„Keine zwei Packungen Skittles sind gleich“ – außer bei einigen [OC]

Data is from possibly-wrong [here](https://github.com/possibly-wrong/skittles/tree/master) and [The Skittle Maths by Clare Wallace](https://www.clarewallace.co.uk/skittles) was on the most recent [A problem Squared Podcast](https://www.patreon.com/posts/130-sly-skittles-152484396)

I took Clare Wallace’s spreadsheet and used python and mathplotlib to graph all the bags she has counted. [Code is here](https://gist.github.com/cavedave/779193b08673c735f2f0a9e048e0f79f)

[https://possiblywrong.wordpress.com/2019/04/06/follow-up-i-found-two-identical-packs-of-skittles-among-468-packs-with-a-total-of-27740-skittles](https://possiblywrong.wordpress.com/2019/04/06/follow-up-i-found-two-identical-packs-of-skittles-among-468-packs-with-a-total-of-27740-skittles)

You could graph summary values like

Per-Color Totals

Color Total Mean/bag Median Std Dev Min Max

Orange 1,411 3.37 3.0 1.59 0 9

Red 1,386 3.31 3.0 1.78 0 10

Purple 1,392 3.32 3.0 1.60 0 8

Green 1,375 3.28 3.0 1.59 0 8

Yellow 1,322 3.16 3.0 1.66 0 8

Per-Color Totals Big bags

Color Total Mean/pack Median Std Dev Min Max

Yellow (Lemon) 5,663 12.15 12.0 3.24 2 24

Purple (Grape) 5,648 12.12 12.0 3.36 3 24

Red (Strawberry) 5,559 11.93 12.0 3.23 3 22

Orange 5,483 11.77 12.0 3.21 2 22

Green (Apple) 5,269 11.31 11.0 3.21 2 22

but I didn’t

You should sue them for false marketing!

Someone has done the math and they sold so many packs that it’s impossible for them to not have made two identical distributions

So if I get a bonus skittle, it’s pretty sure to be yellow?