Y-axis representation bias

[mo-nonym]

Hello. I want to hear if somebody finds the y-axis scale of the graphs also slightly misleading.

Let's take for example the graph for the "Last 7 days". The stat numbers are clear comparisons with the metrics from the previous reference period; days n-8 to n-14. In case of fewer visitors, they show a comparison in numbers like Visitors 53 (-64).

For the graph view, there's no information about the previous reference period. What I find misleading, is that the scale of the y-axis is calculated only to fit best the current period. We see visitor numbers only for the current "Last 7 days" and the bars are always "tall"; even tough there were fewer visits. It gives somehow a wrong impression from the graph.

If the y scale for the current period was at the (theoretic) scale from the previous reference period, on could grasp that the visitors numbers are somehow less, because the bars would appear small compared to the y size of the graph.

Does this make any sense ?