Training Yourself for Path-Independent Decision Making

History doesn’t matter (for you)

Teng Rong
8 min readMay 22, 2021

Walking on the other side of the street.

Path-dependence is a phenomenon that we are all familiar with, even if the term itself is foreign to you. In short, you are path-dependent when your first decision has a disproportionate and lingering impact on your second decision, and so on, and so on.

We naturally have inertia when it comes to our decisions. We need reasons to deviate from our current path, even if we did not have reasons to be on this path to begin with. When you are walking down a street, you tend to stick to one side. In fact, you need some rationale to cross the street: maybe there is a drunk guy on the sidewalk, or there is less sun, etc.

But what you don’t realize is that you had no rationale to be on this side of the street in the first place. It’s not like you conducted a cost-benefit analysis and decided that the west side of Yonge street is better. You just ended up on this side. By random chance.

And now you won’t leave it.

This is path-dependence.

Your decision to stay on this side is dependent on your prior decision to stay on this side. Or in other words, if you started as a blank slate, you may not have chosen the same side that you found yourself walking.

There are many ways to think of path-dependence. You can think of it as decision inertia: its less mentally taxing to stay the course, so we are resistant to change. You can think of it as activation energy: it requires energy expenditure before a change can be effected. Etc, etc. However you think of path-dependence, it boils down to when the following two questions have different answers:

  1. You chose A. You are to choose between A and B, what would you choose?
  2. You have no current position. You are to choose between A and B, what would you choose?

An example would be the decision to buy and sell a stock. Many people (or, I suspect, most people), keep holding a stock only because they bought it before. Astute traders ask themselves a question every morning: “would I buy this stock if I didn’t already have a position?” — and if the answer is no, they sell their position. The logic is that if you wouldn’t buy a stock, why would you hold it?

You hold it because of path-dependence.

When you are path-dependent, you are committed to a position and you need new information to change your mind. This is not an unreasonable position insofar as you used information to make the initial decision. But the real question here is “how much data do you need to change your mind?”

If you pay attention you may realize that there is an information asymmetry. It takes almost no information to enter a position from a blank slate. Once you are committed to this new position, you need a lot of information in order to exit it.

In other words: path-dependence is evidence asymmetry. Logically, the degree of proof you used to enter the position should be the same degree of proof you need to exit it. If your belief is “it will rain tomorrow because it is cloudy today,” then you should not need the National Weather Service to issue a 0% P.O.P. forecast before changing your mind. All you should need is “it is now sunny”.

Note that I am not examining the validity of any of these beliefs and evidence. I’m talking about the degree of proof asymmetry as a proxy of how entrenched you are in an existing position — or in other words, how path-dependent you are.

Take conspiracy theorists for an example. A conspiracy theorist might claim, for example, that magnets stick to vaccinated arms. The evidence is 2 YouTube videos. The degree of proof here is anecdotal with small sample size, so logically, it should take no more than a few YouTube videos debunking it to return the believer to a neutral position. In other words, if it took 2 YouTube videos to help you to enter a position, then 3 YouTube videos should be more than sufficient to help you exit it. But the conspiracist will demand NIH statements and double-blinded experiments, and probably will not change his mind even if those appeasements are provided. He continues to ask for more, and more, and more evidence against his position — and still, none of it matters. He is extremely path-dependent and has already entrenched in his position.

Secondo: the retail investor. Suppose he buys a stock called $XYZ. He bought $XYZ because he read on the internet analysis that it has a lot of near term growth as a tech unicorn. Then, armed with this analysis, he goes into the annual report and cherry picks all the information that supports this hypothesis (see: confirmation bias) — for example, the company just signed a new client. The next year the stock performance is abysmal. The predicted outcome failed to manifest itself. Instead, the company stagnated after the first quarter and the stock plummetted.

And yet he continues to hold. Maybe because “this is just a temporary blip”. Maybe because “its okay, once they turn around I will make my money back.” Maybe its because “the virus made it hard for business to grow but soon $XYZ will be back on track”. Our retail investor is path-dependent. He will keep holding until there is overwhelming evidence against his position, even when he didn’t have overwhelming evidence to enter it to begin with. He loses money, rinse and repeat.

Why is path-dependence a bad thing?

