16 Feb | 2020
Daviano

Expected Value

How it works

So, what is Expected Value?

If your read my Trading risk Management blog post you must have noticed that I mentioned the term Expected Value several times, and like I said, I’d need to write a whole new (and non-esoteric) post just to explain what EV is in a simple and practical way.

So here it is, and here I go...

One of the existing misconceptions that affects not only to the finance but many other industries, is that in order to succeed you need to be a genius, or that you need to understand complex terminology and abstract concepts, and to me this is just one of the many strategies that “professional” fund, money or "wealth" managers (a.k.a. Wallstreet Boys) use to generate an edge over retail or amateur #traders, and I’ll talk about this specific phenomenon later, but as for now, let me ask you something that you should probably know… see the redundancy?

What is a probability?

You might think that in order to understand #probabilities you’d need to have a degree on #statistics or some similar mathematical thing, but this is not at all true. To understand probabilities you just need to exist and maybe have a little bit of imagination and common sense, however (and somehow contradicting myself [which I'll probably do many times in the future]), there are indeed some terms that will help you understand EV way better, so I’ll define them in my own words at the end of this post, and even if at first they might sound confusing, they’re really not.

So let’s get down to the neety greedy (I always wanted to say this).

Expected Value is nothing more than the average outcome if we run an experiment many times. In simpler words, EV is the outcome we expect to see the most if we replicate the same event over and over again. For example, let’s say you flip a coin, this is a classic example. How likely will the coin be heads and how likely will the coin be tails? And if you’re thinking 50/50 (which you probably are), let me shock you with a cool twist... Even though for practical sake we will indeed use a 50/50 chance from here on, I couldn't resist myself to mention that there's actually a small probability for a coin to land on its own edge and the actual numbers vary depending on the coin, it's size, the landing texture, and even temperature, but according to a quick and lazy research I just made on the internet, it seems like there's a 49.99995% probability for a coin to land heads and another 49.99995% for it to land tails, being the remaining 0.0001% of probable outcome a "landing on its own edge" scenario, which would be really cool to watch, but returning to my simplistic me, and sticking to the theoretical 50/50 scenario, what does this means for us?

And please don't underestimate the power of what I am about to share in the following video, because as simple as this concept seems to be by itself, the sophistication of it's implementation and application, might vary on different levels of complexity and abstraction, and in some cases this simple concept somehow ends being the framework for some of the most complex models on fields like theoretical physics, chemistry, mathematics, economics and behavioral sciences, but for us is no more than the understanding of how to become profitable and consistent in trading.

 

 

Probability: The likelihood of an event happening.

Event: A specific outcome, or the combination of several outcomes.

Preferred Outcome: Often called “Favorable Outcome”, it’s just a way to refer to the event that we “want” to see happening, or just the event which likelihood of happening we want to measure.

Sample Space: The sample space is just the sum of all the possible outcomes or events happening in a specific experiment.

Expected Value: The average outcome for a specific experiment if we run it several times.

Experiment: The process of repeating trial over trial in order to see or expect to see the possible outcomes of the experiment in place happening.

Trial: A trial (or iteration) is just a way to refer to a single test of what you want to measure in a specific experiment.

Share This Post