/PIZZA-PIZZA

Primary LanguageJupyter Notebook

Making Great Pizza Is Easier Than You Think

I love to cook. But what I love even more is knowing how to cook—why certain recipes and techniques work and why others don't. As a Chicagoland native I've always been curious as to the inner workings of Pizza. I think there is something about growing up in the only part of the country with two distinct regional pizza styles — Chicago Deep Dish and Chicago Thin Crust — that makes you realize early on that there is no one “right way” to make a pizza. But that doesn't mean there are no wrong ways. I wanted to see what was constant across the many variations of pizza that have popped up everywhere Italian immigrants have settled, and what was truly distinct about these styles, if anything.

To do this, I went to my battle tested source for pizza recipes, Brian Lagerstrom, and calculated the baker's percenatages for each of his recipes, and took note of other attributes like rise and bake times, and method of shaping.

My most interesting finding was that the fat content and the water content of pizza doughs seem to be inversely correlated. Styles with high water content have low fat content and vice versa. I bet that I could come up with my own original pizza recipe by choosing a point along this curve.

The other point that I knew going in but wanted to emphasize was that there are totally recipes — namely Chicago Thin Crust and Sicilian Style — that are really easy to make really well at home as they come together quickly and don't require any advanced pizza tossing skills.

I summarized these key points in two Datawrapper plots. This project was a really fun exercise in choosing how to visualize information, and especially in showing restraint. There were a lot of things I could have plotted and felt compelled to plot - like part-to-whole visualizations of recipes, distributions of other ingredient quantities — that wouldn't have really been useful to the reader.

This project is going on the shelf for now, but were I to revisit it down the road, perhaps when I have more experience with ai2html and scrollytelling libraries, I think I could level up the production quality with some bespoke visualizations and interactives.