What? No. 2000 is the opposite: https://www.cnn.com/2015/10/31/politics/bush-gore-2000-elect...

> The results: The two major conclusions here are that Gore likely would have won a hand recount of the statewide overvotes and undervotes – which he never requested – while Bush likely would have won the hand recount of undervotes ordered by the Florida Supreme Court, although by a smaller margin than the certified 537 vote difference.

Counting over votes would have been utterly ridiculous. It would have election officials try to divine who someone intended to vote for where they punched holes for more than one candidate.

Gore’s whole recount strategy was Bayesian bullshit. Back then, 1-2% of all votes on these punchcard machines were uncountable. Hand recounts will turn up a bunch more votes, creating a poll of votes that can skew them to your advantage through selective recounts that take advantage of different base rates (favoring Bush or Gore) in different precincts. Gore’s original gambit was to request a recount only in counties that had heavily favored him to begin with. Obviously that would result in newly countable votes skewing in his favor at the base rate of the county. But even he never asked for a recount of over votes.

The whole 2000 recount was a shocking parade of innumeracy. The Florida precincts were doing things like including partial recounts in the results. We know from the 2020 election why this would be nutty. Imagine Georgia starting a statewide hand recount, stopping halfway through because they ran out of time, and updating the results with partial data that excluded a bunch of absentee votes, or Fulton county.

I was referring to the full recount[1], and with the larger point being that in 2020 the election didn’t come down to a single state and full hand recounts were actually done. Complaining about a much narrower margin when the single state deciding the election never did a full recount seems far more defensible to me than what we saw this time where the recounts were very similar and the margins in multiple states were at least a full order of magnitude greater, with a backdrop of lying constantly about problems. Hanging chads, over/under-votes, etc. are legitimate concerns – and we can see how many states learned and made them far less common - in a way that fairy tales about mystery ballot drops or dead people voting are not.

1. https://web.archive.org/web/20011217183836/http://www.norc.u...

> Hanging chads, over/under-votes, etc.

These were limitations that existed in the equipment in the time. (The uncountable ballot rate in newer optical scanners is vastly lower.) It only became a “concern” because of Gore’s hand recount strategy. (And burning a ton of time pursing the selective recount, which the Florida Supreme Court put a stop to.)

The fact is that with a 1-2% error rate, you can’t reliably “measure” an election result that’s within a tenth of a percent. The whole attempt to recount spoiled ballots was gamesmanship on Gore’s part. (Permissible under Florida law, which was written by idiots, but gamesmanship nonetheless.)

Again, I’m not saying the hanging chads weren’t a statistical process which was widely recognized, only that they’re a more plausible argument when the entire election comes down to a hundred votes in one state. A recount shifting less than a couple hundred votes is quite possible and at no point did Gore or his party call for disenfranchising huge numbers of people because of it.

Yes, I agree. Given the closeness of the results, Gore only needed to try and manipulate the vote counts in one state.

It's disingenuous to say that Gore would have won if it continued. It was basically a tie to several decimal places -- every recount could swing the balance one way or the other.

I was referring to https://web.archive.org/web/20011217183836/http://www.norc.u.... The key point being that it was a much closer race, with more screwups, and he had a plausible chance of winning the single state needed to flip the result. In contrast, Trump lost multiple states by much larger margins, nothing like the hanging chads or other irregularities, and full recounts didn’t show any statistically significant discrepancies.

> Gore would likely have won had it continued

Isn’t this the exact line the trumpers put out? I hate to be a both sides guy but seriously both sides sound ridiculous when they say stuff like this.

The difference is that it never stopped, confirmed the first results, and had no credible evidence suggesting problems. There was nothing like the hanging chads, and certainly not on the scale needed to flip the results.

Yes. The difference here is that in 2020, the counting did continue, and Trump was so far behind that there was no mathematical trickery that one could do to fix it, especially because he would have needed to overturn elections in several states to get to the required electoral college count.

> we now know, Gore would likely have won had it continued.

This is not correct.

See https://web.archive.org/web/20011217183836/http://www.norc.u...

Now, there is a key difference here as I mentioned: tens to hundreds of votes is a tiny margin, much smaller than anything we saw in 2020 for the presidential race. That’s the larger point: 2000 was a much, much tighter race and there was far more legitimacy to questioning a result which had more confusion in the election process and a margin of victory smaller by an order of magnitude or more.

That's irrelevant here. We have the counterfactual study (i.e. what would have happened if the SC decided the other way) that concluded that Bush would have widened his lead. Stating that Gore likely would have won is factually incorrect.