We are inundated with successful people. The media floods the public sphere with them; their stories, images, videos, interviews, (and more rarely) books. In fact, the public sphere has ended up being the sphere not of the public in the broad sense, but of these people. Because you see, these are very few, and it could not be otherwise; it is impossible by definition. Success today is fundamentally an elitist notion. The way it is understood in the Western world (which is the notion of success I assume in the rest of this article), success requires us to be in some way superior compared to the people around us. If everyone is doing equally well, then there are no successful people. This is why even if you find yourself to be part of some kind of insulated Silicon Valley echo chamber composed of conventionally successful people, you cannot feel your success unless you step out of that environment, so that the difference with the “commonfolk” become obvious. This is the despicable amour propre Rousseau talked about: “self-love” that can only exist by comparison. In short, by definition, most people cannot be successful, let alone all people. If this makes sense, congratulations; you just understood the theme of this whole article.

All too often, when these people become successful “enough”, someone eventually asks them this mundane but crucial question: “how did you become successful?” As an aside, note that these folks may be few, but they are nevertheless all kinds of folks: professors, entrepreneurs, instagram models, chefs. And yet despite this “diversity”, the reply to that question almost always follows a commonplace pattern: hard work, sacrifices, the ability to predict well, the management of one’s emotions, making hard decisions, and all this American “I am self-made” crap. The common characteristic of all these “secrets” is that they are a result of the successful person’s own doing. In other words, it is because they worked hard, they made sacrifices, they made good predictions, and generally, the narrative implies or explicitly states that it all comes down to things they did and chose. After all, this is what being self-made means!

After they humbly let us know how they became so great, they then tell us that we can do it too. How marvelous, how fortunate; why, how encouraging! Hey, it may even seem like a truism: since what produced the effect was events of their own doing, then we must be able to do it too; it’s all about doing and choosing.1

Nonetheless, the devil lurks in this little verb: “can”. To understand that, let’s consider a simpler case than entrepreneurs, professors, and the like: lottery winners. Sometimes even these people go on TV and tell everybody “you can win the lottery too.” And well, that may be true, but it depends on how we interpret “can”. If by “can” we mean “there is a probability—or you can easily create it—however small,” then sure, you can too. You can buy a lottery ticket tomorrow and there will be an infinitesimal but nevertheless non-zero probability that you win the lottery. Thus, indeed you can win it in the sense that yes, it can happen, meaning: it is not known to be impossible.

But there are two aspects that are hopefully obvious when it comes to the lottery. First of all, the probability is unimaginably small (which is why folks have developed all sorts of analogies to help people’s intuitions understand how truly small it is). Second, whether you win or not involves exactly nothing of your own doing, except for obviously buying the ticket. Hopefully, no one interprets “you can win too” as “you can do something—anything at all—to make it happen.” It’s literally just luck. Even if you consider Chaos Theory, if it so happens that you (or a butterfly) do something that causes the result, it cannot be something done intentionally and calculably.

While all that is obvious for the lottery, it is usually not obvious how they transfer in most other domains in life, like entrepreneurship, professorship positions, etc. Because of the way the narratives are presented—I worked hard, I made sacrifices, I, I, I, …—when the phrase “you can do it too” eventually pops up, we do not interpret it as “sure, there is some non-zero probability that it could happen.” No. We do not even interpret it as “there are things that I can do that will push—nudge perhaps—the state of affairs towards my goal.” Nope, that’s still too weak. Rather, we interpret it as “there are things that I can do which will bring about the desired result.” Like, the implication is that if you work very hard, you will become successful. To put it plainly, it is almost certain. “Almost certain” means “unless something extremely strange and improbable occurs, it will happen.” Thus, your life and your future are in your own hands, under your control, unless extremely weird shit takes place.

Well, unfortunately this is not true. The American Dream wants you to believe it is true, and so does your PhD advisor, but it is literally and statistically false. You don’t have to believe me, just look around first, and then at an even wider scale. For every successful person, you will easily find many who are equally or more competent, equally or more responsible, equally good or better decision-makers, etc. You just need to make sure you do not fall into the trap of judging in hindsight. Let’s say, for example, person A made decision (x) (e.g., person A bought some stock S) and person B made decision (y) (e.g., sold stock S). In hindsight, it may be clear that (x) was a much better decision because the stock value skyrocketed. But this is a posteriori thinking, or judging after-the-fact, and it is basically useless when it comes to judging people’s abilities—or at least conscious abilities. After-the-fact thinking also involves biased preferences for people who became our role models later down the line. This is also why you should be careful when praising the surprisingly fancy analyses of famous economists which, however, are presented to us only after a financial crisis has happened. It is one thing to understand and explain an event after it has happened, a completely different thing to predict it, and an altogether different sort of ability to be able to predict accurately with consistency. So, in short what we need to ask is: given what we knew, or what A and B knew, when they made the decision, could it be deduced that one option was clearly better than the other?

