AQTIS - Breakout (Results and Performance

We are thrilled to unveil the latest iteration of our breakout strategy.

In this presentation, we'll focus on two primary areas:

  • First, the performance of the top 20 cryptocurrencies by market capitalization.
  • Second, the average and median performance when capital is distributed across these top 20 coins.

    • Intro

    In our previous presentations, we detailed our methodology for generating results through backtesting and leveraging data analytics to produce our trading signals.

    We also delved into the concept of breakout and discussed how it has been integrated into our trading system.

    If you haven't had the chance yet, we highly recommend visiting the following links before proceeding further:

    Top 20 and Aggregated Performance

    We will explore the results of executing our breakout system on 4H timeframe.

    We will be using the top 20 coins by market cap, which are: BTC, ETH, BNB, ADA, SOL, XRP, DOT, DOGE, AVAX, LINK, LTC, ATOM, BCH, MATIC, XLM, ETC, VET, TRX, FIL, XTZ.

    We have conducted our backtest with older cryptocurrencies that have at least been on Binance's perpetual markets since at least mid-2020, and we have aggregated similar results for over 70 coins.

    Currently, our system scans for opportunities in more than 170 coins.
    However, we will distribute liquidity among the top 20 at most, which is why we are presenting these results.

    Take a look at the market

    We'll start by showing the returns that would have been generated by holding the current top 20 coins by market cap since January 1, 2020.

    The first subplot displays the performance of the 20 coins, each with its own time series.
    The chart is plotted on a logarithmic scale to facilitate visualization.
    As we can see, there are coins that have increased in value by more than 30x, as well as coins that have decreased in value by more than 90%.

    For this reason, we created a second subplot that shows the mean and median performance.
    The median is a more reliable measure as it helps to mitigate the impact of assets with either extremely good or extremely poor performance.

    Breakout Long - 4H Backtest

    Cumulative Returns

    We have generated the charts of the accumulated returns (Cum Returns) using our Breakout trading system, with a 4-hour timeframe and in a Long-only direction.

    The 'Cum Returns' metric means that we start by assigning a value of '1' to each asset, and go 'all in' with the available capital, without leverage, for each trade executed on that asset.
    For each trade, we have a predefined stop loss ranging between 0.5% and 2%.
    Our risk-reward ratio is always equal to or greater than 3, and we let the profits run until the market changes direction.

    As we can see, there are some assets with exceptionally high returns.
    These standout assets are DOGE, LINK, and ATOM.

    Consequently, in our second subplot, we calculate both the mean and the median to get a more nuanced view of overall performance.
    The mean suggests an average return of 18x on our initial investment.
    The median, on the other hand, yields a 6x return on the initial investment.

    Breakout Long - 4H Backtest

    Drawdowns

    We know that after the bull market, all cryptocurrencies suffered a significant drop in value. Most came close to a 90% loss in value from their all-time highs. With our system, we reduced the drawdowns to an average of approximately 10%.

    We can see that there were coins like SOL, DOGE, or ETC that experienced drawdowns of over 30%.
    However, as we see in the second subplot, the average and median drawdowns of the system are less than 20%.

    Breakout Long - 4H Backtest

    Aggregated Statistics

    	              agg_stats mean	        agg_stats median
Start	              2020-01-01 00:00:00+00:00	2020-01-01 00:00:00+00:00
End	              2023-08-24 04:00:00+00:00	2023-08-24 04:00:00+00:00
Period	              1243 days 22:24:00	1301 days 00:00:00
Start Value	      100.0	                100.0
Min Value	      95.9068370876457	        96.82635342986796
Max Value	      2219.23203447836	        815.3650548344369
End Value	      1889.6178759089005	710.7400381334842
Total Return [%]      1789.6178759089005	610.7400381334842
Benchmark Return [%]  448.3786744605557	        138.02478886827691
Time Exposure [%]     20.644776983567475	20.66330064927295
Max Exposure [%]      100.0	                100.0
Max Drawdown [%]      24.36523720742287	        22.89641884078874
Max Drawdown Duration 341 days 12:24:00	        328 days 12:00:00
Total Orders	      357.85	                365.0
Total Fees Paid	      82.48627525116743	        27.654314740267008
Total Trades	      179.25	                182.5
Win Rate [%]	      25.15695962469869	        26.161157168129314
Best Trade [%]	      64.64451346233884	        45.96807125255834
Worst Trade [%]	      -1.059594060593965	-1.0595940605939642
Avg Winning Trade [%] 9.539252829935021	        8.259492508742015
Avg Losing Trade [%]  -1.0295020393688268	-1.029191792022248
Avg Winning Duration  3 days 12:00:59.850000	3 days 15:47:29.500000
Avg Losing Duration   0 days 19:00:55.750000	0 days 17:47:03
Profit Factor	      2.030383238142521	        1.9232290209280567
Expectancy	      9.848275908459652	        3.3514656460124463
Sharpe Ratio	      1.6872546052242097	1.6264157445813832
Calmar Ratio	      4.216640079685391	        3.7833348715194566
Omega Ratio	      1.3075514864272033	1.3138698732146812
Sortino Ratio	      3.12139851099441	        3.0217661368387256

    Breakout Short - 4H Backtest

    Cumulative Returns

    In the same way that we published the results of the accumulated returns in the long-only strategy, we are going to show the results of the short-only strategy.

