Forum Discussion

JohnSkyrim's avatar
JohnSkyrim
Occasional Reader
Apr 16, 2026

Inserting Graph Breaks/ Axis Breaks

I am trying to create a lovely little graph. However, there is a space between two of my points that I don’t want. Is there a way for the x-axis to “skip over” that space, similar to a graph break or axis break? Thanks!

1 Reply

  • m_tarler's avatar
    m_tarler
    Silver Contributor

    A graph or axis break doesn't exist.  As for workarounds, it is unclear which graph you are using and what you mean by "a space" between points.  If you have a literal gap between line segments then presumably you have an X value with no Y value and if that is the case then you could use a helper column to filter those blanks out.  If you are using a Scatter plot with data in one range and then data in another range which may be 1 data set or broken up as 2 different data sets then you could:
    a) probably best and easiest is to create 2 graphs and put them adjacent.  you can even play with formatting to make them look like 1 with a break.  Here is an example I worked up for my guess at the sort of situation you mean.  I have a sin wave from 1-100 and then again from 800-900.  the upper graph is just a straight plot, the lower is 2 plots side by side:


    b) you might be able to use a helper column to force the values / graph you want using interpolation and then use a line graph with predefined 'category' labels.  Although possible it would be a bit of work and likely result in the 'data points' not being the actual 'data points' since you may have to use interpolation and would only work in cases where interpolation between points is possible (i.e. in the above the line between x labels of 0 and 20 is not a single interpreted line).  This would be best for cases where you are taking samples/values at regular intervals with a gap.  Here is a very simple case the upper plot is a scatter plot of the data (on the left) and then the lower is a Line Graph of the data on the right.  In this case the data are the same but in theory you could transform the original data as long as it can be represented using discreet x points: