Improving the strategy and creating an automated portfolio approach

Last days I did an extensive backtesting of the strategy for altcoins position trading. Here my current thoughts on the strategies and on my general portfolio approach.

Improved BMP strategy parameters

I changed the following to improve the strategy performance:

  1. ACC threshold lowered to 7%
  2. TRIG threshold lowered to 70%
  3. Change from TRIG to BTC/USD only on BMP threshold level and not on HMA Alts change.

These changes yield the following performance calculated from Jan 1 2016:

  • BMP Strategy: 942x
  • Alts: 157x
  • BTC: 15x

Notes:

  • Lowering the ACC threshold reduces a too early entries on alts. We need to be careful however on that threshold, because the risk of overfitting on the past is very high. I would say that ACC phase cannot be completely automatic, there is a lot of margin to the single investor to improve the performance of this phase.
  • Lowering the TRIG threshold improve the odds to catch a local alts top.
  • Changing from TRIG to BTC/USD only on BMP avoid to get out too early from alts and losing potential additional profits.

Pure HMA BTC and Alts strategy

In my quest of a simpler and fully automatic trading strategy I tried to backtest simple strategies based only on the HMA55/80 signals. The first results I got were shockingly good.

Backtesting HMA
HMA vs BMP Strategy

I tried two strategies:

HMA BTC  strategy

This strategy looks at daily HMA55 and HMA80 signals compared to the BTC marketcap. When BTC marketcap is higher than HMA80 it switches to a 100% BTC position. In contrary, when BTM marketcap is lower than HMA55 it switches to 100% USD position.

The performance of the strategy is 528x, which is very impressive, taking in account that staying in USD or BTC position is less risky than playing alts.

Moreover using BTC marketcap instead of BTC price takes in account also the inflation, which should be a more reliable trading signal.

HMA Alts strategy

This strategy looks at daily HMA55 and HMA80 signals compared to the total alts marketcap. When the marketcap is higher than HMA80 it switches to a 100% alts position. In contrary, when the marketcap is lower than HMA55 it switches to 100% USD position. The 100% alts position can be created by selecting at least 20 mid-cap/high-cap coins with sound FA background.

The performance of the strategy is a whopping 15’635x, which is incredible!

Stress-testing the strategy

The HMA Alts strategy performance is so good that is literally asking to be cautious. One important point has to be considered, entering and exiting a 20+ coin position is more complicated and prone to noise than trading only BTCUSD. I tried to introduce some noise by delaying the position switching by one day. This simulate a real world situation where you have delay between the signal and the action.

 

Backtesting HMA with delay filter
Effect of a one day delay on performance

This delay significantly reduce the strategies performance. HMA Alts reduces to 961x which is still a bit higher than the BMP strategy. HMA BTC performance is reduced to 76x.

Final thoughts

Another point that has to be considered is that the history in crypto is very short and the HMA 55/80 strategy could be overfitted to the latest years and maybe the performance could be lower in the future.

HMA strategies are faster and trades more often between the positions and are more time sensitive. These strategies are very good candidates for a fully automated approach. I will look in the future to further improve the automation side on these strategies.$

Said that the performances shown are too good to be ignored. In the website and the automatic tweets you’ll find from now on also the HMA strategies phases.

The BMP strategy is slower and less time critical. I believe this strategy is more suitable for trading on micro/low caps coins. These markets cans be very profitable but they are very illiquid. A slower strategy that allows time for accumulation in my opinion is very important for these kind of coins.

Creating an automated crypto portfolio approach

How to put all these possibilities together? What is the more sensible approach? I propose the following portfolio, split between this strategy allocations:

portfolio strategy
Proposed portfolio strategy

10% – USD

This allocation stay all the time in USD, it allows to keep some “dry powder” in case of big/unforeseen opportunities. Moreover it increase protection in case of big market crashes.

20% – HODL Passive Income

During alts market cycles I try to accumulate freeroll bags of passive income coins: PoS, MNs, DEX coins. The rewards from this allocation is sold every month to USD and exit forever the crypto portfolio. The goal of this strategy is to create a Real Live income stream that can be used and/or reinvested in traditional markets.

10% – BTC Bots

This allocation is used by price action BTCUSD bots. 50% of this fund is allocated to a 12h swing trading bot, 50% to a 30min scalper bot. I use a bot because I’m an investor, not a trader. This strategy is still in trial I will write a performance review in the future when I have more data.

20% – HMA BTC Strategy

The BTC HMA strategy is less performant but also safer, because it stays in USD or BTC, which are less risky than alts positions. Therefore it has a place in a diversified approach.

20% – HMA Alts Strategy

More risky but with a much higher potential reward. The alts positions will be equally weighted rebalanced every hour by Shrimpy, this should further increase the performance.

20% – BMP Strategy

This strategy will be dedicated to low/micro caps altcoin plays. It will not be fully automated but it’s the most fun, because it involves FA work, and activity which I enjoy I lot.

Conclusions

As you can see my approach is a work in progress. This is inevitable in crypto, where everything is in motion.

The pure HMA alts strategy looks very promising, I will follow this strategy from now on on my Shrimpy test portfolio.

I will stick however to my BMP strategy for my micro/low caps play, which represents now a considerable portion of my portfolio. After the BMP strategy full cycle end I will try to gradually reach the proposed portfolio strategy allocations.