  • it impairs your ability to make impartial, rational decisions
  • it locks you into current states and positions to your own detriment
  • it makes you less receptive to new information, less responsive to change
  • it reduces small random deviations that carry little downside and the potential for asymmetric payoff

Path-Independence is a lifestyle

Remember back in March 2020 when the world ground to a halt? Of course you do.

But do you remember when Warren Buffett bought billions of dollars of Delta Air Lines in early March— and then sold it all in April? He even took a big loss. “We have sold the entire positions,” he said. “When we change our mind we don’t take half measures or anything of the sort.” (source)

Ask yourself whether you could have done the same thing. Could you change your mind about seven billion dollars in the span of less than a month? Most of us couldn’t — especially not at a steep loss.

Warren Buffett, like all the great investors and traders, is decidedly path-independent. When there is evidence that he was wrong, he exits the entire position and moves on. No dawdling, no “but maybe it’ll come back up”, no gambling on an uncertain future. These people can change their minds in an instant. George Soros, according to Nassim Taleb, changes his mind in the middle of the day for positions entered in the morning. Every day in Soros’ life is a brand new day and he is unencumbered by his prior decisions. This path-independence makes him a master speculator, just like how Buffett’s path-independence helped him become a great investor.

(If you want another Buffett example, look into how he switched from IBM to Apple in 2018).

So why are you so stuck to your current positions?

We are naturally path-dependent. It helps us conserve mental energy. It helps us make decisions in uncertainty (stay the course!). It reduces existential and life anxiety by making things feel more known than they are. It may have been a valid strategy when humans lived in small hunter-gatherer tribes, but we are now in the midst of modernity. There is complex systems with trillions of bits of information zipping past us every second. We can no longer afford to use this strategy.

In the past, path-dependency has helped me lose a lot of money, and made me stay in places and jobs I didn’t want to stay in. There were stocks where the writing was on the wall, but I did not act because I demanded a steep asymmetry between proof-to-enter and proof-to-exit. I realized this, and then I committed to train myself to be more path independent.

How to (slowly) become path-independent

I use two strategies to help myself be less path dependent. The first strategy is used in day-to-day life to reduce my psychological inertia against change. It also introduces helpful randomness to my life, which among other things makes my life slightly more interesting.

Strategy 1: make many small changes without reason.

If I have decided to do something on a particular day, sometimes I will remind myself that I should change the date for no reason whatsoever. This is because I had no reason to choose this day in the first place, so I should be free to un-choose it just as easily. This applies to the choice of restaurant, choice of the dish within the restaurant, choice of road to take (among equally safe highways), the choice of writing in the morning as opposed to writing in the afternoon, etc, etc.

Doing this helped me realize how much of my life and my natural tendency is path-dependent (most of it!). Every time you do something, it becomes more entrenched in your behaviour. So if you are lugging the burdens of your prior decisions through mundane life, then you are going to be path-dependent when it really counts. So choose to make a change for no reason whatsoever: do something just because you feel like it. Un-do something for the same. Change your plans at the last minute (as long as it does not impact other people). Be fickle. Be path-independent.

Strategy 2: pre-define an exit condition.

To un-entrench yourself in a position, it helps to define the conditions that, should you encounter them, you must exit the position. With this strategy, you decide ahead of time when and why you must exit so that there isn’t any inertia to keep going down the same path. With an exit condition, your path is conditional, and your decisions become less path-dependent.

If I buy a stock, for example, I might say “I will buy $OPQ but I must exit if the YoY top line growth falls under 5%, or by May 2023 if the company still haven’t recovered their previous clients”. If I am deciding to go to law school I might say “I will drop out if I fail more than one course in the first year, or if tuition goes up more than 50%”.

If you are a trader you might recognize this as a stop-loss (for the non traders, its a price floor where you must sell). This strategy is a generalized stop-loss that can be applied in many aspects of life. If you are trading a stock you might not sell but for your stop loss. Your mind and its path-dependence exposes yourself to the possibility of ruin. The stop-loss protects you against catastrophic loss by making it difficult to be path-dependent.

— —

All my articles are dedicated to the public domain under the terms of the Creative Commons Zero licence. Please translate, copy, excerpt, share, disseminate and otherwise spread it far and wide. You don’t need to ask me, you don’t need to tell me. Just do it!

--

--