What the author (and really everyone in the Valley) is referring to as "luck" is that fundamentally the whole process is stochastic: there's an unavoidable randomness. Probability is involved. Now, mini rant here: whenever people talk about something being random there's a sense that it's COMPLETELY random: the probability distribution is uniform and anything can happen. Lottery tickets, throwing dice, flipping coins. And, not to put too fine a point on it, that is COMPLETELY, UTTERLY WRONG.
Saying there's probabilistic uncertainty in something doesn't mean the distributions are uniform. It just means given all the inputs you can't always predict all the outputs. No matter what you do, no matter what happens, the result can be one of many things. 1 or 0. Success or failure.
This doesn't mean you can't DO SOMETHING about it though, and that's precisely what people are doing. Figure out the patterns, cut through the uncertainty, and try to find your way to the promised land.
What the author's describing using the language of social science and valley speak is the same thing that the machine learning/AI/robotics community learned in the past few decades. They started out with formal logics and rigid rules, and they learned that to succeed in a random world, you have to embrace the randomness. Build it deeply into your systems and processes and all of a sudden you start performing better than your wildest expectations.
Model, measure, evaluate, pivot, repeat. That's pretty damn familiar. What's it describing? A closed loop control system. Change the terminology a bit and you're talking about a Kalman filter with a feedback controller. What's a learning algorithm but a way to fit patterns to noisy, partially random data?
And that's what this is all about. Luck doesn't mean it's all random, just that you can't control everything. And being good doesn't mean you eliminate the randomness, it means finding the patterns and making randomness work for you.
Is there randomness in poker? Sure. Of course. But there's a shit ton of skill too (I know, because I don't have any of it). Don't confuse a poker game for a game of dice.
This whole conversation is so stupid. Dropping out isn't the same thing for everyone. You really can't talk about dropping out without looking at context and consequences, namely why and how they dropped out and the options that are available to them afterward.
I mean, say you're a kid from an upper middle-class family who's parents are paying for your education at some top-tier college, and you drop out because you have a startup that's growing and making money or got funded by some deep-pocketed investors. Or you're a genius who's been recognized as such by one of the world's most famous and well-connected venture capitalists and given a grant to go off and do something entrepreneurial.
Then you REALLY CAN'T FAIL. Say your company goes belly-up. Who cares? The school doesn't care, you just go back to class and your parents start paying tuition again. You graduate with your degree, except now you have valuable experience and connections that help you get internships and jobs and maybe funding for your next startup.
You dropped out because you had a plan and a social and financial safety net that meant you had meaningful choices about your future. Dropping out was taking a shortcut, the life equivalent of jumping over the edge on Rainbow Road in Mario Kart 64. If you make it, you're hugely ahead! And if you don't, well you get picked up and put back on your original path and maybe you're a little behind but really it's no big deal.
Now think about an alternative scenario. You're a low income kid who was the first in your family to go to college. You're working 20 hours a week because your student loans cover tuition but nothing else at the giant-ass state school you go to. Your parents (you're one of the lucky ones whose family stayed together) are cheering for you but they're also laid-off and unemployment is running out and they're not sure how they're going to pay for dad's pills and anyway they're not sure you're cut out to be in college and they're worried you'll fail. So you drop out because you're struggling in some of your classes, and the shift manager at work offered you more hours and you say why the fuck not because the extra cash can't hurt.
What do you think your options are when that company cuts back and lets you go in a year or two? Think you've saved enough after paying down the interest on your loans that you can go back to school and not work? Think that 2 years of credits and work experience at a dead end job is going to translate into a high-paying position at some solid company?
The point is, if you have a specific plan in which dropping out will advance your career and life interests and also have enough of a backup that you can recover from failure, then you should totally take a break and go do something exciting.
If you don't, if a college degree is your lifeline to a better life than your parents' and you're skating on a razor's edge to be there in the first place, then just Stay In School, kid.
And now the rest of you stop debating the value of finishing college as if it meant the same thing to everyone and get off my lawn. Go back to work.
Saying there's probabilistic uncertainty in something doesn't mean the distributions are uniform. It just means given all the inputs you can't always predict all the outputs. No matter what you do, no matter what happens, the result can be one of many things. 1 or 0. Success or failure.
This doesn't mean you can't DO SOMETHING about it though, and that's precisely what people are doing. Figure out the patterns, cut through the uncertainty, and try to find your way to the promised land.
What the author's describing using the language of social science and valley speak is the same thing that the machine learning/AI/robotics community learned in the past few decades. They started out with formal logics and rigid rules, and they learned that to succeed in a random world, you have to embrace the randomness. Build it deeply into your systems and processes and all of a sudden you start performing better than your wildest expectations.
Model, measure, evaluate, pivot, repeat. That's pretty damn familiar. What's it describing? A closed loop control system. Change the terminology a bit and you're talking about a Kalman filter with a feedback controller. What's a learning algorithm but a way to fit patterns to noisy, partially random data?
And that's what this is all about. Luck doesn't mean it's all random, just that you can't control everything. And being good doesn't mean you eliminate the randomness, it means finding the patterns and making randomness work for you.
Is there randomness in poker? Sure. Of course. But there's a shit ton of skill too (I know, because I don't have any of it). Don't confuse a poker game for a game of dice.