How To Tell If We're Beating COVID-19

COVID-19 reporting often feels like a moving target because case counts change so fast that today’s numbers can be misleading tomorrow. The central insight here is that the key question isn’t just “How many cases are there?” but “Is exponential growth still happening—and has it started to end?” Exponential growth is dangerous precisely because it can’t be easily “eyeballed” once you’re mid-spread; the epidemic can look like it’s still accelerating even when the growth rate is already slowing. That uncertainty matters because the point at which exponential growth stops largely determines how many additional people will become ill.

To make that turning point easier to see, the analysis introduces a global visualization built from real country-level data. Instead of plotting cases against time, it plots the number of new cases (the growth rate) against the cumulative number of cases, with both axes on logarithmic scales. Under exponential growth, new cases are proportional to existing cases, which makes the relationship appear as a straight line. Countries that keep following that line are still in the exponential “rocket ship” phase. Countries that peel away and drop off the main pattern have started to beat back the spread—an “emergency eject button” moment that can be detected even when a country’s current case totals are low.

The method also relies on two practical adjustments. First, it uses logarithmic scaling so that changes by factors of 10 (rather than fixed increments) remain comparable across countries with vastly different case counts. Second, it emphasizes catching changes early by focusing on growth itself—such as the number of new cases over the last week—because slowing growth is much harder to spot when looking only at total cases. When weekly new cases flatten or decline, the epidemic is leaving the exponential zone.

The visualization is designed to highlight deviations from exponential growth, making it easier to answer whether public health measures—testing, isolation, physical distancing, and contact tracing—are actually working in the data. It also reinforces a sobering expectation: countries that haven’t yet had high case counts may still follow the same exponential trajectory later unless interventions interrupt transmission.

Still, the approach comes with caveats. Log scales can visually distort differences, and the choice of axes can make rebounds harder to see after downturns. The graph shows detected cases rather than true infections, and rising testing can make case growth look faster than it would under stable testing. Data quality varies by country and is delayed because growth rates are averaged over recent days to reduce noise. The result is intentionally conservative: it aims to avoid overreacting to short-term fluctuations, so downward trends are more likely to reflect real progress.

Overall, the takeaway is that tracking the rate of change—new cases relative to existing cases—offers a clearer signal about whether an epidemic is still accelerating or finally starting to slow, which is what planning and public understanding depend on most.