In thinking back on the most effective developers I've ever known, there's a pretty wide range of backgrounds. Though I guess most of them do have some sort of university degree, not all of them were in CS and most were not top-tier schools.
I can say with confidence that I've known serious over-achievers in school who went on to become non-achievers in life, and vice versa. In other words, I'm not sure that what school someone attended is a good indicator of their ability or their enthusiasm for the work. The more top-tier school alumni I meet, the less confidence I have in the predictive value of that data point vs. other data points.
I have worked in finance, and I am currently working on my Master's degree in financial engineering. A lot of finance-oriented companies highly value their code. Now, yes, there are projects like QuantLib someone can contribute to, but if you someone was hired directly out of school and had never contributed to Open Source, chances are you aren't going to find any Open Source projects associated with that person. That person's employer probably doesn't allow it. With that said, the other case may be that they simply don't have the time to work on anything outside of work. Again, Wall Street is like this.
Ummm, there is a huge difference between a 9 and a 10. The horse that wins the race by a nose wins 10 times more prize money than the second place horse. Was the 1st place horse 10 times faster? 10 times stronger? No, he was a nose faster - and it made a huge difference.
Think about this in terms of search engines. are Googles search results 10 times better than Yahoo. Absolutely not, but they are a nose better and thats why we all use Google.