The point being: the optimal level of fraud is never zero. At some point, it's more costly to wring out those last few instances than to just accept that yes, sometimes people steal your pens. Or vote their demented grandmother's absentee ballot.

And the reason we don't worry is that it's not systematic. You should worry about people stealing pallets of office supplies, not individual notepads. And similarly, individuals deciding to vote for grandma doesn't affect outcomes as long as people in both parties do it.

What matters is wholesale fraud: parties racking up tens or hundreds of thousands of nonexistent ballots. And while one hesitates to say "impossible", that's pretty damn close in this age of modern record-keeping.

There may be a reason that the last of the old party machines, with their legendary graveyard votes, died at the dawn of the computer age.

So yeah, I'm sure you'll find six ballots here or twenty there that seem to have been cast for people who didn't request them. That wouldn't change the outcome of any of these races, particularly since you're only looking for Dems who did it (and I guaranteed Republicans did too)

Is it right? No. Shame on you! But that's not why Trump lost. And there is no system that can prevent every single such case. I could probably vote on my mother, my sister, or my Dad's driver's license, if the poll watchers weren't super careful.

And the costs of such a system would be enormous--exactly the kind of mass, scary, this-is-what-they-use-to-take-our-guns database that normally freaks conservatives out.

Come to think of it, guns are a good example here: we could virtually eliminate gun crime by banning guns and going house to house to grab them. We'd save a whole lot of lives that way, prevent a lot of disability & fatherless kids.

But the costs to liberty aren't worth it.

Anyway, to go back a few tweets and resume my earlier point: it's not enough to think maybe there was some idiosyncratic small-scale ballot theft. You need a large, systematic theft, the voting equivalent of "stealing pallets and pallets of pens"

As my colleague @henryolsenEPPC has laid out here, it's not just that we have no evidence that this kind of systematic fraud happened, we have affirmative reason to think that it didn't in, for example, Philadelphia. https://www.washingtonpost.com/opinions/2020/11/10/we-can-be-confident-trump-voter-fraud-claims-are-baloney/

If there was fraud, we'd expect it to be more consistently for Democrats, rather than, for example, INCREASING Trump's 2016 margin in Philadelphia.

We'd expect to see it in cities rather than suburbs.

We'd expect turnout to rise in cities, but not rural areas.

But finally, the whole argument that mail-in balloting allowed rampant fraud doesn't actually make a great deal of sense.

Okay, yes, many ballots went to people who no longer live at that address. But this happened across the state. How does the party get their hands on them?

Does the party have better information on who lives where than the secretary of state? How does the party predict who won't vote? How do they get the ballots--which are in private homes--without ever, once, accidentally asking a Republican to help them out?

Also, if this happened, it will be super easy to prove: get a list of all the absentee voters, and start calling them.

It seems somewhat telling that for all the online speculation about ABSOLUTELY RAMPANT FRAUD, Republicans are trawling with huge rewards for evidence then showing up in court to argue about whether to count a few thousand PA ballots that arrived after the statutory deadline.

Anyway, that's why I can acknowledge that someone, somewhere, probably voted dear departed Grandma's ballot (either the way grandma would have wanted, or in a giant [expletive deleted] to the hideous old bag) without needing to put the FBI on Defcon 1. In case you were wondering.