The number of deaths may be off but that is likely to be dwarfed by the much higher number of total cases that haven't been confirmed. Between this study, the

The number of deaths may be off but that is likely to be dwarfed by the much higher number of total cases that haven't been confirmed. Between this study, the LA study, and the Santa Clara study there could be 10-80x as many cases. That is more significant that the death count being off by a couple percent when it comes to determining the fatality rate.