**5. Conclusions**

As a means of extending previous literature, this paper has analyzed the profitability of a significant number of SMA TTRs on a wider range of energy markets and a period that includes the ongoing COVID-19 pandemic.

Using daily data for Brent and WTI spot crude oil prices and for the energy fund XLE and splitting the data into a pre-pandemic window (1999–2019) and a "pandemic period" (January–March 2021), we employ 14,100 SMA crossover TTRs for the longer pre-pandemic period and 1575 SMA crossover TTRs for the shorter pandemic interval.

To the best of our knowledge, this is the first research attempt to investigate the effectiveness of technical indicators on both main crude oil markets, as well as on a relevant energy exchange traded fund, in comparative perspective between the pre-COVID-19 pandemic and the pandemic period.

Overall, we find that technical trading rules can achieve high abnormal returns for all three energy markets over the COVID-19 pandemic period (annualized excess returns over the BH strategy of approximately 20% for WTI, 284% for Brent, and 83% for XLE, without including transaction costs), and only for the Brent market, some small abnormal annualized excess returns (of about 2%) over the 21-years of pre-pandemic period.

However, these excess returns encountered over the pandemic period are not strong on the WTI and XLE markets when their significance is tested by the standard Brock et al. [25] bootstrapping methodology with 1000 iterations, while for Brent market, excess returns gained by TTRs hold strong in all subperiods against the standard bootstrapping methodology. Over the pandemic period, excess returns achieved by TTRs on the Brent market are still high in magnitude and remain statistically significant after transaction costs are included in estimations. Over the pre-pandemic period, the small excess returns achieved by some technical rules on the Brent market are eroded by transaction costs and thus have no economic value. Similar to Taylor [36], our findings could thus reflect a relationship between technical rules' performance and market conditions.

Nonetheless, the abnormal return achieved by the best-performing TTR on the Brent market over the 1 and <sup>1</sup> <sup>4</sup> years of pandemic period no longer holds strong against White's [30] Reality Check test. Thus, we find that SMA TTRs are not consistently profitable in the three energy markets once the data-snooping bias is accounted for.

However, while our results allow us to confidently conclude on the weak form efficiency of the WTI crude oil and the XLE fund markets throughout the 1999–2021 period relative to the universe of TTRs that we apply, and also to sustain the conclusion that TTRs do not add value on the Brent market beyond what may be expected by chance over the pre-pandemic 1999–2019 period, we refrain to also attest the weak-form efficiency of the Brent market over the COVID-19 pandemic. We feel that the performance of TTRs on the Brent market during the pandemic period needs further investigation, as most technical trading strategies achieve high excess returns over the benchmark buy-and-hold strategy, these excess returns hold when their significance is checked by the standard bootstrapping method and are also unaffected by transaction costs. The excess return gained by the optimal TTR only disappears after adjustment for data snooping is accomplished via the employment of White's Reality Check procedure. In this particular situation, the RC test might be too conservative and thus prone to type II errors. By presenting the results of all 1575 tested TTRs on the Brent market over the pandemic interval (both with and without transaction costs), and by showing that an overwhelming number of these strategies have been able to achieve abnormal returns of high magnitude on the Brent market during the COVID-19 pandemic (96.20% of strategies are still substantially over-performing even after adjustment for trading costs is made and thus have economic value), we sustain that survivorship bias is already mitigated. The adjustment made on the bootstrap *p*-values via the RC procedure could thus be too severe.

Consequently, while similar to Psaradellis et al. [23], we did not encounter enough evidence to be able to reject the weak-form efficiency of the three energy markets (Brent crude, WTI and XLE) for the whole 1999–2021 period, it would be hazardous to completely dismiss the above argument in the case of the Brent crude oil market over the COVID-19 pandemic period, when TTRs seem to have benefited from the extreme evolutions that characterized the market. As such, future studies on the Brent market efficiency during crisis are needed to sustain the right policy formulation process.

This paper makes several contributions to the existing literature. The first contribution is to revisit the predictive ability and performance of technical trading rules (TTRs) on some oscillated energy markets. This is the first paper to include both main grades of crude oil (WTI and Brent crude) and a relevant energy fund, XLE, and to assess the trading rules' performance over two different subperiods: pre-COVID-19 (January 1999–December 2019) and COVID-19 (January 2020–March 2021). The second contribution thus consists in presenting proof of TTRs' performance during the historically turbulent COVID-19 pandemic period for crude oil markets. Previous studies mostly refer to the WTI crude oil and cover periods no more recent than year 2019. Other contributions consist in the large universe of tested TTRs (14,100 over the pre-pandemic period, and 1575 over the pandemic period, respectively) and also in estimating the relevancy of results by evaluating the performance of the universe of TTRs while considering both naïve [25] (Brock et al. random bootstrapping method—with 1000 iterations) and more severe methods of accounting for data snooping effects (White's Reality Check procedure—also based on 1000 iterations). In addition, this strategy allows easy comparison with previous findings that have employed one of the two (or both) techniques. A fifth contribution consists in also estimating the economic value of results by allowing for transaction costs, while a final contribution stems for the identification of distinct financialization process between the two main crude oil markets, WTI and Brent. As such, research findings further imply that there is evidence to the existence of a more intense financialization process within the WTI crude oil market, whereas the market for Brent seems to be more impacted by shifts in global supply and demand. This has important implications for both the right choice of oil price forecasting methods by policy issuers and also for identifying the accurate policy measures.

Policy makers must thus consider these market characteristics that the study encounters for effective oil price forecasts and for efficient policy issuance. Moreover, as the financialization of the WTI market and the approaching expiration date for WTI contracts for delivery in May 2020 [37], coupled with insufficient storage capacity have determined its historical and unforeseen plummet on April 2020 into uncharted negative territory [38] regulators of WTI commodity market (i.e., The U.S. Commodities Futures Trading Commission), should also consider tighter measures (i.e., mandatory reporting of high volume trades, short selling restrictions, etc.) to prevent the recurrence of such events.

**Author Contributions:** Conceptualization, C.T.; Data curation, C.T.; Investigation, C.T.; Methodology, A.A.; Software, A.A.; Validation, C.T.; Visualization, C.T.; Writing—original draft, C.T. Both authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data for crude oil prices is publicly available from the Federal Reserve Bank of St. Louis's (FRED) database, which collects data from the U.S. Energy Information Administration, while data for XLE is publicly available from Yahoo! Finance.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

