Via Althouse, the AP has done an encompassing investigation and found that the provably fraudulent votes were not numerous enough to change the election results in any state. The AP has a leftish bias - we will return to that presently - but I suspect they are approximately correct. I think even if this was slanted that people recognise there is a limit to what you can overlook and pretend is fine. Yet there are two issues here and always have been, and I believe conflating the two causes us to see the issue wrongly.
If there are mistakes, whether by incompetence or design, then those are a problem for elections in and of themselves. We should want to get things as clean as possible, and perhaps just as important, not have any appearance of chicanery or incompetence. People need to believe it's worth voting. So whatever the final total, there were clear appearances of error, and identified places where error or cheating could have occurred. Were these exaggerated by people who were certain that there must be widespread fraud who believed every rumor? Sure. But there were things that even I shook my head at, thinking "this has to at least look better for the sake of our mutual sanity."
Whether Trump was robbed is a separate issue from whether things were done correctly. He certainly conflated them, as did many of his supporters. And opponents. Look at the AP article, which also focuses on the "does Trump look bad" aspect rather than dividing the two issues. That may be sexier, but it's not good reporting. (I don't want to make this messier than it has to be, but the issue above could itself legitimately be split into two, reality and appearance, but I don't think that is necessary here.)
Voter ID is an excellent example here. Apparently Voter ID laws don't favor one party over the other in the total, even though conservatives are certain those would help them and liberals are convinced their voters will be cast out. So it may be mostly symbolic to insist on ID, but it is a symbolism I strongly approve of, whether it helps my side or not. It gives some reassurance that everyone is trying to do their best here.
For the vote totals there is the issue everyone seems to be most focused on,which is counting up the votes and seeing if Trump had the election stolen from him, was blowing smoke, or some of both. The final count is all they see. I have problems with that. There are Type I and Type II errors - that is, denying someone their legitimate vote and giving someone an illegitimate vote are equally bad. Whenever we are filtering voters, we want to reduce both types of error. There are also no guarantees that votes that smell bad all went one way. Especially when it's just incompetence, who knows who will benefit? Also, what people believe - even very smart people who get paid lots of money to assist campaign strategy believe about which side a policy favors - may not be accurate. You could make good cases that both Republicans and Democrats have been adopting tactics they thought would work that have blown up in their faces, so why not voting as well? We take educated guesses that suspicious activity in highly Democratic districts likely favor Democrats, yet even at that, do they favor them 55-45 or 90-10? If 10,000 suspect votes are at issue, it doesn't mean "those 10,000 votes could have swayed the election" if the margin of victory was 8,000. In fact it means it almost certainly didn't. Just because it might technically be enough votes doesn't mean there is any reality in which that actually happens. Incompetence is messy and cheating doesn't work as well as the cheaters hope.
I hoped that at least one state was shown to have been stolen (as Kerry stole Wisconsin) just to get everyone's attention to take election cleanness more seriously. I think we got some improvement this time, but I'm not sure how much.
Moral: Everyone should live in NH or VT.