an envelope that usually has a barcode and which you have to sign for the ballot to be eligible. That gets checked for arrival, and the fact that you casted a ballot is recorded so that your credentials are only used to vote one time. The people processing that cannot see how you voted, they just count that you did and check that all of requirements are met (the equivalent of people working the polling stations).

If your ballot has been accepted, only then is it moved to a second processing area where it is opened and your folded ballot removed. The opportunity for any handling where both your ID and how you voted are exposed is minimized, optimally.

My VERY blue town and the blue towns around it all turned it red this election. We all thought it was weird, but figured it was just the REGULAR kind of rigged like gerrymandering or dead people.

How do we not end up with a recall election after all this shit, assuming it's all formalized as charges by Mueller? I feel like I know about 5% of what's going on despite rabidly following this shit for the past two months and I still find stuff like this article that is appalling months later. And this is the PUBLIC record. Fuck.

I think they just want to be 100% sure before they report it. It would be a pretty ugly thing to get wrong.

I saw this article, but where is the evidence to back up their claim of 22 M voters deregistered?

They will never report it. It would be too destabilizing because there is no way to have a do-over election. Can you imagine the shit show of Americans knowing the election was hacked successfully for Trump and that he still gets to be president?

There's no provisions in the Constitution for a recall election. Of the President, or of any member of Congress. There's only two ways to remove the President, and they both require that 2/3rds of the Senate get their shit together.

This is what's a big piece of crap, and the thing that will turn millions of people off to voting.

Most people don't really have any say in who will be elected.

Dude, the story gets way worse than that. Florida was the only state at the time to hire a private company to purge its voter rolls of felons. A different company had performed the service in 1998, charging $5700 for the job. The company that was given the job in 2000, despite having the highest bid submitted, ended up charging over $4 million.

Florida Secretary of State Katherine Harris, the person in charge of overseeing the state's election and also co-chair of GWB's Florida campaign, requested five times that the analytics company update its list to cast a wider net.

By the time the final list came out, it contained 173,127 names, roughly 1% of Florida's electorate. Only a third of those could actually be confirmed as felons. 88% of the people on the list were black.

This is one of the most blatant cases of election fraud I've ever seen, and most people have no idea it ever happened. Everybody thinks the whole issue was about "hanging chads."

More info

I also highly recommend checking out the documentary Unprecedented: The 2000 Presidential Election.

Edit: Unprecedented is on YouTube, plz ignore shitty repost title.

I really hope there's an implicit "/s" at the end of that. Poe's Law is dead these days.

That's what it looks like to me

He went to the voter section of the county office and waited in line for about two hours to be told that he couldn't re-register and he was not allowed to vote. Most of the people in line were younger voters or minorities

There's no provisions in the Constitution for a recall election. Of the President, or of any member of Congress. There's only two ways to remove the President, and they both require that 2/3rds of the Senate get their shit together.

The people who wrote the Constitution probably assumed that subsequent generations would be smart enough to govern themselves, and would probably find it intolerable to adhere to a 200+ year old document to resolve an unforeseeable systemic crisis.

Of course, they also knew how to handle illegitimate governance.

Let's team up with the people who did these attacks so that we can find out what really went on and stop it happening again, after all, they're the best people for the job!

I've seen multiple people make similar claims in different states... fishy indeed

Yes, I am very concerned by the potential for this type of thing too. But my guess is that your ballot was targeted because of current or past voter registration identifying you as a likely democratic voter, at the envelope-checking stage, not because your ballot was opened and then discarded.

I mean, I really don't think Putin and Trump have ever spoken about this matter directly. Putin probably used their conversation to game out Trump's level of knowledge of the subject and his general demeanor.

That said, at this point, it wouldn't surprise me if a tape came out of them talking explicitly about flipping the election.

Don't advise it but the comments on the Hill are funny.

"The publication said most of the hacking attempts in South Carolina likely came from automated computer bots."

not Russia

OMG people, bots are sentient and they are voting Republican! ...according to Trump Supporters who are down with the cyber.

There's some funky shit happening in the PA election also. Northampton County has a majority of registered Democrats and still went Republican. Like a huge number of Democrats.

