Currently Online 
drkohler said:
Not isn't. It has equidistant horizontal lines between orders of magnitudes and the numbers don't fit. That is not logarithmic. |
The y-scale of that graph starts at 10 -> 1.0.
Canada sits at 2.85 on the Y-axis. 10^2.85 = 708
The actual value was 713, log(713) = 2.853
I can change the background to make it more clear. I find the 'squeezing lines together' graphs harder to read so I just put 2 and 5 (at .3 and .7) to illustrate the plot is log ( y ) and have an easier way to estimate the actual value using 10^y.
Anyway @SpokenTruth has the numbers covered. I'm trying to compare countries by growth rates and stability of reporting while putting less focus on actual numbers. There are advantages and disadvantages to any graph. Week over week comparison varies wildly at low reported cases and any outliers will distort the graph twice. 7-day running average is more stable than the 3-day average I use, yet is also more unresponsive to see growth trends and 7-day averages are messed up for a week with outliers.
And then you have countries like France where it becomes more of an artistic impression to figure out trends lol.
