CMR I take it means Case Mortality Ratio. That’s usually taken as the ratio of confirmed cases to mortality. I’ve never seen it anything like that low anywhere there’s been an outbreak. So looking at California’s figures right now, 1,438 total deaths for 37,787 confirmed cases is about 3.8% which is more usual.
The infection fatality ratio IFR is all infections, detected and undetected, versus total deaths, and is therefore lower than CFR. IFR from a simulation looks like 0.1% as an outbreak is spreading unrestricted, in that by the time there are x deaths, the model predicts that there are about 1000x infections in total, detected and undetected. But this is misleading through being a static snapshot of a developing situation, because some of those 1000x will die over the coming weeks, by which time the number of infections will have increased. Neither CFR nor IFR can be properly calculated until infections come to equilibrium, and that’s a long way off.