How the algorithm works, what it optimises for

The Crypto Autopilot looks at several months of past price data to inform its decisions. It takes into account historical rise and falls, but over several months, not over days. It doesn't have "momentum" detectors (and when we tried adding some it generally performed worse).

Its objective is to build “stable growth” over the medium to long term (2.5-5 years). Stable here means “without drawdowns” (a drawdown is a top-to-bottom decline over a specific period). It optimises for a 4th generation risk measure - iVaR - which can roughly be described as the time it takes to recover any losses. You might also have seen a simplified version of it be called “the Ulcer Index”.

It’s no magician, though, and can’t go against the market, but that’s what it optimizes for.

Its ability to recover faster from losses is particularly evident in times of trouble. If we zoom in on the period from 1 Feb 2020 to 30 April 2020 we can see how it behaved during the Covid-19 related crash

First, the Autopilot “dropped” less than either the C20 or Bitcoin. Then, by April 8 the Autopilot had recovered to its value from 1 February, when C20 & Bitcoins were still suffering 20% losses (and Bitcoin would need until May 6 to recover, C20 until July 27!).

Why optimize for minimum risks and not maximum returns?

Funnily enough, optimizing to reduce risks actually also helps improve the returns. We don’t want to bore everyone with the details and the math, but if you are interested, you’ll find more academic information here and in the product sheet of our partners.

In their product sheet, our partners put it plainly: “The demonstration of such significantly superior performance from our portfolio construction framework compared to the EuroStoxx 50 is merely a fortunate consequence of our methodology, because our framework does not explicitly optimise for outperformance.” But in academic literature downside volatility-managed portfolios & and a high degree of diversification (which the Autopilot also seek, and you can find info about the D-Ratio here) are linked to outperformance.

And when the algorithm is back-tested on the EUROSTOXX 50 over 16 years, its outperformance is hard to miss.

As you can see on the chart, and can read in our article "Time in the Autopilot and returns", this does however take a while to be visible and one should not expect immediate outsized gains.

What are the assets it can choose from?

We give the Crypto Autopilot an "investable universe" of Crypto. We currently track about 200 cryptos and have "whitelisted" a little over 50 to date (March 2021).

The criteria for inclusions are the following:

• High market cap: a relative sign of quality
• High liquidity: because we want to be able to sell your position if we need to, and illiquid coins would see too much slippage
• No privacy coins (Dash, Monero): because the regulators really don't like them and there is a serious risk they will be increasingly delisted from exchanges
• No joke coins (Dogecoin): they are undoubtly fun, but we want to invest in projects with a real future
• Mostly no forks (Bitcoin Cash, Ethereum Classic): most of the forks don't appear to really have a future and don't have a lot of developer backing, so they don't make for very good investments.
• No stablecoins (Tether, USDC, EUT, EURS): the idea is to help you access the value of true crypto projects with groundbraking possibilities. Not to have you hold digital cash.
• No commodity-backed coins: the idea is to help you access the value of true crypto projects with groundbraking possibilities. Not to have you hold gold or silver.
• No derivatives or synthetic products: we like the simplicity - and relative safety - of investing directly in the real thing.

This is not investment advice, nor a solicitation. Crypto markets possess a high level of risk, including volatility and regulatory uncertainty. Past performance does not constitute a guarantee of future results in any way. You are solely responsible for doing your own financial, legal, tax, or investment research before taking any actions.