Testing an Investment Theory

In a simulation of course

Dec 24, 2021

• Home
• Blog
• Projects
• Cookbook
• RSS Feed
• Donate
• Other
  Library Links Webring

A regret that I have was not cashing out my LBC from Odysee several months ago when the LBC that I had was worth somewhere between $150 and $200 (compared to the $10 it is worth now). This regret is not all that big since I didn't invest in LBC using money rather I accumulated it through the use of a service that I once enjoyed* so even just ten bucks is ten bucks more than I put into that.

But of course I recognize that similar things happen to people that actually have put money into one system or another and do loose that money when they miss an opportunity like I did. Recently I've been enjoying listening to Joel Salatin's podcast, Beyond Labels, and at least once he's shared the story of his father who went into the stock market when trading first made easier for the general public who one afternoon received the call from his stock guy who said that he had sold his stuff that morning he would have made a $10,000 profit but by that afternoon things had changed and at that point he had lost $15,000. From that point on the Salatins didn't invest money into anything they didn't understand which is probably wise for most people.

But what if there was a way to profit form investing in cryptocurrencies or stocks without the level of risk normally associated with it? The other day I came up with an investment strategy that had the goal of not missing an opportunity for profit, then curious about how it would work I designed a simulation that would put my strategy to the test.

The Strategy

A strategy where the goal is to not miss out on an opportunity for profit is one that must take advantage of all opportunities for profit so my strategy does exactly that. The most simple way to describe it would be to say that an investor using this strategy would sell off excess shares each time his investment's worth became larger than its original worth. For example, if the initial investment was $100 and its value rose to $101 then the investor would sell off the extra $1 worth of his investment leaving the original, but now smaller, $100 worth left, if the value of the investment dropped then the investor would simply wait for the value to grow again and withdraw later.

We'll see how this strategy does.

The Test

Of course to test this I could have written a program to monitor the price of something then thrown some money at it so it could play with it but I don't have that kind of money or patience and it also turns out that letting a computer make a bunch of trades involves jumping through a few hoops, so instead I reminded myself how to use pandas then spent an hour writing a program that would simulate investing using that strategy on historical data.

The data I used was bitcoin prices from around March 2020 through March 2021. Why bitcoin? That's what I felt like using. Why March 2020 thru March 2021? I wanted a year of recent data and March 2021 was when the dataset I used ended (for those of you reading this in the future after that dataset is updated again I used version 7). Why can't I say the exact dates I tested? I didn't care enough to convert Unix time to human time.

The dataset recorded the price of bitcoin every minute, so my simulation analyzes the price at each minute and will simulate selling small amounts of bitcoin each minute that the price is right. The simulation started out with $100 worth of bitcoin (however much that happened to be at the chosen start time) and determines what kind of profit would be seen if the strategy were carried out, it also reports what the profit would have been had the investor simply let the money sit then withdrew it at the time that the simulation ended. I ran two main tests, one where the strategy was tested over multiple short intervals of time (1 month) and another where it tracks how well it would do over longer periods of time (1 month to 1 year).

Test Results

One Month Simulations

If I were publishing this somewhere important then I would have put it all into a nice easy to read table, but this is my website and I can put as much effort as I want into things here.

Each of the months in these results represents an individual month where the strategy was carried out independent of each other. "BTC sold for" represents the amount of money that the simulated investor sold the bitcoin they sold for. "Sold BTC value" is the value that the bitcoin the investor sold is worth at the end of the simulated period. "Initial investment value" represents what the $100 worth of bitcoin from the beginning of the simulation is worth at the end of the simulation, while "Value left invested" represents the value of bitcoin left unsold at the end of the simulated period. "Profit from new method" represents the profit that the simulated investor made from the investment strategy I came up with while "Profit from traditional method" represents the profit the investor would have had had he simply left the investment alone then withdrew everything at the end of the simulated period.

As we see in the above results, my strategy isn't terrible over short periods of time like these. The traditional way of investing is still almost always better but the results are still comparable so it isn't too bad. Interestingly the two months where there was a loss rather than a gain my method would have saved you more money than the traditional one would have.

Long term results

In these results each month builds after the last, ultimately it is one test simulated over a one year period that we pause to analyze each month.

In this test we see that my strategy is terrible for investing long-term. Yes the investor did still profit using my method, but by the end of the year that profit was around a quarter of what it could have been using a traditional investment method and using my method would clearly be a mistake.

Conclusion

So the question to ask now is what is my strategy good for? The short and simple answer is nothing, and that shouldn't surprise anyone, I have no experience in investing, no education in finance and spent very little time thinking this through so the odds of me coming up with something that would beat what the people who have made their living off of this do are microscopic.

But there is one scenario where my strategy has an advantage and it is exactly the scenario it was designed for since it is what happened to me with LBC. Look back up at month twelve of the last simulation, imagine you're the investor (using a traditional method) looking at your $100 investment that has turned into over $900, you'd be pretty happy with that but you'd also think to yourself that if it has gotten that high surely it will get higher so you decide to leave your money in a little longer and plan to take it out once it hits $1,000. What if that day never comes? What if the value of your investment drops to virtually zero? You would have lost your $800 profit as well as your $100 investment. Had you used the strategy I came up with instead you'd still have $200 to show for it all after the price drop, it won't make you rich, but it will protect you from losing out due to your own greed.

You could say that my strategy would be great for people who are afraid of sudden price drops like that but really those people shouldn't be investing at all. Single stock (or in this case coin) investments are virtually gambling, in fact your money would be safer spent at a casino since there you are at least guaranteed to get some sort of a return. Why would any sane person invest in something they expect to lose value?

If you want to gamble then go all in with a full understanding of what you may loose. If you want a more stable investment then diversify your assets, do what the wealthy do to stay wealthy, there is no need to reinvent the wheel.

* I no longer endorse Odysee/LBRY for the same reason that Bryan Lunduke condemned them back in August.