A friend posted this meme on social media recently. I'm posting my answer here, unedited, because--well, I'm feeling lazy; and it's a neat illustration of how overly simplified arguments don't really answer much of anything.
I don't know that this is quite so ... encouraging? or discouraging! as it first appears.
I think the numbers are coming from here: https://www.prb.org/usdata/indicator/deaths/chart
The final number, deaths for 2020, I'm less sure of. CDC keeps a count here: https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm. As of today, November 23, they have 2,551,852 deaths from all causes.
If you do the math, use either of the November figures to calculate total deaths for 2020, you'll come to something similar to previous years: 2,837,145 if you use the number and date from the meme; 2,848,397 if you use the number from the CDC's website today.
Here are the two questions though that throw all that up in the air. First, there's an ongoing upward trend in deaths as the U.S. population ages. Given that trend, and how old the population was at the beginning of 2020 (compared to previous years), how much should we have expected the death rate to increase in 2020?
Second, the unfortunate fact of the matter is that deaths do not get tabulated right away. There's a lag time between when a death occurs, when it's recorded, and when it hits the tables for a big organization like PRB or the CDC. So the number of deaths suggested by projecting from the meme, or the CDC's current numbers, is definitely too low, perhaps a lot too low.
(That's why elsewhere, the CDC does not publish death counts for recent months--because they know they don't have all the data in yet. E.g., here are counts for 2019 and the first six months of 2020, suggesting that they're not sure how many people died in July-October: https://www.cdc.gov/nchs/nvss/vsrr/provisional-tables.htm.)
Tl:dr, we'd need more data than I can find publicly available online to ACTUALLY know if we're seeing more people die than normal, or about the same, or less.
That's what I wrote. But here's a thought to close with. We'll be able to tell, eventually, when all sorts of other numbers come in, whether all this was worth it, or not. We might even be able to project right now whether it is worth it. Certainly if we end up with death totals that are similar to what we could have anticipated without COVID, there will be some questions about those "excess COVID deaths" that the media likes to talk about. Probably the reason for a similar death rate to normal--if it does end up similar--will be that we're seeing fewer deaths from regular flus, etc., most likely due to people avoiding each other more.
But all of that is moot to the larger point, which is this. Whether you're a COVID believer, or a COVID denier, you can't simply pick up numbers that look like they make your point. You have to think about the context of the numbers, and consider what other relevant numbers are needed to complete the picture of the truth.
That said, I should note that the totals for the first six months of 2020 (see last link above) are not super encouraging. They suggest about 200,000 more people died in the first half of the year than died in the previous year. Wildly extrapolating, let's say 400,000 more people die this year than in 2019. That would mean this:
That's a pretty sharp uptick compared to previous years, and I have trouble imagining that the U.S. population aged *that* much--suggesting that indeed, the extra deaths are due, directly or indirectly, to COVID and the fears and restrictions it entails. (Of course, we may then see a corresponding downturn in deaths in 2021, since there will be fewer elderly people and people with preconditions alive, proportionally, than before.)
Will it actually turn out this badly, when all the numbers are in in another eight or nine months? I don't know. How bad would it have been if we had done nothing? I doubt we'll ever know. How much better could we have done if we had done things "perfectly", whatever that means? There's no way of knowing that.
Tl;dr (again), and with apologies to Mark Twain: "It's easier to convince people with a little evidence that they're already right than to get them to consider all the evidence, especially when it doesn't give them certainty."