The rate of infection largely depends on the initial rate of growth within the country, which in turn depends mainly on cultural and environmental factors. Thi

The rate of infection largely depends on the initial rate of growth within the country, which in turn depends mainly on cultural and environmental factors. This spreadsheet https://docs.google.com/spreadsheets/d/1AeG07Bt1fOA-msWYb9jWAl_bdo2ThBea7t2NzjFp2to/edit#gid=0 https://docs.google.com/spreadsheets/d/1AeG07Bt1fOA-msWYb9jW... compares two countries, each starting at an initial observed death rate of 1.0 / million. (The rate of growth here is (1 + Rt / serialInterval) or about the average daily spread to others) The first country has a higher daily growth rate and locks down immediately, starting a process that will have its final effect on the deaths per million in 18 days. The second country delays lockdown by 30 days and needs to do far fewer changes to reduce the growth rate to 1.01, and yet it doesn't surpass the deaths per million of the first. You can get an approximation of (1 + Rt / serialInterval) but delayed by about 18 days if you look at the ratio of total deaths between two days (will need to be smoothed due to reporting variations). If you want a more realistic model you can get the same thing by tweaking R0 and intervention strengths/delays at https://covid19-scenarios.org/ https://covid19-scenarios.org/ This is a simplification (since it assumes that the entire population is a spreader rather than those infected 3-6 days before) but the conclusion doesn't change when you make the model more complex, and it explains why different countries seem to be doing so differently. The initial rate of growth completely dominates any other factors. This image generated with covid-scenarios: https://imgur.com/a/BRpkb0I https://imgur.com/a/BRpkb0I describes the variation in infected depending on R0 given everything else is the same with 2.5 being the average, 2.0 the lower bound and 3.0 being the upper bound.