Excellent question RAYMOND CHIMHANDAMBA. Data models are used to estimate information that we do not have, or that we consider suspect. In this case, excess mortality rates for your country do not match reported reported COVID cases x COVID case fatality ratio. A data model would take into account reasons why this might be (such as disease related deaths, due to an overtaxed healthcare system, and the like).
My question though would be, why would we want to rely more on modelling when we could observe the reality on the ground or try to make that data more reliable. Is that not a better and more beneficial scenario. Models are just that, an intelligent assumption of the real thing