    Also shown on a logarithmic scale.

    We can see that in this case there are no assets with extreme results, so the average and median results are very similar.

    Additionally, we can see that our short-only system began to yield better results starting from the summer of 2021.

    Breakout Short - 4H Backtest

    Drawdowns

    The drawdowns in the short-only system are a maximum of 25% for some of our assets. The rest have a drawdown lower than that value.

    We can see that during the bull run from January 2020 to the summer of 2020, the short-only system experienced its largest drawdowns, and then again during the spring of 2021.

    As we see, the average and median drawdown of the system is around 8% when the market moves against the short system.

    Breakout Short - 4H Backtest

    Aggregated Statistics

    	              agg_stats mean	        agg_stats median
Start	              2020-01-02 00:00:00+00:00	2020-01-02 00:00:00+00:00
End	              2023-08-24 04:00:00+00:00	2023-08-24 04:00:00+00:00
Period	              1241 days 03:12:00	1298 days 00:00:00
Start Value	      100.0	                100.0
Min Value	      96.15708902110939	        96.26711596272446
Max Value	      747.160871326915	        683.5534071241005
End Value	      702.600441693444	        617.4638147926273
Total Return [%]      602.600441693444	        517.4638147926273
Benchmark Return [%]  450.81861818823046	138.02478886827691
Time Exposure [%]     20.520292603219993	20.31463794621689
Max Exposure [%]      100.0              	100.0
Max Drawdown [%]      19.41094141928539	        19.82768016882038
Max Drawdown Duration 225 days 17:36:00	        222 days 18:00:00
Total Orders	      355.05	                346.5
Total Fees Paid	      20.690080840191296	18.38492245338495
Total Trades	      177.65	                173.5
Win Rate [%]	      25.87199402113536	        26.41603415559772
Best Trade [%]	      31.15892436780996	        30.3297914515747
Worst Trade [%]	      -1.060406060606087	-1.0604060606060866
Avg Winning Trade [%] 7.835316571577081	        7.367906385265911
Avg Losing Trade [%]  -1.0350319411722653	-1.0361008528284228
Avg Winning Duration  3 days 14:54:25.850000	3 days 08:33:24.500000
Avg Losing Duration   0 days 17:17:17.150000	0 days 15:49:31
Profit Factor	      2.419756493901569	        2.348656597460951
Expectancy	      3.3938355179335984	3.020159570364563
Sharpe Ratio	      1.758282790010045	        1.7701231360198824
Calmar Ratio	      4.100543098306568	        4.053032277693095
Omega Ratio	      1.3027546539015342	1.306675272598731
Sortino Ratio	      3.2391572020460515	3.253921803735735

    Synergy of Long and Short Strategies: Achieving Balanced Portfolio Performance

    Returns Correlation

    We present how both strategies behave in an overlapping manner.

    We can see that the long-only strategy had better performance up until the spring of 2022, whereas since then, the short-only strategy has had better performance.
    In 2023, the long-only strategy had excellent performance, and in the recent weeks, the short-only strategy has regained strength.

    Drawdowns correlation

    A correlation close to 1 would mean that both strategies produce gains and losses at the same time, which is not desirable. A correlation close to -1 would indicate that when one strategy produces drawdowns, the other reduces them, and this would be the ideal case. In our situation, we get a correlation close to 0, indicating that there is no relationship between the drawdowns of the strategies.

    Drawdowns correlation between long and short only strategies:
avg       0.919045
median    0.906873

    Conclusions

    Our Breakout Trading System allows us to capitalize on every sharp market movement.
    We are achieving very good results in both the Long and Short directions for all the analyzed assets.

    The breakout is a robust strategy that provides us with gains and reduces drawdowns and exposure time on both sides of the market.
    Using the statistics of the median values, we would achieve more than 5x on both sides in our portfolio if we had started in January 2020, with a maximum drawdown of 20% and 20% exposure time.

    Obviously, in this backtest, we are using the capital allocated to each asset independently, which is not efficient in terms of optimizing the use of liquidity.
    In an optimized portfolio, we would rebalance the capital allocated to each asset periodically, and we could share capital among multiple assets while adjusting our risk-reward ratio for each trade, which would allow us to increase potential gains while keeping risk adjusted.