I see it as putting them in context of a trend, perhaps the forecast doesn’t have to go so far but it makes a difference that the change in trend is also represented on the chart
that’s why it is useful to have the longitudinal data, so it isn’t just looking at a single day number but instead seeing how it fits onto actual and possible trends, isn’ it?
Okay, so to use your analogy - we’re not watching the full game here and we’re not seeing all the goals. This is something like showing you a 30-second clip of the game every 10 minutes with the scores blurred out, and you’ll find out the full results at the end of the game in two weeks.
The trend isn’t calculated accurately for the reasons mentioned above, and the methodology used to collect the test results is also non-representative. That’s the point.
Trying to predict trends is fine when there’s sufficient data to do so, but it isn’t being done properly here, and the set of data is also incomplete.
Try tossing a coin 10 times, and 10 times again, and 10 times again, and record the number of heads each time. It’ll be all over the place. It’s a similar set of data points we’re dealing with here.
But again, one could make the same complaint about announcing the daily totals in the first place. At least this is putting them is some kind of longitudinal context of possibilities and actualities
It isn’t even the goal here. Perhaps the problem is that you’re missing the point.
Just think: Taipei City is so short staffed that mayor Ko has issued a patriotic call to retired medical personnel: your country needs you. 5000 NTD per 8 hour day, to help out in quarantine and testing facilities.
What point do you believe I’m missing? @hansioux is using a flawed method and very incomplete data set to attempt to extrapolate the number of cases into the future and then making claims like that the curve is being flattened.
Absolutely not, those are datapoints (actually aggregation of several datapoints) resulting of actions on the field. You have to interpret them in context which might evolve.
It’d quite funny how on this site, the first thing people do when someone says something they don’t like is look for grammar mistakes instead of an actual response.