> Sports are the exact opposite, they are almost completely unpredictable. If code was almost completely predictable, we wouldn't spend so much darn time squas

> Sports are the exact opposite, they are almost completely unpredictable. If code was almost completely predictable, we wouldn't spend so much darn time squashing bugs. If users were predictable, we wouldn't need to spend time prototyping or A/B testing or anything like that, we'd just build exactly what our users wanted from the start. If sports were almost completely unpredictable, sabermetrics wouldn't be as large and a profitable field as it is, and we wouldn't have statisticians working in other fields (such as everyone's favorite pundit Nate Silver) who got their start modeling sports. I totally appreciate the other arguments he's making about why the sports industry presents a potentially lucrative one right now, but the point about unpredictability seems like a nonsequitor. I'd like to think he's trying to make an interesting and meaningful point with it, but it's not coming through clearly at all.