I haven't looked at the source website you sourced this data from, but I assume all the numbers are correct (although, there's a few rounding errors in the provided tables, such as the year 2011 having a deaths per capita of 9.56 using the numbers in your table, which would round up to a value of 9.6).
Is there a particular reason to use an average figure for the previous 10 years and not, for example, the previous 20 years, or 5 years? This is because the size of the year 'grouping' can shift the average figure for the deaths per capita. If one were to take the average of the previous 5 years to 2020 (from 2015 to 2019 inclusive) then that would mean the average deaths per capita would be 9.02 (using the values in the last column of that table) with an associated standard deviation of 0.25, and would therefore imply that the increase in deaths in 2020 would be 4.77% above the preceding 5 year average.
I think that, given the year 'grouping' of 10 years is probably arbitrary, and the 2020 record is only 1 year's worth of data, I would think that it would be better to compare the 2020 data against yearly data, rather than data spanning multiple years. (If Covid were to last say, a decade, then it probably would be good to compare that Covid 'decade' with the previous decade)
To better visualise the yearly trend (done via Excel) with the data in your table, the graph below shows that from 2010 to 2019, the deaths per capita has actually shown a decreasing trend, with the deaths per capita in 2020 presenting a large jump compared to the previous year.
If you do the percentage change in deaths per capita per year, you would come up with a graph looking something like this:
Which shows a roughly 9.4% increase on deaths compared to 2019 (using the values provided in your table).
Also, if we were to go back to the decade grouping argument and statistically analyse the years 2010 to 2019, the figure of 8.62 wouldn't be classified as an outlier as the number of 8.6 wouldn't fall outside 1.5*IQR (interquartile range) as shown below, and for reference:
https://en.wikipedia.org/wiki/Box_plot
Whilst the decrease in deaths per capita in 2019 compared to 2018 is visually distinct (4.6% reduction), if compared to the 2010 to 2019 group it does follow the trend of decreasing deaths per capita over time. From the graphs I would probably think that there's further questions:
1. What has caused the decreasing trend in deaths per capita over the past decade?
2. What caused the larger than normal reduction in deaths per capita in 2019 (I'm not sure what you mean by your statement:
"meaningful "pent-up" mortality")?
3. What caused the abnormally large increase in deaths per capita in 2020 that appears to break the trend in questions 1 and 2?
In response to your Finland post, if you were to graph that, it actually demonstrates the opposite to Sweden, with an increasing trend in deaths per capita over time:
I find the Finland data quite interesting, given the fact that it's virtually opposite to the Sweden one.
