 |
| Eight years ago, Julie and Colby got married. I still have my boutonniere from the ceremony. Congratulations, Julie and Colby! |
And on a completely unrelated topic: churrascarias.
Two years ago, I postulated that churrascarias can only exist if they are larger than a certain size. Without that minimum size, there aren't enough customers to reliably eat an entire "sword of food" (I couldn't come up with a better phrase), and so the meat has to either not be available (they just don't cook all the dishes), or they have to keep it in some sort of warming place to keep it hot, even if not necessarily fresh. So, with that in mind, and assuming that all restaurants have the same density of tables, we can use something like
mapfrappe.com to make this:
 |
| A comparison of churrascaria footprints, using Estancia in Austin as the unit of measurement. |
Based on this, it looks like the Austin
Fogo de Chao is roughly 1
Estancia,
Pikanhas in Richmond is about 1/3 Estancia, and
Galeto is about 1/2 Estancia. I was thinking about coming up with something more quantitative, but I'm not completely sure how to correctly account for the saturation and quantization effects in yelp rating data. The easy thing is just to take simple average ratings: E=4.05, FDC=4.32, P=3.71, G=3.87. Another idea is to fit power-law to the rating data, under the assumption that "better" restaurants have purer high-rating to low-rating values, and then looking at the power-law index to see which one is steepest: E=1.357, FDC=1.877, P=0.809, G=1.031. Maybe I'll come up with a better idea tomorrow. In any case, this does seem to have the same relation: bigger=better.
No comments:
Post a Comment