90k register Dems 70k registered Republicans. Still went Trump by a few percentage points. It's fucking bizarre.

Edit:

Luzerne County is a better example. 105,487 Registered Dems 68,603 Rep 20,569 Other. 77,508 Trump 51,454 Hilary

Doesn't make sense.

That doesn't sound like "some 400 pound guy in his mom's basement."

With any other President, do you think a report like this- just this single report- would be the biggest news story for a month?

As of November 8, only one relatively minor vulnerability was left.

Computer security doesn't work that way. You can't quantify the unknown. Measuring "attempts to penetrate the firewall" is only direct attacks at the computer security system itself. What about unknown exploits of IIS 7.5? https://www.cvedetails.com/vulnerability-list/vendor_id-26/product_id-3436/version_id-92758/Micro... - what records are there of hacks that removed their own records? Let alone hacks against the application code itself, SQL injections or similar data exploits.

If you consider that almost everyone expected Hillary to win, including Putin, things start to make a lot of sense. Putin wanted Hillary's victory to be tainted. That's why Trump kept saying it would be rigged, that's why the Russians claim they wanted to give Trump Jr. evidence linking Hillary to Russia, Christ, that's why Trump himself said "No puppet, no puppet, you're the puppet". He really thought she was.

I don't think you understand the concept of a margin of error.

It sounds to me like he does.

being in the margin of error means the polls were extremely close to the results,

Not exactly. What it means is the sample size was insufficient to conclude that the difference between the actual result and the poll average result was statistically significant. That's not the same as saying the poll was accurate.

What the person you responded to was saying is that for a given sample size there's an interval around the sample mean where random sampling errors like inadvertently selecting more Rs than Ds or vice versa than is true for the full population can account for the difference between the poll and the actual results. That interval is defined by your confidence level (say 95%) and the sample size.

The import part of that is random sampling errors. If the sampling error is in fact approximately random, which is a reasonable assumption for a situation like this, than there is an equal likelihood that the poll would error above or below the actual result.

What the OP was saying is it's improbable that nearly all polls would error in the same direction, if the error in the polls was simply due to the significance/confidence intervals about the mean that arise from random deviation of the sample mean due to sampling error (which is what the margin of error quantifies for a given confidence level).

Gore v Bush came down to a few hundred votes, 537 to be exact. Also, his brother ordered a review of the voter rolls to 'remove felons' and instructed the analytics company to err on the side of removing those that were questionable despite the company itself discouraging the idea.

Yeah this really could be run of the mill for any system on the internet; more context is needed.

Remember that a lot of voters reported issues with voter roll data, even here on Reddit. Mistakes in addresses, initials, names, DOB's that prevented them from voting. Why hack the vote when you can prevent people from voting in the first place? Everything is coming full circle. All they would've had to do was target Democratic votes in the swing states.

And if they never report it, it will happen again and again.

I voted in the Democratic primary in SC and by the time the election came around all of a sudden there was a problem with my registration and I wasn't registered to vote anymore.

I registered to vote at 18 and voted in almost every local, state, and federal election, all at the same location. The only difference was that I voted in the Democratic primary rather than the Republican primary in 2016. Very fishy.

Ok so in 2016 the county has 198,921 registered voters. 94,915 Dems 68,812 Republicans 35,194 Other. The election results are that 71,384 Trump 65,936 Clinton.

Previous election results from the area indicate that the Democratic vote number lines up well with previous voting, but Republican voting doesn't. Even with a so called populist candidate it doesn't make sense.

You're telling me all Republicans voted that day and a few Independents? In a county that historically goes blue? W Bush lost the county in 2004. Obama won the county both in 2008 and 2012.

There would have had to have been almost 12,000+ vote change between 2012 and 2016 to get that change. It's insane.

We don't see vote swings like that happening in the rest of the state. Most of the other counties where this happened Trump won by 1-2%. Trump won this county by almost 4%. It just doesn't make any sense.

You don't understand.

They BOTH consider Democratic voters cybersecurity threats.

Votes cast may not have been altered but targeting likely Dems and voiding their registration results in the same thing.

No, they didn't. Unless you've got some proof to back up your silly claim.

