The story much of the world tells us is that we need to get a head start at everything if we want to be successful. If you want to be a musician, better start practicing shortly after you’re out of diapers. Want to be succeed at sports, if you haven’t started deliberate practice to get your ten thousand hours in before kindergarten, kiss that dream goodbye.

These stories that society tells us are reinforced by the popular celebrities like Tiger Woods, who did start golfing at an extremely young age. This specialization myth is where David Epstein wants to step in and present an alternative argument with Range. The purpose of range is to help us understand how valuable late specialization is and how much we’ll benefit from a wide range of experience in life.

Epstein introduces the book by comparing Tiger to Roger Federer to show us an alternative narrative to early specialization. Unlike Tiger Woods, Roger Federer didn’t start playing tennis shortly after he stopped needing diapers. Federer played many sports and didn’t start taking tennis seriously until he was 11. To conventional wisdom this put him far behind his contemporaries on the quest to gain the requisite skill to become the player he is today.

Despite this handicap, Federer is one of the most dominant players in tennis today.

Debunking The Cult of the Head Start

Epstein starts Range by debunking the cult that’s arisen around getting a head start to be successful. He cites a bunch of research that shows that from college administrators trying to choose successful students, to HR people choosing successful job candidates, the longer you’ve done the job has no effect on the quality of your predictions1.

This is because both of those domains are “wicked” learning environments. Daniel Kahneman and Robert Klein identified wicked learning environments and “kind” learning environments2. Kind environments are those that have well defined rules and the choices you make immediately result in feedback about the quality of those choices. Chess is a good example of a kind environment. The rules are known, and you get feedback immediately about the quality of your move.

In contrast, wicked environments have ill-defined rules and the feedback may come long after choices have been made. Wicked learning environments don’t have repeating patterns, and sometimes even success isn’t quite clear.

Epstein says that this is why computers are so good at chess. They know the rules and are better at the pattern recognition that chess requires. Asking computers to succeed in wicked environments is knowing that you’re setting them up for failure. This is also why chess is one of the environments that early specialization pays off. It takes a long time to build up the library of patterns that chess has and the earlier you start building your library the better you can be at chess3.

When we know the rules and answers, and they don’t change over time — chess, golf, playing classical music — an argument can be made for savant-like hyperspecializated practice from day one. But those are poor models for most things humans want to learn. 4

Now that we have the basic ideas of range introduced to us, let’s take a look at the advice that Epstein gives us as we look to apply them in our lives.

Applying Range in Learning

How much do you like to watch your kids struggle with learning a new school concept? If you’re like me, and most parents, not much. We end up giving them hints so they can get to the answer faster. We show them shortcuts and fast and easy methods to get through their work.

Unfortunately all of these things are doing the learning of our children a disservice. While we all may feel good when we learn things quickly without struggle this is not what we want if we value long term retention5. Epstein cites a study done with monkeys that showed the more hints you gave them the worse they learned the subject6. Instead of learning, they relied on the hints.

Epstein also called into question many of the worksheets that get sent home from school. Even right now my oldest has a list of multiplication to work on regularly. Range showed that worksheets that focus on the single topic are worse at helping people learn the concepts than using a method called interleaving7. Interleaving would have my daughter doing some multiplication, then a bit of addition or subtraction, and then other math skills. This type of work will cause more struggle intitially, but will help her learn the concepts of math better in the long term8.

Building Teams with Range

Probably one of the biggest question that comes from Range is for companies that want to build teams that can solve novel problems. Range showed that specialized people tend to suffer from cognitive entrenchment, which is when you get stuck in the known solutions and fail to even listen to options that fall outside your experience9. On teams with only specialists, you have no views outside of the collective narrow experience. This resulted in a few studies that showed labs with diverse backgrounds and training were more likely to come up with breakthroughs and take the outside view10.

The outside view probes for deep structural similarities to the current problem in different ones. The outside view is deeply counterintuitive because it requires a decision maker to ignore unique surface features of the current project, on which they are the expert, and instead look outside for structurally similar analogies. 11

Epstein cited scientific breakthroughs that relied on analogies from baking, to sewing. The only reason teams were able to solve these hard problems is because members had vast experience outside the field they currently specialized in and recognized the structural similarities to the current problem they were trying to solve. If you’re working in a wicked learning environment, which most of us are, relying on experience from a single learning domain can be disastorous12.

