Floating area line chart - Shading the area between lines in a chart without hitting the minimum on the y axis

I am looking to create a graph that uses data lines with shading between them to display something like this:

image

The closest option I found was an “Area line chart” however its lacking in options to customise it to what I am trying to achieve. (Cant attach more images as examples due to being a new user on qs community.)

Is there a way to get around these limitations:

  • The main issue is that the bottom line always shades down to the minimum. Is there a setting that only shades between data lines and not down to the x axis line?
    • Trying to set the color to white for the lowest line to hide the shading doesnt help due to the compounding colors of higher value lines.
  • The colors of higher lines mix with lower lines which makes setting a consistent color scheme very difficult and the use of a Legend completely useless. Is there a way to make the colors appear opaque as they do on a “Stacked area line chart”?
  • There doesnt seem to be a way to display lines on the area line chart without them contributing to the shading. I would like a way to show a mean line (as per the first image) without it affecting the shading.

(Note about the “Stacked area line chart”) I also found a post with a similar problem to myself

The proposed solution was to use a stacked area line chart and shade the bottom in white.

The problem for my case is that there still doesnt seem to be a way to display the mean line without stacking with the other lines.
Also (more importantly) I would really prefer to retain the ability to hover over the graph and display accurate data points for each date. Doing so on a stacked area line chart would display incorrect numbers if I were to compensate the values to make the shading proportions look accurate.

1 Like

Hi @ilya
as you said it is a bit tricky


But i guess there is no simple solution right now.
BR

Hi Erik,
How did you create the chart shown above? I am struggling with the same problem and was wondering if I could adopt the same approach.

Best,
Magnus

please have a look at Mean_Analysis

Thanks very much for this, Erik.