We need to go back to paper ballots.

Right, but the majority of the country isn't so there majority of votes don't matter

The national polls were closer than usual. They got the popular vote nearly dead on. What the predictions based on the polls didn't do, was accurately model which particular states would shift in which direction and get the electoral college numbers right.

Within individual states, the polls were off in both directions. Which mostly cancels out, but I some states count more than others, so it might be worth checking the ones that were off in Trump's favor.

This is the paper trail they cite:

https://medium.com/siip-campaigns/trump-campaign-allegedly-violated-constitutional-law-to-alter-e...

An attack like that is definitely automated so it doesn't even have to be particularly sophisticated to reach those numbers.

The high number is useless on its own, as a successful attack only needs one attempt and a professional hit would be less likely to leave any trace.

All it really points to is: investigate and follow all leads but don't assume anything. Through all of this there is still the run-of-the-mill criminality that tries every virtual door to find vulnerabilities.

"Our leverage in the election quite candidly goes up as the voting population goes down."

Those types of injustices will hand leverage to the Republicans so long as the electoral college exists. The national popular vote is what should decide the executive branch's leader.

Republicans have been anti-democracy for decades. Republicans have been anti-democracy for decades. They're the opposite of what they claim to be, when they dress up in the flag's colors (the same flag which principles they trample).

The margin of error is the result of the size of the sample. You might, on accident, have more Democrats in your sample, than Republicans. It's very unlikely that all the polls had a sample with less of the exact same voter.

Yeah I tend to agree here. Every single important person who would have been in a position to know if true (former President Obama, former CIA Director Brennan, former FBI Director Comey, former Director of National Intelligence General Clapper, former Homeland Security Secretary Jeh Jonson, NSA Director Rogers) have all said that the Russians interfered, attacked our democracy, meddled in the election with a specific objective. Yet they all say with certainty that the votes weren't altered. Stating a negative definitively is almost an impossibility. They always go right to the edge, but are almost defiant when they're asked if the ballots were effective.

I think it will remain as classified as anything can possibly be and these men will take this shit to their deathbeds. Admitting it would be admitting the Russians broke us. The ramifications would be staggering. It would be so destabilizing that it could very well rip the country apart irreparably. Sad to say, I think the election results were tampered with.

If you have poll A, they use sample A1. You have poll B, thet use sample B1. You have poll C, they use sample C1.

They all have a margin of error, because none of their sample is the every voter-group. It is likely that sample A1 wasn't perfect representable for all the voters and that it counted an higher percentage Hilary voters. How likely this is, can be seen in the margin of error.

It is possible that A1 and B1 both, randomly, missed a few Trump voters. It's somewhat less likely that they missed the same percentage of Trump supporter.

It's less likely that sample A1, B1 and C1 weren't a good representation of all the voters, in favor of the same person.

It's very unlikely that most of the polls had a Hillary-biased sample. A bias for some reason, of the same size.

A result just within a margin of error is possible, a result just within all the margin of error (on the same side), is increasingly unlikely.

Yet they all say with certainty that the votes weren't altered.

The Trump admin hasn't actually checked any voting machines.

The DHS hasn't investigated whether voting machines were hacked in November, and says it doesn't ...

Unfortunately, the provisions for this sort of thing require either action before the electoral college votes, or for the electoral college to do their fucking jobs and not just blindly vote on party lines. Once they vote there's no legal procedure other than removing the president.

The Trump Administration and GOP leadership are actively blocking any commissions from actually auditing anything relating to the election at all.

Nobody will be reporting anything because there's no ongoing audits and the department in charge of doing the audits have no plans on doing them either.

They can't see how you voted but they can predict how you might vote.

Yeah, it's like if a burglar struggled to break a door open with a sledgehammer and you counted each swing as a separate attempt.

To fix the problem, and then undo all this bullshit as well.

It would cause the GOP to fall apart because they did nothing when these issues were questions.

Then we could finally move on as a country.

All of them in California (and not in the states where 80k votes would've gotten Hillary the presidency) I take it?

North Carolina was polled in favor of Clinton, and was won by Trump with a 4.5 percentage point bump.