I’m pretty sure that if you play this frankly demoralizing game for even just a week, you will figure out that yeah in fact there does not seem to be a reason that person A is where she is primarily because of their own doing. I am obviously not saying that competence plays no role, that what they have done does not matter at all, or that you may as well just sit all day staring at the ceiling since it’s all luck anyway. No, it’s not all luck. Competence, skill, dedication, and most of all obedience; all of these matter, and their contribution to someone’s success is non-trivial. What I am saying is that the luck factor is non-trivial. It usually plays a significant, and in many cases the primary role (although it is of course difficult to calculate precisely). One of the best examples of the role of luck is found in hockey:

[I]n many years 40% of hockey players selected into top tier leagues are born in the first quarter of the year, compared to just 10% in the fourth quarter; an early birthday can make you up to four times as likely to be a pro hockey player [...] Those born in the first part of the year are a little older and so on average bigger and faster than kids in their league born late in the year.

As the narrator tells us, we might think that “as they grow up, this difference should eventually shrink to nothing, but it doesn’t.” Why?

Because the young kids who share the most promise are given more time on the ice and enter more tournaments, where they receive better coaching and improve their skills. And these advantages compound year after year, so by the time you get to the pros, birthdays are heavily skewed towards the start of the year. But does any professional hockey player feel thankful for their birthday? Probably not.

Here is another, personal example. My great uncle graduated at the top of his class of the then (and still) highly selective and top military academy, the Hellenic Military Academy (Evelpidon). This was quite an achievement for meritocracy if you consider that he came from a middle-of-nowhere village somewhere in Peloponnese. But he was a bit unlucky: he died in the Civil War. Oops… That unfortunate event did not influence just him; it carried over to his brother, my grandfather. At the time, my grandpa had just entered the, again, highly selective Law School of Athens. But since his brother died and they did not have his salary to help the family anymore, my grandpa had to drop his studies and go back to the village to work the fields. Oops… To make things worse, my grandpa was objectively (and also according to both him and everyone else in the family) not cut for such a job. This was to be expected as he had not been doing that from an earlier age. Thankfully, in collaboration with my grandmother (talk about women’s contribution to breadwinning), they were good enough. But you could be the best farmer at the time and it wouldn’t matter; you wouldn’t make much anyway. Thus, their children did not grow up exactly affluent.

So there you have it: two people who grew up in a pretty poor and socioeconomically lowly family, made the leap through their own talent and effort. They had been the only people on both sides of my whole family to ever get into tertiary education institutes before my generation. And yet, luck undid that. One unlucky event carried over from my great uncle to my grandpa, and there from him to his children, from his children to their children, etc.

To get back to our point, what these successful people fail to bring to the foreground of public awareness is how much luck has contributed to their own success. They may do it out of ignorance, or more probably—because these people are usually not dumb—they do not admit it because it is not in their best interest to do so. For example, as a PhD advisor, it serves you well if your students believe that their success is primarily determined by how hard they work. Why? Because you can beat the odds through aggregation.

To see this, let’s return again to our lottery example. Sure, each individual ticket has hardly any probability to win. However, if you buy all the tickets, then you suddenly have a 100% probability to win. More specifically, and this will become relevant shortly, the probability that at least one of the tickets you hold will win becomes 100%. More realistically, the more tickets you buy, the more you increase your chances that at least one of them will win. This should be fairly intuitive, but it is sometimes hard to transfer that intuition to career relations.

You see, your advisor shares part of your success; credit, papers, money, etc. This is also true with regard to every student in your lab. Now, it is true that if you work hard, you will increase your probability to succeed, and so also your advisor’s. In reality, your individual probability may increase only slightly through hard work. But if many people work hard, then collectively they increase significantly the probability that at least one of them will succeed, similar to buying more lottery tickets. Of course for any one of these people—like you or any one of your lab mates—that does not matter at all, because each of you structurally holds only one “ticket”. But for your advisor this changes a lot, because he holds stake on all the “tickets”! As a lab becomes bigger, even a slight increase in each one’s probability to succeed increases the advisor’s cumulative probability to succeed tremendously. So, it serves an advisor well if his/her students think that working hard helps them a lot; it may not help them a lot, but it helps him/her a lot.

