Category Archives: Opinions

When should you use first principles thinking? When should you avoid it?

“First principles thinking”is a fairly popular phrase out here in the Bay – and for good reason: it’s a powerful mental tool.

Many of the most effective people I work with are exceptional first principal thinkers.

It’s also an area of growth for me. See below for some reflections on the idea.

What is first principals thinking?

The best description I’ve heard is that is it reasoning by root cause rather than reasoning by analogy.

Instead of trying to figure out how something is like another thing, it is more about keeping on asking “why” about the specifics of one thing.

For example, if I were trying to build a cheap rocket to get to Mars, I could look at the price reduction trajectories of other similar complex machines, or I could do a part by part analysis of each component of a rocket.

The first is reasoning by analogy, the second is reasoning through the specific issue at hand.

Yes, Elon Musk is a major proponent of first principles thinking.

Why is first principles thinking useful?

Thinking by analogy is useful in that it is efficient: it quickly allows us to make guesses about unfamiliar issues by comparing them to familiar issues.

However, thinking by analogy is also noisy: the things you are comparing – especially in the case of really hard problems – will never be truly the same. So you are inevitably losing information by making erroneous implicit assumptions.

Ideally, first principles thinking will increase the amount of accurate information at hand, which will help with sound decision making.

Why is first principles thinking dangerous? 

It takes a lot more mental energy to think down to the root causes of every issue you face.

Moreover, the ability to execute sound first principles thinking will most likely require some subject matter expertise. While it’s possible to get to root causes without being a content expert (this is what junior strategy consultants often attempt to do), I’m skeptical that you consistently identify all root causes without deeply understanding the content.

Often times, I catch people who pride themselves in first principles thinking making major mistakes. This is especially true when they are talking to me about education, which happens to be something I know a lot about.

The world is an ever complicated place, and if you don’t have a handle on research and a broad set of real life facts, you can end up with high confidence in first principles that aren’t anchored in reality.

When should you use first principles thinking?

My hunch is that using first principles thinking is most useful in areas where: (1) you have subject matter expertise (2) you have time to devote to thinking about the issue (3) you have external feedback mechanisms that will tell you if you’re right.

I think reasoning by analogy and / or trusting experts / or simply not having an opinion (an under used option!!!!) is best in situations where you don’t have the requisite expertise, time, or feedback mechanisms to conduct first principles thinking.

How can you getter better at first principles thinking?

I’ve been thinking about this a lot lately; here’s some suggestions:

1. Get into friendly arguments with people who are good first principle thinkers and watch them work through issues. I have 3-4 great first principle thinkers in my life and I’ve learned a ton from watching them think. If you don’t have them in your life, listen to podcasts and read blog posts by people who excel at first principles thinking.

2. Use thinking devices such as getting to root causes by asking “why” 5+ times before you settle on viewing something as causal or true.

3. Cultivate anxiety about not being a good first principal thinker. Whenever I’m confronting a thorny issue in education, I always pressure / chastise myself to take the time to do first principles thinking before voicing an opinion.

Good luck.

You Shouldn’t Have an Opinion on Most Issues

Not sure what inspired this post: Vaccinations? Health care? Climate change? Middle East policy? Superbowl play calling?

Anyways, the tagline is this: you should not have opinions on most issues.

Next time you’re at a friend’s dinner party, and someone asks you:

“What do you think about X?”

Almost every time your answer should be:

“I have not studied the issue deeply. But a lot of people have spent their lives studying the issue. You should read their work.”

If your new dinner acquentaince presses on, you should say:

“Alright, alright. I really don’t know, but I do like talking about X, so I’m quite happy for us to both pretend we know something issue, and there’s a small chance we’ll learn something, and it might keep us from having to play Jenga.”

Also, if expert opinion is divided, this doesn’t mean you get to weigh-in without doing your homework.   Instead, you simply note that experts are divided on the issue and you’re not qualified to determine who is right.

This is not to say that you should never question expert opinion. If you are willing to go deep on analyzing complex arguments and reviewing research, you should by all means dig in and try to come up with an opinion. If you don’t devote (at least!) a few months to this project, please keep your opinions to yourself.

To be clear, I don’t think experts are always right. Nor do I think that laypeople can’t sometimes prove experts wrong.

I just think experts are more likely to be right than laypeople, and that the laypeople who prove experts wrong usually have explored the specific issue pretty deeply, or they have personal knowledge of the issue that the professed experts do not have.

Of course, there are many issues that are about values and not about data and research. But I still urge you to be careful, as there are people called philosophers who have thought long and hard about how to select values and then align these values to actions.

In Sum

When it comes to opinions, there’s no free lunch. If you want to have an opinion, do your homework.


There do seem to be some issues where the wisdom of crowds (foolish opinions) is more accurate than expert opinion.

After an expert has identified such an issue, please do feel free to articulate your opinion on an issue which you have no real knowledge.


I’ve succumbed to opinion fever way too many times. It’s a plague that affects most law school students.

Some of my most notable foolish opinions include:

  • I once weighed in on whether or not the Fed should utilize NGDP targeting.
  • I used to hold strong opinions about when the singularity will occur.
  • I argued for socialism as a freshman in college, and my arguments were mostly based on the price of airline tickets from Chicago to New Orleans.
  • I supported the Iraq war.
  • I thought we could close Connecticut’s achievement gaps by giving more money to school districts.

I could go on, but I worry that you’ll stop reading this blog.

I promise that I’ve gotten (somewhat) better.

Unfortuantely, I have a high risk profile for ungrounded opinions. It’s a daily struggle.