Agreed. In fact, I think it's impossible to get useful evidence for any causal relationships between the founding team and success of a company, because there a

Agreed. In fact, I think it's impossible to get useful evidence for any causal relationships between the founding team and success of a company, because there are feedback loops and confounds everywhere. Here's an example of feedback loops. Suppose investors are biased against young founders. When you look at outcomes, you'll see a higher success rate from the young founders that do get funded because they had a higher bar to overcome the bias. And, of course, investor biases are (mostly) the result of historical outcomes. So eventually investors will become biased in favor of young founders, and the bar for funding will be lower, and you'll have a lower success rate. The system creates a time-delayed (by a decade or so) negative feedback loop that maintains a rough homeostasis of success rate for every kind of company. Economists realized long ago that you can't predict stock market movements by looking at fundamentals, because all the other investors are looking at the same fundamentals, so the consensus is already priced in. This seems study has the same problem, in addition to a data set several orders of magnitude smaller. Don't change your plan based on any such study.