The implication is they would have died soon, like in the next 6 months.

About half of Covid deaths are in nursing homes. The median lifespan after entering a nursing home is 5 months, and the mean is 13 months.

Coronavirus is very infectious, so if it enters a nursing home, it could easily spread and wipe out everyone who was on the brink of dearth, or on a fragile state.

According to The Economist, the average lifetime lost per covid death is over 10 years.

https://www.economist.com/graphic-detail/2020/05/02/would-mo...

Well, it's worth keeping in mind that the risk of death from a coronavirus infection is comparable to the annual risk of death from all causes. Coronavirus infection though is a once per lifetime risk, whereas all-cause mortality occurs every year.

Just as a random comparable, the annual risk of getting cancer if you're 35 years old is around 0.1%, and the annual risk of having a heart attack is around 0.5% (higher for men, lower for women). All of those diseases would have similar average lifetime lost as covid, but the risk is continuous, not a once per lifetime occurance.

Sounds like they took the life expectancy of certain age groups without taking into account individual preexisting conditions whatsoever.

Not according to the article: "Then the authors accounted for other illnesses the victims had, in case they were unusually frail for their age."

Presumably he means they would've happened in the same time period otherwise it's a meaningless statement. On a long enough timescale everyone who died of Covid would've died anyway.

Whether it's meaningless depends on the timescale. For zero months its very meaningful while for infinity months it's completely meaningless. And in-between it's meaningful, e.g. they would have died in the next 12 months.