Fair enough. When it comes to career advice, I prefer anecdotes to large data sets, because there are far too many variables, and it's too easy to torture numbers.
More seriously, anecdotes are terrible when giving advice to fresh high school seniors on their way to graduation. When people think of degrees in certain fields, they gravitate to the successful people (“I’ll study acting and make millions like the famous actors I know”, etc). They don’t think of and aren’t even aware of how many people get those degrees and fail hard.
That’s why they need to see the placement and salary stats to understand what their up against. Once they understand a French Comparative Literature degree has approximately no economic value, then they can make that decision knowingly. Papering over that by pointing out the one leader that happens to have that degree does more harm than good.
At the end of the day, people hire you. Not data. After all, who says that a teenager can tell apart badly interpreted data any better than poor anecdotes?
Teenagers should be able to understand data by the time they pick a college or career.
I think there's a world of difference between "I want to study acting and I'll end up successful like Brad Pitt" and "Hmm, my cousin is in marketing, and seems to have created a great career there, I should talk to them about that."
Anecdotes about people you don't know, and career stories you don't fully understand aren't helpful, but frankly I don't think massive data sets on how much the average graduate earns within two years of graduation (for example) are all that helpful either, except as a broad way to compare yourself. It's such a large number of people, you don't really anything particularly useful at an individual level (even though there are plenty of good uses for data like that).