Yes. A decision can only be evaluated using what is known at the time, not what is later known in hindsight. Betting on an inside straight in poker is a bad bet, even if in one particular instance you got the card needed.
How do you know in advance, as a 20something, that you're doctor material, or lawyer material, until you actually go through med school or law school?
What exactly is the "safe" route in life, Nostradamus?
As a 20-something, without going through med school or law school, you don't know if it's something you want to do. As a society, we want this hypothetical 20-something to go to med school or law school to find out. Therefore, we should make failure be an acceptable outcome, minimizing the risk for the potential student. In this case, that means publicly funding their education, such that dropping out doesn't place the student with a massive debt.
-------
For context on the PhD probabilities, a professor will typically be supervising 3-4 students at any given time. A professor's career is typically about 30-40 years, and the duration of a PhD program is about 4-6 years. Therefore, over the course of a professor's career, they will supervise around 25-ish freshly minted PhDs. With an average 2% annual population growth, the population will roughly double over that time period.
So on average, a professor will produce about 25 PhDs, after which there will be one empty position due to the professor retiring, and one new position due to a larger need for professors. Two spots for 25 people is not a good bet, if you do a PhD program will the goal of becoming a professor.
That's my point. You want to talk about personal "bad decisions", but what's a "good decision" and why, given that all of the lucrative careers require a series of risks in order to enter the field and become successful in it?
> we as society can avoid putting up additional hazards
Agreed, but I wasn't arguing against that, just against the criticism of individuals for having personal ambition and taking risks.