In 1984, there was a famous experiment by Cleveland and McGill to work out which charts were best for people to perceive quantitative data accurately.  We'll do an updated 2019 version of this experiment.  I will show lots of different Power BI visualisations of the same dataset; the audience will vote for the viz that they think is "best" - most insightful, attractive or clear -  and then explain their choice.  Hopefully we'll come to some conclusion about choosing the right chart for the data.  We'll look at the usual chart types: line; bar; scatter and their variations: stacked, clustered, dual-axis, but also some more exotic ones; Sankey, chords, Mekko, bumps and slopes.

The video is not available to view online.