The IFR for the unboosted in System 2 is 2.4/171.2=.014 whereas the IFR for the boosted is 1.3/72.5=.018.

There is no reason to do any maths here, the numbers are already presented in per capita terms. Fatalities are presented as 2.4/1000 and cases are presented as 171/1000. This is not 171 cases, which would make sense to divide if there were 2.4 deaths; this is cases per population. If the numbers were presented raw, then you would do the division and come up with 0.14. Perhaps the "Cumulative incidence" threw him off so hard he missed the "per 1000 residents" part?

the numbers were not presented raw. This is dunking super hard on a straw man, at best

The unboosted group of System 1 (yellow dashed line) is clearly an outlier; it looks nothing like the unboosted group of System 2 (blue dashed line). The unboosted groups track each other for days 7 to 14, then mysteriously diverge on Day 14. WTF is that about?

"WTF is that about" maybe it's about people who get unpredictably sick. Don't hate-a the data!

The unboosted groups should track each other if these are large, diverse populations. So we have a problem.

If these are large, diverse populations with the same kind of patients in the same circumstances with the same treatments, nutrition, and overall protocols applied when they fall ill ... Which they are not, necessarily, so we have again no problem.

It seems increasingly like Kirsch is a gayop to make pesudo-intellectual antivaxxers seem stupid. He has shit takes too often.