The only corona predictor you need
We’ve seen many projections about the potential spread of coronavirus in different countries around Africa over the last couple of weeks, some that look convincing, others that seem patently mad. I’m not going to link to any of them because I’m not qualified enough to know which side of the sanity line they are.
Gabriel Goh‘s Epidemic Calculator, however, is awesome. It’s a bit initially overwhelming with all its sliders and variables, but all you really need to know is that it’s preset with some general numbers that reflect current knowledge about Covid-19 (what percentage of the infected need hospitalisation, how many die, how long people are infectious for) which you can update as new information comes up. The key slider to play around with is the big one marked R0. This is the number of people one carrier infects on average, and what Goh’s tool does is show that just reducing this number a tiny bit, from say, 2.3 to 2, can reduce the peak number of hospitalisations from 200 000 to just 12 000.
It is, obviously, an imperfect model (a low R0 depends on social distancing going on forever, and new data on mortality and hospitalisation rates is coming out every day). But it is a great showcase for how a little effort can make a big difference.