The other reason successful people do not want to admit the role of luck in their success is simply because it robs them of some credit. It feels much better to claim that “hey, this is 100% the result of my effort, competence, ingenuity, empathy, strategy, etc.” than admitting that well, perhaps only 40% or even 60% is my own doing (and authorship is always relevant in such matters). After all, no one wants to give promotions or accord higher salaries to someone just because they got lucky. If you just got lucky, then the implication is that you do not deserve any rewards.

Just in case you want to know the jargon people use for the effect we’ve been discussing, it’s called “the survivorship bias.” There are many ways to explain the origin of the term, but one analogy is to imagine that there is a plane with 100 people in it, which crashes and only one survives. Now, imagine that survivor saying that “if I survived, you can survive too.” Again, if “can” means you can control it or bring it about, then it is as preposterous as the lottery case.

In general, the survivorship bias is the logical fallacy in which someone focuses only on those who succeeded (e.g., the winner of the lottery) and ignores all those who failed. Our discussion is a more constrained case of this fallacy: thinking that your success is your own doing when to a big extent it is random. In any case, the media is set up to perpetuate this fallacy since they only show us those who succeed. For instance, every time someone wins the lottery, they may bring her on live TV. Of course what they do not do is take an interview from the million other people who did not win, and who had equal probabilities.

But the survivorship bias is only part of the story. At its basic and most common form, the survivorship bias involves individual elements. For example, an advisor telling an individual student that she can do it because he did it. But this narrative can be extended to whole sets of people. This is when the “you can do it too” narrative turns into “you can all do it too.” Let me explain this with an example.

A while ago, a professor got into a Philosophy class and told students “you can all become Philosophy professors.” Then a student raised his hand and objected that this is statistically impossible. That in fact if you consider the average number of available positions every year across all of the U.S., that number could barely accommodate the students of even this one class alone. Imagine now opening this up to all classes across the U.S., and the probabilities of this particular class which that professor addressed fall hard to the ground. Unavoidably, this forced a “minor” qualification on the professor’s part: “No, I just meant that each one of you could become a Philosophy professor.”

Well this “minor” clarification is of monumental importance (and in fact still imprecisely phrased). It turns the “you can all do it too” to “any one—and only one, or at least far from all—of you can do it too”. The difference is best illustrated by turning again to the lottery. You see, if all of us buy a lottery ticket, each one of us has a probability—tiny as it may be—to win the lottery. But it would be patently laughable to assert that all of us could, simultaneously, win the lottery. That probability is literally a square 0.2

While this is obvious for the lottery, it is not so obvious for professorship positions, especially when you are on the receiving end of the kind of pep talk that that Philosophy professor blurted out. It simply gives the impression to students that indeed all of them could simultaneously become professors. The same is true for the American Dream. The U.S. governments and industries have been telling people “you can all become rich and successful.” But in fact no, this is statistically impossible. At best, each one of you has some probability to become rich and successful. Of course what they don’t tell us is how small that probability is. “Them” includes the U.S. government, your bosses, your Philosophy-professors-turned-motivational-speakers, and your PhD advisors.

Lurking underneath everything we discussed is (in)equality. Let’s return to the lottery example. Under the assumption that each one buys exactly one ticket, then the probability to win is equal for everybody. However, this is neither true for the American Dream, nor for professorship positions. When it comes to both, if your dad is rich and “successful”, you have—statistically speaking—a way higher probability to become rich and/or “successful” and/or a professor. For instance, it’s a well-known fact that in the Yale Law School there are more undergrads from the top 1% of the economic rank than all of the bottom 50%. Under this light, we should be careful how we interpret the phrasing “I got into Yale.” In general, there are a myriad of factors giving people an edge—which successful people conveniently forget or ignore—that are nothing of your own doing.

In general, inequalities due to socioeconomic and political disparities are well-known. Those then give rise to the famous demand for equal opportunities, which is a call for justice. I think we can all agree that justice is more than “all children should have the opportunity to enter the same competition.” We can probably tell that this alone does not guarantee justice, but what else is missing? Daniel Markovits, a Professor of Law at Yale, finds the answer in the notion of investment. He came into the center of the Internet spotlight when he took part in a famous debate about meritocracy at Oxford. In it, he said:

Meritocracy in fact is not, as we just heard, that advantage turns on effort and talent. It turns on effort, and talent, and one more thing: investment [...] Meritocracy, when it works as designed, favors those whose parents are in a position best to invest in them; who have the money, who have the skills, who have the free energy and space to do so.

