Hmm, try looking at <excess deaths> as a computation that takes the delta with a computation based on a control period. (unpacking: It actually measures the de

Hmm, try looking at <excess deaths> as a computation that takes the delta with a computation based on a control period. (unpacking: It actually measures the delta with a predicted trend. The trend calculation projects what current deaths should be, based on the situation in the past - which you can see as the control period. It's not perfect, but this should, reasonably, exclude most other factors fairly well. )