BMI 625 Final Project

Jelle’s Marble Runs and Marbula One

Marble racing on a track, with commentary
In the style of Formula 1 motorsport
Two-day competition:
- Day 1: qualifying with marbles one at a time to determine order for Day 2
- Day 2: the race (or “grand prix”)

Demo

Jelle’s Marble Runs received sponsorship from Last Week Tonight with John Oliver during the Covid-19 pandemic.

The Data

Sources

Marbula One Open Data Repository for race track data
Dr. Randal Olson’s personal blog for season one results

First six rows of the merged data set
Date	Race ID	Track	Turns	Marble Name	Team Name	Lap Time (ms)
15-Feb-20	S1Q1	Savage Speedway	13	Clementin	O'rangers	28.11
15-Feb-20	S1Q1	Savage Speedway	13	Starry	Team Galactic	28.37
15-Feb-20	S1Q1	Savage Speedway	13	Momo	Team Momo	28.40
15-Feb-20	S1Q1	Savage Speedway	13	Yellow	Mellow Yellow	28.70
15-Feb-20	S1Q1	Savage Speedway	13	Snowy	Snowballs	28.71
15-Feb-20	S1Q1	Savage Speedway	13	Razzy	Raspberry Racers	28.72

Audience

Anyone interested in marble racing or Jelle’s YouTube channel
- i.e. kids and amused adults
Some curious physicists?
Fans of Formula 1 who need some kind of an adrenaline fix

Graph Type

Raincloud plot - hybrid of box plots, violin plots, and jittered points/dot plots/scatter plots
- Combination resembles a raincloud and gives the chart its name
Enhances traditional box plots through multiple modalities:
- Conveys distribution shape
- Measures of central tendency
- Preserves raw data information
Considered a transparent visualization

Representation Description

Start with the question: what is the relationship between the number of turns on a track and a marble’s lap time?
This question evolved to: how is the distribution of lap times related to the number of turns in a track?
Two ways to do this:
1. Use qualification laps for each marble because there is no interference from other marbles.
2. Use race lap time because more turns might cause more collisions with other marbles, and thus be more realistic.
For this visualization I used the qualification lap data.

Methods and Composition

Start with a scatter plot of number of turns vs. lap time.

A distribution may be more interesting than a line of points. Also, some of the race tracks have the same number of turns. I will encode that information with boxplots and color.

Want to also see distribution shape. The number of turns is not continuous, so treat as a discrete ordered set of categories and adjust the x axis.

I am also interested in the points themselves, so convert this to a raincloud plot.

Where is the individual champion of the first season of Marbula One?

Finishing touches on the graphic to make the color scheme colorblind-friendly and use fonts that evoke a racing theme.

There is actually a fair amount of variability among lap times, with clusters around certain times. Could there be a relationship between the number of turns and the distribution of lap times?
The boxplots give us a numeric summary for the distribution of the points, but different shapes of point distributions can yield the same boxplot. So I decided to add a violin plot to clarify the shape of how the points are distributed. Also, the turns variable is not really continuous, so I will treat it as an ordered categorical variable.
This gives a good idea of how the lap times are distributed both in terms of shape and numerically. But we may also be interested in the individual points themselves, for example, to see if a particular marble is a high or low performer. A raincloud plot would combine all three of these elements.
It would be interesting to see how the individual champion of the first season of Marbula One performed on this plot. I will highlight the winner (Speedy of the Savage Speeders) with a red dot.
I changed the title font to “Righteous” from Google fonts because I thought this aligned with the racing vibe. I changed the color palette to use a Tol-inspired palette that ought to be more colorblind friendly.

How to Read

Raincloud plots combine:

A ‘split-half violin’
Visualization of central tendency and error (i.e. a boxplot)
Individual data points.

Strengths: inference at a glance, statistically robust, data transparency

Next Steps

Make the same graph with race data
Make paired raincloud plots that compare race lap times with qualifying lap times
Add interactivity so audience can look at individual cars and mouse over statistics