So, in a broad sense, equal opportunities are both: (i) equal opportunities for action, and (ii) equal opportunities to get support.

Nevertheless, this leaves a lot of injustice unaccounted for. Say we have children A and B. Even if both are given the same opportunities and support, and if both devote the same effort, A may still end up in a socioeconomically unequal position compared to B. Why? Again, it is the reason we have been discussing all along: luck. But that alone does not illuminate the role of luck enough. You see, when people think of luck, they probably think of e.g., the lottery, or not dying in war like my great uncle, or in rare cases of being born in January; we discussed all those things.

But there are less obvious factors causing inequality. Perhaps the easiest to consider is talent. Talent is by definition inborn. It’s a natural ability or trait, and so (just like Locke’s natural rights) you are born with it. So, what exactly is of your own doing if you are born with perfect pitch? Some people argue that when it comes to perfect pitch, one dominant factor is the native language. For example, native Mandarin speakers have a higher probability of developing perfect pitch because Mandarin (unlike many modern Indo-European languages today) is a tonal language. Ok, so now nurture is entering the picture. But does that change anything with regard to justice? Is there anything of your own doing if you are born in a household or in a society that speaks Mandarin?

As a consequence, equal opportunity is a necessary but not sufficient condition for justice, and John Rawls recognized that fact.3 In his magnum opus A Theory of Justice, Rawls writes:4

Intuitively [the system of equal opportunities] still appears defective. [...] [E]ven if it works to perfection in eliminating the influence of social contingencies, it still permits the distribution of wealth and income to be determined by the natural distribution of abilities and talents.

And of course this natural distribution is random (and non-uniform) and thus unjust,5 as Rawls tells us in exactly the next sentence:

Distributive shares are decided by the outcome of the natural lottery; and this outcome is arbitrary from a moral perspective. There is no more reason to permit the distribution of income and wealth to be settled by the distribution of natural assets than by historical and social fortune.

But even natural assets are not the end of the story. There are even less obvious inequalities caused by characteristics which—if we’re speaking influenced by the world of finance—are not “assets”. For example, your taste. Generally speaking, you do not choose what you like doing. Just as you do not choose to like a girl you see on the street, you also do not choose to like AI or botanology. Even the phrasing is awkward; nobody incarnates a degenerate version of Ash Ketchum by saying things like “AI, I choose to like you” or “Debby, I choose to like you”. No, people say things like “I like AI” or “I like Debby.” So, our very language indicates that it is not of our own doing. Not only does it not have to do with our effort, skill, etc., we also do not even get to choose what we like. We may again get into all sorts of nature vs nurture discussions around the causes of taste, but these are irrelevant because in any case we simply do not get to choose. Accordingly, if you love doing AI and the guy next to you loves doing botanology, it is probable that in this day and age, you are making way more money than him. But how much of that is your own doing?

One may of course argue that sure, you cannot choose to like AI, but you can choose to do AI; otherwise stated: “you can do it too.” That is true, but if your heart likes botanology, you are still almost always at a disadvantage when it comes to the superficial criteria for success our society promotes. Some argue that you could potentially be as technically proficient in AI as a guy whose heart loves AI. Sure, this could happen—in theory at least—due to all sorts of combinations of reasons: you may be equally or more intelligent, or equally or more responsible, equally or more lucky, etc. But even in the off chance that this is true, what about your soul? Yours will probably suffer, especially by comparison to that other guy. Consider an analogy with sexual preferences. If you are a gay woman, indeed you can choose to only engage in straight sex. And in fact you may be equally skilled in bed as a straight woman. But what about your soul? That life would be soul-crushing…

This is a form of injustice that I have literally never heard anyone talk about. I don’t pretend to have a solution, but let’s start by accepting that it exists! Whether the society you are part of values or not what you like has nothing to do with your own doing. It is up to what Rawls called “the natural lottery,” which, like the normal lottery, is all luck. And should you happen to like something the society does not value, then you are at a disadvantage.

1. There is an implication of course, which is that if we do not succeed, then it is only us who we have to blame. This creates a whole series of problems which I do not take up in this article, but Michael Sandel addresses them in his book The Tyranny of Merit.↩
2. Assuming that not everyone has the same number, which in practice is always true.↩
3. Michael Sandel, who studied Rawls carefully, concluded that meritocracy is definitely not sufficient and perhaps even unnecessary.↩
4. A Theory of Justice, Belknap-Harvard Press, Revised Edition, sec. 12, p. 64.↩
5. Technically speaking, this requires the assumption that morally arbitrary => unjust, but this is arguably something Rawls agrees with, and the same is true for me.↩