Earlier quoted context omitted.
That volatility is perfect for a mean reversion strategy.
I've been testing this out over the past 6 days with $500 to see how it would actually work while keeping my current amount in. I've been working a bit more on the micro level, but it still is pretty consistent in its volatility (lol). So far I have a 4.6% total return on trying to capitalize on the volatility. That's much less than if I had just held at the original buy. I'm going to keep doing it though to see how…
> So far I have a 4.6% total return on trying to capitalize on the volatility. That's much less than if I had just held at the original buy.
If you compare the returns between those two strategies (buy and hold vs mean reversion), make sure you include comparisons of their beta profiles. Their risk measures are going to be very different.
> When you hold yourself to rules, it also decreases maximum losses too.
Speaking of rules, do you have a maximum tolerable drawdown for the strategy, or a number of consecutive losses at which you stop loss or retire the strategy? In order to add more rigor to your work (and so you know there is an element of empirical strategy here instead of just luck), you should conceive a set of priors for the strategy that allow you to set a hypothetical win rate. If you deviate too far from the win rate, or too far from a drawdown as mandated by your risk management rules, you should shut down the algorithm pending a review of its inputs and retire it if it's no longer working.
> Right now if the volatility and my returns hold for even just two years, it could be going from $500 to over 100K.
There are a few hypotheses implicit to your thought here:
1. Market volatility for the target cryptocurrency will remain functionally stable for the next two years,
2. Your strategy will remain functionally stable for the next two years,
3. There is sufficient liquidity to allow you to continually compound your trading strategy's assets with its returns for the next two years, from an initial outlay of $500 to over $100,000, without hitting capital constraints along the way.
Those are all testable hypotheses (which, technically, you're involved in doing), and I can't tell you if they're realistic. I wouldn't count on all three of them being correct though.