In the planning phase of projects experience also harms us since the more familiar we are with a project the more likely we are to be optimistic about it’s success13. Maybe this is why software developers always assume projects they run will hit timelines and budgets, while knowing that most projects don’t hit either of these metrics. We assume that we won’t succumb to all the problems that other projects deal with, we fail to take the outside view.

When we look at forecasting outcomes, Range showed us that the average expert was bad at forecasting no matter their experience, education, or how much information they had on hand14. This is because the more they know about a subject the more likely they are to fall into confirmation bias15. Experts are more likely to double down on their incorrect views in the face of contrary evidence instead of looking at it and adjusting their ideas and methods.

The best forecasters turned out to be non-experts in the subjects they were asked to predict. But more than this, the best forecasters didn’t tie up much value in being right. Instead they made their predictions and then looked to the people around them to show them why they were wrong.

The best forecasters view their own ideas as hypothesis in need of testing. Their aim is not to convince their teammates of their own expertise, but to encourage their teammates to falsify their own notions. 16

Unfortunately, this isn’t normal human behaviour. We want to be right and we want others to view us as someone with knowledge. When we turn out to be wrong, we feel like we loose status and that hurts our ego.

Epstein encourages us to combat the failings of experts by taking facts from them not opinions. Experts are great at facts, and this is why they should be part of our teams, but opinions and forecasts should be put together by people not tied up in being right17.

Grit and Match Quality

Grit is another fairly hot topic, which I’ve written about when I reviewed Grit by Angela Duckworth. The way that grit is framed today is all or nothing. Either you stick with things or you don’t and if you don’t we need to fix that because grit is a sign of success.

The problem is that the other side of grit is the sunk cost fallacy. Sunk cost fallacy is the idea that we take all the past expenditures of time and money into account when we decide if we’re going to continue down a path. Instead of taking the advice of choosing and never wavering18 we should discount sunk costs and choosing the best path now.

If I had followed the gritty ideals, I would have moved from my counselling degree to a masters degree and then into a counselling practice of some sort. This would have been my path because I had just spent tens of thousands of dollars on the education I had. Instead I threw it away at 28 and finished teaching myself programming, even taking a job as a junior web developer while I finished my internship in my degree program.

I was lucky in many ways that I had paid for school mostly in cash and had less than $10k in debt when I came out. I was lucky that in the previous 10 years of working I had about 10 different jobs ranging from equipment operator, to live theater, to canoe salesperson, to outdoor guide and instructor. I hadn’t picked a specialty right out of high school and wasn’t worried about the perceived head start my peers may have had19.

The truth is that early specializers get a quick jump ahead in the short term but after about 10 years, it all evens out20. If early specializers picked wrong and went for a speciality they didn’t stick with, then they were further behind than the late specializers who found something they did enjoy and stuck with it.

That’s because the early specializers had poor match quality. It’s not that they lacked grit, because once they changed they often stuck with the new path and were very gritty about it. They just had a bad match and discovered it through experience then made a change21. We see this principle in The Dip by Seth Godin as well. Godin spends the book showing you how to decide when to quit, and encouraging you to quit things that aren’t working.

Getting Started with Zettelkasten

If you’ve been wondering about what Zettelkasten is and how to start organizing your notes with this excellent system then this course is for you. I’ll walk you through how I use this system to develop my reading research for posts like this one here. You can also become a member to get all my courses.

$99 USD (30-day guarantee)

In Range, Epstein cites the spending of the US Army on West Point candidates. The short version is that the less money the Army spends on a student in the form of scholarships the more likely they are to stay as officers22. Those that get scholarships are likely to serve out their required time, which means the Army broke even on the education, and then head off to a better fit with their great education. It’s not that these people were really bad candidates, but they had a bad match quality. A better question to ask them would be, would they pay for their own education at West Point or are they only taking it because of the scholarship?

…a mind kept wide open will learn something from every experience.23

So instead of asking if someone is gritty all the time we should be asking where they are gritty24. This is the power of context in action. Sometimes we are very gritty, and sometimes we aren’t25. When we see where someone is gritty, it’s a decent approximation for match quality and we should encourage them to dig into it.

