Quick Reminder: Don't Compare the Final 2020 Popular Vote Totals with Non-Final 2024 Vote Totals

My quick glance at the posted vote totals suggests that there are about 16 million votes not yet counted.

Right now, Harris is at about 67M (N.Y. Times data), but about 45% of California votes aren't yet included in that tally, plus 30% of Oregon votes, 35% of Washington votes, 35% of Arizona votes, and some more in other states. Given that so far the 54% of California votes that have been counted are split 5.6M for Harris to 3.9M for Trump, that suggests the remaining 46% will add roughly 4.8M to Harris's tally and roughly 3.3M to Trump's.

I went through the N.Y. Times map and entered the data from the states which weren't listed as >95% reporting, and projecting from current totals, it seems like Harris is likely to get probably about 9M more, for a total of about 76M. Trump's popular vote will likewise grow considerably beyond his current 72M, to about 79M. That expected split (76M to 79M) might be compared to the 2020 final results, which favored Biden by 81M to 74M; it thus appears that we will ultimately see a likely swing of about 5M votes, give or take a million or so, I'd guess.

I mention this because I've been seeing people suggesting that the decline from Biden's 81M to Harris's 67M is highly suspicious. But, again, that doesn't make much sense, because that's comparing final 2020 totals to far-from-final 2024 totals.

Of course, I'd love to see more reliable estimates than my quite rough calculations. UPDATE: The University of Florida Election Lab estimates that 158.5M ballots were counted, which suggests that there are 19.5M ballots left to count; that's higher than my estimate of 16M left over, but still consistent with the broad point that the current totals are far from final.