The Formula for Success

On exceptional achievement

Aug 08, 2022

In Arthur Jensen’s analysis of genius, he studies the patterns behind famous geniuses. His research concludes that genius is the result of what he calls the ‘multiplicative model of exceptional achievement’. His idea is that exceptional achievement comes not from any particular factor, but as the result of a multiplicative relationship between multiple traits:

“Exceptional achievement is a multiplicative function of a number of different traits, each of which may be normally distributed, but which in combination are so synergistic as to skew the resulting distribution of achievement. Thereby, an extremely extended upper tail of exceptional achievement is produced. Most geniuses are found far out in this tail.”

The general idea is that by combining a number of unique skills or traits where you’re highly above average, you become much more likely to achieve exceptional outcomes than if you only had a couple, or if you were merely decent at a large number of things.

Jensen identifies 3 of these synergistic traits that are common in the exceptionally successful:

Intelligence
Productivity
Creativity

This multiplicative model is an accurate and under-recognized pattern in successful people. However, there are many more traits that can result in these multiplicative effects than the 3 Jensen lists. And while there’s little we can do to increase our intelligence, other multiplicative traits are easier to improve. We should collect and develop as many of these multiplicative traits as we can.

Multiplicative traits fall into 2 categories: extreme dispositions and extreme skills.

Extreme Dispositions

In Vitalik’s post on convex vs concave dispositions, he discusses 2 common worldviews.

In the concave worldview, the optimal policy is to take some middle ground. In the convex worldview, the best strategy is to stick to an extreme.

Most people naturally have concave dispositions – they find work-life balance, pursue conventional career paths, and have big 5 personality traits that are near the mean.

But typical dispositions are unlikely to result in atypical success; extreme outcomes are most likely with extreme inputs. Indeed, many successful people are well known for their unusual traits. These people often have convex dispositions that put them in the extremes of the distribution.

Here are some more convex dispositions:

Working 60+ hours a week
Having very high conscientiousness
Consuming far more information than average
Never eating sugar

There are many contexts in which concave policies are superior (Vitalik gives plenty). But for a large number of traits that are correlated with success, being convexly disposed is better. Not all dispositions are easy to change, but others are tractable, and working toward becoming more convex is worthwhile.

However, sitting at a single extreme is often a bad strategy. For example, being very disagreeable can be useful for thinking independently, but can also make you an asshole. Instead, what you frequently want is not to stand at one extreme or find some ideal mid-point, but to stand on both extremes of a policy simultaneously. On a meta-level, you should be concave about your object-level convex dispositions.

Peter Thiel captures this idea in his interview with Tyler Cowen:

COWEN: Let’s say you’re trying to select people for your Thiel fellowships, or maybe to work for one of your companies, or to start a new company with. Just you, Peter Thiel, as a judge of talent, what trait do you look for in that person that is being undervalued by others? The rest of the world out there is way too conformist, so there must then be unexploited profit opportunities in finding people. If you’re less conformist, which I’m very willing to believe, indeed would insist on that being the case, what is it you look for?
THIEL: It’s very difficult to reduce it to any single traits, because a lot of what you’re looking for, are these almost Zen-like opposites. You want people who are both really stubborn and really open-minded. That’s a little bit contradictory. You want people who are idiosyncratic and really different, but then who can work well together in teams. And so, this is again, maybe not 180 degrees opposite, but like 175 degrees.
COWEN: This is why you like Hegel?
THIEL: I don’t like Hegel that much.
[laughter]
THIEL: I think if you focus too much on one or the other end of it, you would tend to get it completely wrong. I like to get things where you get these combinations of unusual traits, so if you have people with some really interesting, very different ideas, that suggests we’re in the idiosyncratic category.

My favorite example of these “175 degree” traits is being both very short-term oriented and long-term oriented. Or as I say, ‘move fast and be patient’. Think of the most successful founders of the past few decades: Musk, Bezos, Gates. They needed to have both excellent short-term execution and long-term vision. Musk needed the clarity of vision to build a multi-decade mission of getting humans to Mars, in addition to the short-term focus to work intensely, day-after-day.

Extreme Skills

Another area for accumulating multiplicative traits is in your skill set.

Scott Adams has this idea of talent stacks where he points out that becoming the best at a single skill is very hard, but becoming the best at a unique intersection of skills is far easier. This is because there’s a large number of valuable combinations of skills to have, and few people have any particular combination. In his case, he combined his business skills, work ethic, risk tolerance, and sense of humor to create Dilbert, one of the most successful comedy strips. While he’s not the best along any of these dimensions, he’s good enough at each that their multiplicative effect made him enormously successful.

But don’t just become good, aim to push your talent stack to the pareto frontier. Develop your skill sets until there isn’t anybody who is as good or better than you at every skill in your talent stack.

For younger people still deciding their career path, I recommend developing a talent stack related to an exponential technology. Technologies moving along exponential curves will have immense impact and face talent shortages, providing unusually high chances for massive returns. Here are some interesting examples:

Cryptography, distributed systems, and economics
Machine learning and law
Computational biology and philosophy

The more synergistic skills you have in your talent stack, the better your chances of extreme outcomes.

Conclusion

Extreme success is the output of the multiplicative relationship between a number of outlier traits. Convex dispositions and talent stacks are 2 categories of multiplicative traits that can be developed. Combining valuable talent stacks with convex dispositions leads to even greater multiplicative effects, pushing you even further out on the success distribution.

So start multiplying.

Trace’s Newsletter

Discussion about this post