The Laissez Faire Algorithm:
The algorithm’s performance at the time of writing stands 131% YTD, a return that has been both exciting and unprecedented. We are currently trending above the five-year expected average return of 121%. As of 2/2/2020, Laissez-Faire now has a 1 year Sharpe ratio of 2.06! A Sharpe above 2.0 is generally considered phenomenal, and for relativity; the S&P500’s Sharpe ratio (holding U.S. stocks) is historically around a Sharpe ratio of 1.0. In layman’s terms, this more or less means it’s twice as ‘worth it’ to have the algorithm trade as it would be to hold the SPX.
Statistically, Laissez Faire is still performing within the first standard deviation of its annual performance, meaning that it is in the normal distribution for where it should be performing, although on the higher end. It’s still evidenced that Laissez Faire’s abnormally large returns are due to Bitcoin having so much surplus alpha to capture. It’s designed to capture that in BTC while limiting the downside, and as long as the asset remains volatile the algorithm should still continue to make impressive returns.
Concerns:
Besides the normal concerns over risk with Laissez-Faire, like exchange risk, abnormally large drawdowns in BTC, and security risk; I’m most concerned over the issue of liquidity in the Bitcoin market. I made adjustments to the algorithm in July 2019 to optimize it for the liquidity problems I was beginning to see, but that has caused it to be in a position less frequently and occasionally miss moves due to not getting filled on orders.
Currently, the net value traded is a bit over $100,000, a sum that should be minuscule in the scheme of a global market. However, at certain moments $100,000 is enough capital to move the market, meaning Laissez-Faire can’t always just slip in and out of positions, but actually affects the price of Bitcoin. Shown below is the CoinbasePro (GDAX) order book that LF executes trades on. If you look on the ask side (red side), there are only 13 Bitcoin ($120,000) for sale within a $10 range. This means that if LF were to place a market buy order, the price of BTC would move nearly $10. Currently, moving the price by a couple dollars is inconsequential for the most part, however, it has negative implications long term. If the algorithm is moving the market by $10 with the capital it’s trading now, what will it do to the market when it’s trading one million?
Solutions:
In the future, if the liquidity problems with the algorithm don’t improve it’s likely more adjustments will have to be made to the algorithm itself, or it will have to be moved to a more liquid exchange.
Strategy Based Solution:
Changing the ordering functions for LF on a fundamental level would likely fix at least part of the problem. One of the ideas I’ve had is removing the lower-timeframe ordering functions and focusing on higher-timeframes and laddering orders on a wider spread. (1-3% rather than .1-.5%). This could possibly make the algorithm less profitable and the average amount of trades would go down, while the amount of time in a position would go up.
Exchange Based Solution:
The ideal option in my mind would be to move to a Bitcoin derivatives exchange, such as www.bitmex.com or www.deribit.com. These exchanges have nearly ten-times the liquidity when compared to GDAX, meaning there is usually around one million on the bid and ask side. The problem with this prospect is that CFDs (contracts for difference) are not legal in the United States. They are legal in most countries in the world, but due to post-2008 regulations and the presence of huge U.S. broker lobbying, they are essentially banned in the U.S. There is possibly a workaround to this by registering Laissez-Faire in another country, however, I believe that takes away from the legitimacy of it and isn’t really likely at this time.
CFD regulation and the global impact – A comprehensive guide
Cryptocurrency Markets:
It’s likely that in the next one to three months there will be a market-shattering moment in Bitcoin. There are several things that evidence this, such as the weekly historical volatility being at the same levels it was in March 2019, October 2016, and early 2015. These areas preceded massive moves to the upside on Bitcoin. The halving is looming around the corner in May 2020 and the market is remaining strong going into it. If demand for Bitcoin remains the same leading into May (it is very likely to), the halving will create a supply shock event that has the potential to create another parabolic move similar to the ones in 2012 and in 2016.
The U.S. Government is auctioning off $37 million dollars of Bitcoin beginning on February 18th, a catalyst that could potentially absorb short term Bitcoin demand but ultimately would be good for adoption over the long term. As Bitcoin becomes more adopted by the hands of the wealthy, its likelihood to be used increases. As written in previous articles, adoption is the main source of demand for Bitcoin, and adoption needs to remain increasing in order to create upside.
Credit Suisse’s 2019 wealth report reminds us of the fact that there are now 46 million millionaires. It’s worth keeping in mind that there will only ever be 21 million Bitcoin, and there are currently 18 million in circulation. As of today, there are not enough Bitcoin’s in circulation for each millionaire in the U.S. to have one whole coin.
A Hedge against global stability?
It’s possible Bitcoin has been adopting the characteristics of an uncertainty-commodity. I first realized Bitcoin’s positive correlation to gold in the middle of 2019, which itself is usually a hedge against market uncertainty or currency uncertainty. Over the last couple of quarters, I’ve noticed a negative correlation between global stability and the price of Bitcoin. For example, Bitcoin had several >5% green days on announcements of tariff increases, trade war extensions, virus breakout announcements, and war threats such as the one in Iran recently. It’s interesting that there has been such obvious price movements on Bitcoin around these events, but is it really a hedge against global crisis? Maybe. Many people believe Bitcoin is a democratically liberating form of money that puts the power of banking back in the hands of the individual. Would Bitcoin be around if world governments collapsed? Maybe. If a government’s currency collapsed, as instanced in Zimbabwe’s hyperinflation, https://en.wikipedia.org/wiki/Hyperinflation_in_Zimbabwe , it would imply Bitcoin becoming infinitely more valuable. As of today, China is injecting $174 Billion into their market to prop it up, and with money printing comes monetary devaluation.
https://www.nytimes.com/reuters/2020/02/03/business/02reuters-china-health-cenbank.html
However, that scenario seems less likely. For now, it seems Bitcoin is only reacting to instability because people believe it’s a hedge.
Here’s Forbe’s getting the memo that Bitcoin is becoming a digital-analog for gold a couple of months after us.
https://www.forbes.com/sites/investor/2019/08/23/bitcoin-is-the-new-gold-trade-war/#18c150da5953
Summary of Accounts:
In March 2020, the algorithm will have been running for a full fiscal year. At that time detailed reports of every member’s account will be sent out and able to review. Around this time I’d love to hold a celebration as well!
3 comments
Nigel
Nice article! Very good info.
Shawn Barr
Great article Alex!! The celebration in March is a great idea!
Jason Johnson
Good ideas, it might also be that a number of other algorithms are triggering ahead of us with a looser criteria exhausting liquidity ahead of our order flow. It might be interesting to know when the algorithm placed unfilled limit orders so we could get a ratio of filled to unfilled orders over the 1 year history and see if the ratio is decreasing. We could be getting gamed by other players on the same index who loosen their criteria to capture liquidity from tighter algorithms.