Mental Models for Crypto

Ramblings on the ways I like to do research, and how to optimise them.

Mar 07, 2022

If someone were to ask me to simplify the journey of life, I would say that it's a long line of decisions(ultimately - a tree of decisions). In which each decision taken would pivot the line somewhere else. These decisions will then change the outcome of the future. So the question on my mind is — how can we make the "best" possible decisions which will lead to favourable outcomes?

Let me introduce you to the concept of Mental Models, how these models can help us make decisions, and how they can increase our chances of reaching our goals. In this case, how can we make decisions which will impact our journey in crypto positively? We can make decisions that probabilistically increase our chances of success, and here's how to do it.

The idea of mental models is to find a way to assess how certain things work. Then by realizing the way a certain thing works, it will then help you make better decisions in the aftermath. Basically, by not repeating the same mistakes we can maximise our efficiency.

That means, by being able to understand how a certain model works (or protocol) we can guide our decisions towards being convictions. These decisions more often than not have various commitment lengths, and tradeoffs will have to be made no matter what. These tradeoffs could be small, but might also have massive implications in other situations.

In most cases, we just go with the flow and don't ask ourselves questions as to why — a great example is latching onto narratives without understanding the base catalysts for a certain narrative in crypto. Most people might do a bit of searching, either on CT or on google to understand the 'idea' of a project. However, that's the furthest most people get. This doesn't help with convictions in any way, and will usually — in most cases, lead to failure.

So what can we do instead? We can use already established mental models to simplify the concept of a certain project. By this, I don't mean just knowing that x enables x, but rather identifying the core components of a certain protocol and how they interact. As well as what catalysts are within scope. This should in most situations give us clarity in regards to what conviction is to be made.

The first model we'll be looking at is First Principles. The concept of first principles is to find the most efficient way of solving a problem from scratch. The best way to utilise this in relevance to crypto, is to use it to compare similar protocols — ask yourself, is this way of doing x more efficient than this way of doing x. If it is, then that should certainly light something in your brain, and add to your conviction. As to figuring out the most efficient way, you can ask yourself certain questions.

What systems support the protocol? How do they function?
Are previous versions of this protocol more efficient? Can we innovate on it?
If we can innovate, what could be done differently?
Are there competitors fitting these parameters better?

If you find yourself realizing that this version is more efficient than previous ones, then there's already clear value found. However, if you find yourself realizing that previous versions fit the parameters better. Then, more often than not, you should probably move on. Unless there are some other catalysts in favour of that version.

The second model that can be very useful is One Level Higher. The concept of this model is to figure out if a certain protocol is changing the entire underlying concept more efficiently, or if it's just changing one cog in the system. For example, is the protocol changing minor points after forking an already existing protocol, or are they changing the entire underlying layer — in a more efficient way? In a lot of situations, redoing something in a trivially different way doesn't add much value. However, if you innovate on something in an entirely different way, there's a higher chance of creating something original, that can add value. What you're identifying here is basically the level at which a certain project is optimising.

Is it also vital to continue to look at the protocols in which you've gained conviction. Here you should reassess over time which catalysts are coming, what narratives are being played out, and how to put yourself in the best possible position to deal with them. If you realize that there's a problem somewhere, you should try to address it. Thus if you realize that your original conviction was wrong, then there's no harm done in changing it. This will eventually lead to what we call Regret Minimization. Basically, you should always pick the long term decision that you'll regret the least in your current state of mind. Therefore by continuously reassessing your decisions, you'll be able to identify the best ones. Hence by choosing the decisions that come with the least amount of regret in your current situation, you'll be able to live a life with less regret.

Let's pivot into something a bit more personal, the way I like to work, and whether or not I could optimize this. I've had this conversation before when people have asked me to describe the way I work. The most suitable way to describe my workflow is by using the concept of Flow State. The best way to describe flow state is to see it as being "in the zone". It's basically a state in which you're totally immersed in a certain thing with the entirety of your focus. This is usually seen as immensely positive, and I tend to agree with this, but let's look at the other side of the coin. Are there negatives with this way of working? By going into a flow state I tend to bury my thoughts in certain information, and sometimes shut off for other, newer information. This can, and has, led to me sometimes not noticing something that I probably should have. I think something I could do a lot better, is to ingest new information by taking advice continuously on things that I'm working on. That means prioritising certain avenues while realizing that I need to allocate time to update myself on what's happening in the now.

I think a good way of doing better at this, is to look at the Pareto Principle. The idea of the Pareto principle is that the majority of your results come from a minority of causes. Basically, the idea is that the vital few will change the outcome greater than the majority. So spending a short amount of time to better yourself in certain areas, gaining updated information by researching something new, or even acquiring the advice of others — can ultimately lead to a better outcome.

Lastly, let's take a look at Bayesian learning theory, especially Bayesian inference. The idea of this theory is to update the probability for something, as you garner more available information or evidence in support of it. In crypto, we can use this to revise our convictions, as newer more up-to-date information becomes available. If you realize that a decision you took in the past was wrong, after you gained new information, then you have the ability to reassess and change your conviction to one that fits your updated state of mind. This will eventually lead to the probability of a certain conviction playing out being much higher (Bayesian Probability).

This was obviously quite a bit different from what I've written in the past, but hopefully, you found some value in the words within this article.

We should be returning to regular programming later in the week, as I continue working on pieces on Juno and LayerZero.

The Depths

Discussion about this post