Giving Yourself Range

To give yourself range, you need to build your own education across disciplines. You need to devour information from fields that seem to have nothing to do with what you do regularly. Epstein talked about how many well recognized scientists were also excellent muscians, magicians, fiction authors, or bakers. The deeper you can dive into the wealth of human knowledge the more range you’re giving yourself.

The larger and more easily accessible the library of human knowledge, the more changes for inquisitive patrons to make connections at the cutting edge. 2627

We need habits of mind that allow us to dance across disciplines28. As I wrote, if you’re going to be successful you need to have lots of mental models. In fact, if you don’t have at least three mental models to apply to a problem you don’t understand the problem enough to solve it29.

As you become more experienced in a domain, you need to recognize that you’re more susceptible to sticking with only the methods you know. Instead, have strong beliefs held loosely and be willing to entertain new arguments30. I’ve previously written about this when I talked about becoming a master in your domain. You’re ready to prove you’re point when you can argue both sides with equal knowledge.

Take the time to track down the hunches you have because over time they’re likely to lead towards something interesting31. Feel free to go cross-discipline with your career, because you’re not getting behind, you’re increasing your value as you broaden your experience.

In professional networks that acted as fertile soil for successful groups, individuals moved easily among teams, crossing organizational and disciplinary boundaries and finding new collaborators. Networks that spawned unsuccessful teams, conversely, were broken into small, isolated clusters in which the same people collaborated over and over. 32

Should You Read Range by David Epstein

Before we dive into my recommendation I have two more quotes for you from the book. The first deals with the comparisons we make between ourselves and others.

Compare yourself to yourself yesterday, not to younger people who aren’t you. Everyone progresses at a different rate, so don’t let anyone else make you feel behind. 33

Epstein brings this up as we compare ourselves to others in our fields. Yes, maybe you picked a speciality late, but you’re far more likely to have better match quality and thus be just as successful long term.

Second, on automation and our work.

The more constrained and repetitive a challenge, the more likely it will be automated, while great work will accrue to those who can take conceptual knowledge from one problem or domain and apply it in an entirely new one. 34

If you want to be of value in the future, then you need to focus on areas that computers can’t match up to the value that the human brain can provide. An easy but great example of this would be any type of speech recognition device and how often they get what you’re saying entirely wrong. Yet, almost only of us can understand a small child speaking in language that is much “worse” than an adult speaking.

So, should you read Range? Yes. If you’re looking to understand how to not be overcome by confirmation bias, Range will help. If you want to build teams that can solve problems, then Range will give you ideas about how to do that. If you have a diverse team, Range will help you structure it in a way that will maximize the utility of the generalists and specialists you already have.

I found Range to be well written and it gave me a bunch of ideas about how to continue my own wide ranging education and make sure that my children aren’t pushed into a speacialty early.

Purchase Range on Amazon


  1. Page 20 
  2. Page 21 
  3. Page 26 
  4. Page 30 
  5. Page 85 
  6. Page 87 
  7. Page 94 
  8. Yes I started doing this with my daughter and yes it annoys her. My explanation from the book didn’t smooth it over at all but she follows along with my random practice 
  9. Page 32 
  10. Page 118 
  11. Page 108-109 
  12. Page 107 
  13. Page 112, 113 
  14. Page 219 
  15. Page 227 
  16. Page 227 
  17. Page 225 
  18. Page 64 
  19. I was almost the oldest graduating student my year. There were two ladies in my program that went back to school once their kids were older but outside of those two I was the oldest by about 5 years 
  20. Page 130 
  21. Page 134 
  22. Page 138 
  23. Page 153 
  24. Page 160 
  25. For more on the power of context see Tipping Point, Atomic Habits and Quiet 
  26. Page 189 
  27. This statement seems especially tragic given how much of the science produced is locked behind paywalls so that almost no one gets to read it unless they pay huge amounts of money or are in education 
  28. Page 49 
  29. Farnam Street has a stellar set of mental models to learn 
  30. Page 257 
  31. Page 259 
  32. Page 280 
  33. Page 290 
  34. Page 53