I just look at the data as it is published. A good way to look at infection data is using a logarithm scale. This clearly shows if the rate of increase each day is increasing (very bad), stable (bad), or decreasing (good).
Spoiler alert: it's good news.
"Cases" are number of cases reported in mainland China (M. China) and rest of the world (ROW), from John Hopkins University.
Y-axis is plotted as a logarithm, so straight lines on graph show the exponential rate.
Coloured lines on chart show examples of specified daily increase in number of cases reported. They are not lines of best fit.
So we can see that in the early stages the number of cases increased at about 50% per day, or more than doubled every two days. From about 28 Jan the rate decreased to about 25%. Recently the rate of increase has dropped even further. Conclusion: the rate of increase is no longer exponential but is moving towards a null rate of increase.
The second graph shows the daily rate of increase for each day. There is large scatter in the early days when the number of cases was low. More recently the data points have become less scattered and show a clear trend downwards. Conclusion: the rate of increase in number of cases is going down, so it is moving in the correct direction. This is very good news.
"Daily change" for mainland China (M. China) and rest of the world (ROW), from John Hopkins University.
Coloured lines on chart are linear lines of best fit.
Lines extending beyond data points are extrapolated and therefore not data-based but are speculative.
It's important not to read too much into this data. There could be further outbreaks in countries with poor medical care, for example, or a cluster of outbreaks elsewhere. But the trend is going the right way. By the end of February it is possible that we will start to see an overall decline in the number of cases (ie the trend becomes negative). I am optimistic that this will occur before the end of February but it may take some time to recognise.