As far as they know none of the 150,000 attempts were successful - but how certain can we be? They should be releasing the number of people who were turned away for not being registered for the last 4 elections.

Michigan, Wisconsin and Pennsylvania are the other states I really want to hear whether they were subjected to the same number of attempts.

I don't think you understand the concept of a margin of error.

you said "happens to be the same year that almost none of the the polls came close to predicting the final results."

but being in the margin of error means the polls were extremely close to the results, in fact statistically close enough that the deviation from the actual result is accounted for by possible errors in the sample, size, etc...

I'm from Arkansas which is pretty red already. But my area is VERY liberal and has been urbanising for years now. We even legalised weed this year. Everyone was very surprised that the area voted red this year. Idk just seemed weird. Even the pretty religious red groups didn't like Trump that much.

Well it doesn't have to be a direct attempt to help Putin so much as sow distrust in the American democratic process weakening the US in the long run, which helps Russia indirectly. If they "know" Clinton is going to win but can help with the narrative that Clinton is rigging the election by making poorly hidden attempts at hacking heavily Trump favored areas it gives an appearance of Clinton attempting to manipulate the election. While half of America has a brain and would see right through it, Trump's half of America would absolutely attach to the conspiracy and run with it, potentially adding more people who are on the fence about him. As we have seen, there is nothing about Trump or his supporters that is good for America or the world, minus Russia. The fact that Trump won might actually be better for us, in that ONE way (makes me sick to even say that). If Clinton had have won than a lot of this may have never seen the light of day and god only knows what could have happened in 2020 or later.

Yep, I am confident at this point that Trump is an illegitimate president. This is the biggest crime on our country's history.

I'm the only one I've ever heard voice this thought out loud, (or in print in this case) but Kobach had a crap ton of information already as the founder of Crosscheck (Interstate Voter Registration Crosscheck Program). I'm suddenly curious as to how he ended up in the administration, and whether or not he illegally shared the info he collated with a little Russian man.

" Crosscheck supposedly matches first, middle and last name, plus birth date, and provides the last four digits of a Social Security number for additional verification."http://www.rollingstone.com/politics/features/the-gops-stealth-war-against-voters-w435890

This especially infuriated me in Ohio, where Jon Husted made of show of not providing Kobach with any information...because it turns out he's been handing it over since 2006.

List of states that participated. If yours was one of them, shit probably got weird. https://healthofstatedemocracies.org/factors/intercross.html

And what if they were not? What if the polls, all of them, had it spot on?

A shitloads of statisticians, from different parties, different towns even different countries all used their own sample. All used a sample size enough to say something with some certainty.

And all the statisticians, from all the different States, from all the different papers, all had the wrong, in the same year Russia tried to hack the 2016 election?

"Remember we're competing in a Rigged Election, This is a Rigged Election folk" - Donald Trump

Americans are dumber than anyone expected.

Lawsuits need to go to the supreme court on this. Republicans are cool with Russia hacking them to victories at the small price of treason and the end of the American way. So we have to turn to the judiciary. People need to sue in all these states that the hacks were "attempted", challenging the states to prove that their votes were not stolen, or flipped by some Russian machines or that they wont be in the future.

Maybe that is why Trump was so insistent he won the popular vote. Maybe Putin guaranteed him, not just an electoral college, but a popular vote victory.

Edit: Added stuffs

Or, just name the joint unit the Impenetrable Cyber Unit. Problem solved.

Somebody must know. If they stopped attempting to break into the machine at say-- 10am of election day-- that would show that they successfully hacked it. Otherwise they would be attempting until the results were announced, right?

Doesn't it just mean one known minor vulnerability was left?

yes. So any Zero-Day / Unreleased wouldn't be counted. Further, custom apps like this can have unique logic bugs that would be in another category yet.

It's not always about changing the results. Even failed hacking damages voter confidence in the system, and that's very much in line with Russia's goals.

Got proof? Just because your daddy said it doesnt mean it's true.

Sure but 150k times is a lot.

Why didn't the system lock out after the first ten? Why didn't this send out red flags and alerts!?

State polls don't usually have smaller samples and they are predicting outcomes for smaller populations as well.

I am beginning to think the GOP won't investigate this matter further Because they are scared of what they may find out.

