The Lesson to Unlearn
paulgraham.com
paulgraham.com
This is extremely tech centric view of the world. If your work function cannot be sold with minimal marginal cost - e.g. technical service provider, consultants, or other service industry work - this strategy may not work for you. This is true for most of traditional engineering, or other non-tech STEM functions in orgs.
As the cost of customer acquisition over net revenue will be significantly higher for you than the incumbent - so might as well join them - and to do so, you will likely need to hack the bad test.
If the SAT were so easily hackable by using the right test-taking techniques, rich parents would have hired tutors to teach their children the tricks, rather than paying people to actually take tests for them, which caused several famous people to go to jail.
I would wager over 95% of rich people do hire tutors (or send children to expensive tutoring programs). Since they don't get sent to jail, we overestimate the percentage of the latter.
The SAT is hackable. In my time, both the SAT and the GRE analogy questions had the same "flaw":
You are given 5 choices. Two of them will always have identical relationships - so neither can be the answer. There is always a nonsensical choice (something like "Dog : mathematics"), so you can rule that out. You're left with two choices. Even if you don't understand the question, you now have a 50% chance of getting it right.
Similarly, for quantitative questions, a number of problems are easily "solved" by working backwards. I don't need to know how to solve a quadratic: I just plug in the choices into the equation and see which gives the correct answer. This is so "common" that people consider the strategy as legitimate. However, the goal of the question is to see if you can solve a quadratic - not evaluate one.
A lot of quantitative problems tend to have at least one nonsensical answer. You can rule it out.
And so on.
Long story short, if you simply checked answers A, C, E for every question you apparently had a very good chance to get a passing grade.
I remember these, but I am actually most proud of being one of the top 100 scorers on the Putnam competition, definitely not a hackable test!
Sadly, I didn't know Microsoft hired top Putnam scorers, no questions asked, at the time (1970), so I missed out on being a Word programmer. Oh, wait! That's a good thing!
1. Usefulness - Is it a useful thing to study and remember - for an engineer medieval history is utterly useless.
2. Interest - is the student interested in the subject beyond grades and usefulness?
3. Signaling - you need good grades to get to the next stage in your life.
4. Energy and Time - students have other things that they want to do outside of school.
Students are making a tradeoff on all these dimensions while optimizing for the grade - this in itself is no easy thing to do.