We could even mail them in...

More likely they already know and don't want anyone else to find out.

In other words, likely dem voters were deregistered.

Because that's not how these things work. Port scans, for instance, could be considered an attempt but have very little recourse from an ops perspective.

if you fail at collusion (as with ALL of your endeavors/relationships) I guess it's not treason!

If you have 30 polls all saying the same thing, their margin of error becomes smaller. Ofcourse, you can't really translate it this way: phrasing is different, other methods and the possibility that you ask some people twice. But, in essence, the margin of error of different polls should (to some degree) even eachother out.

It does seem pretty bad - do we know how common these sorts of complaints were in previous elections? Just wondering if it's confirmation bias since we started looking for it once the election results were so far off the expected result, or if it's actually something unique to the 2016 election.

At the time, foul play was simply inconceivable. People were scattering to find other causes. All possible, but somewhat unlikely.

The fact is: it isn't that inconceivable that there was some degree of foul play at work.

You know what, a week ago I might try and correct all the false conspiracy theories you jammed into this post.

But today.....who cares. Trump is going down hard.

That's why the question was being asked: was this a password attempt, a port scan, or something else? Saying "they did 150k hacking things" doesn't mean a damn thing.

Most likely outcome is a provisional ballot that was possibly never counted. If you're a polling worker you have to trust the database and the best you can do is a provisional ballot.

though sc wasnt one of them but notice Michigan one of them odd swing states.. said no. Had hilary won Michigan which she lost by 10k votes, she would be president. dems havent lost Michigan since 1980

Ahead of elections, states reject federal help to combat hackers

and

With Russia already meddling in 2016, a ragtag group of obsessive tech experts is warning that st...

Princeton prof shows how it would be easy

A person's Reddit posts would give you a good indication of which way they were going to vote. You could look at which subreddits someone posted in, the content of the comments, what articles they were in response to, and more to build up a picture of a user.

I suspected that something like this would start trickling out eventually, after all these months of calming voices claiming that no voting machines were compromised.

Voting machines were not only compromised; I think that the only reason such opaque systems are used at all is so that the numbers can be fiddled with.

If Americans did it, it's tantamount to treason. If Russians did it, it's a fucking act of war.

I'm also curious as to exactly why Karl Rove freaked out so vividly on Fox on election night in 2012. He was almost behaving like somebody who'd been guaranteed a fix and didn't get it.

Investigate and follow all leads. Anything less is foolish.

The indifference to the security of the election process is something that needs to be addressed. The whole process is way too vulnerable, from foreign and domestic sources equally.

if it was easy to vote in red areas but in blue areas you had to wait in long lines and your registration was lost so you filled in paper ballot that was not counted.

I would call that hacked.

This is all true and we definitely need to pour support for all elections right down to some obscure board position in local council.

But it is ridiculous to suggest the president is actually irrelevant or somehow not dominant in our lives...trump alone fucked over the Paris accord, he stacked a cabinet that is hellbent on upturning our established society. Trump alone can sign into law the madness coming out of the congress - even if Clinton was in the WH with no congressional majorities she could've still Veto'ed their laws and kept gorsuch away from the court.

The president is really, really important- not especially when they are good- but in particular when they are awful.

The narrative will shift.

"Umm, it's actually not illegal for the Russians to disqualify voters via hacking the registrar's database. Since no votes were actually changed, it's not cheating."

There's a comment I saw once about the absurd cognitive dissonance that exists in Trump land around the Democrats (and Hillary in particular).

Hillary Clinton is simultaneously a serial killer with a body count higher than Jack the Ripper, while also dying from every disease known to man. Democrats orchestrated the greatest voter fraud in American history, with 3-5 million illegal votes, but were too inept to place them anywhere but California, the safest D state after the District of Columbia.

This seems totally normal.

All samplings would have an underpresentation of the same group.

And we are living in a world, with a president who colluded with Russia to win the election, a time when Russia almost openly attacks western elections and a time when the most unpopular candidate in ages was able to get 306 electoral college votes. I think nothing is too conspiracy-ey right now.

That's a really shitty brute force attack if it could only get run through 150,000 attempts.