Gate Research Institute: ETH Strong Rebound, Moving Average Breakthrough Strategy Annualized Return Reaches 127%

Introduction

This quantitative biweekly report (from April 25 to May 12) analyzes the market trends of Bitcoin and Ethereum, comprehensively utilizing indicators such as long-short ratio, open interest, and funding rates. The quantitative section discusses the application of the "moving average convergence breakout strategy" in the ETH/USDT market, covering its logical framework and signal determination mechanism. Through systematic parameter optimization and backtesting, the strategy demonstrates robustness in trend identification and risk control, with clear execution discipline, and overall outperforms simple holding of ETH, providing a practical framework for quantitative trading.

Summary

• In the past two weeks, BTC and ETH have risen in sync, with BTC increasing by about 34% and ETH rising by over 60%.

• The long-short ratio of ETH fluctuates significantly, indicating that the rise of ETH is accompanied by strong short-term trading and market speculation, and the bears have not significantly retreated.

• The increase in the BTC contract position amount is relatively slow, while ETH showed a stronger rise at the beginning of May.

• The overall contract market saw a concentration of short positions being liquidated in early May, while long positions faced reverse liquidation on May 12, reflecting increased market divergence under high leverage.

• The quantitative analysis uses the "moving average convergence breakout strategy", achieving a return of up to 127% under optimal parameter selection.

Market Overview

1. Analysis of Price Volatility of Bitcoin and Ethereum

BTC and ETH as a whole have shown a steady upward trend since mid-April and maintained a relatively consistent upward rhythm until early May. During this time, BTC rose from about 78, 000 USDT to nearly 105, 000 USDT, while ETH climbed significantly from about 1, 600 USDT to around 2, 600 USDT. It can be seen that ETH has risen significantly higher than BTC, showing stronger price elasticity, especially in early May, when the two jumped simultaneously, and it may be that with the slowdown of tariff policy, BTC has also come out of a wave of repair. BTC has a higher price, less volatility, and a relatively stable trend; ETH, on the other hand, rose more and reacted faster. Originally, the market lacked bullish expectations for ETH and its performance was relatively lagging behind, but after entering May, as the Pectra upgrade was approaching and the tariff policy was eased, ETH increased in volume. This round of changes reflects the market's renewed focus on the value of ETH allocation in the short term. 【 1 】【 2 】

Image 1: BTC price rises to nearly 105,000 USDT, while ETH surges significantly to around 2,600 USDT, with a more dramatic increase and response speed.

In terms of volatility, both BTC and ETH experienced significant changes in their overall fluctuations from early April to mid-May. In mid-April, BTC's volatility frequently peaked, indicating an active market sentiment and sharp price adjustments; subsequently, towards the end of April and early May, the overall volatility tended to converge, reflecting a brief period of market stability. However, around price surges, ETH's volatility saw several sharp increases, even briefly exceeding BTC, indicating that it experienced stronger short-term fluctuations during the upward trend. Overall, BTC's volatility is relatively more uniform, while ETH's fluctuations are concentrated around several key points in time, especially before and after price breakthroughs, suggesting it is more susceptible to capital-driven influences.

Figure 2: The volatility of BTC is relatively more stable, while the volatility of ETH has experienced several sharp increases.

Overall, ETH has shown greater price increases and more concentrated volatility changes in this market trend, indicating its strong price responsiveness at critical moments; meanwhile, BTC has displayed a relatively stable upward trend with a more dispersed volatility distribution, reflecting its relative stability amid market fluctuations. Although both assets have jumped in price simultaneously, their volatility characteristics and rhythms still exhibit significant differences, highlighting different market traits and dynamic structures. In the short term, attention can be sustained on the capital flow and volatility changes of BTC as an important indicator of market risk appetite.

2. Analysis of Long-Short Ratio (LSR) for Bitcoin and Ethereum Trading Volume

The Long/Short Taker Size Ratio (LSR) is a key indicator that measures the trading volume of long versus short taker orders in the market. It is typically used to assess market sentiment and the strength of trends. When the LSR is greater than 1, it indicates that the volume of active buying (long taker orders) is greater than the volume of active selling (short taker orders), suggesting that the market is more inclined towards going long, with sentiment leaning bullish.

According to Coinglass data, the prices of BTC and ETH have shown a significant upward trend in the past two weeks, but in terms of LSR, the two have shown varying degrees of divergence. BTC's LSR rose slightly in the early stage of the rally, but remained fluctuating around 1 as a whole, and even fell below 1 around May 10, showing that even if the price continues to rise, the short trading volume in the market has also risen simultaneously, reflecting that some investors choose to lay out short orders or hedge operations at a high level, and the market has not formed an obvious unilateral long structure, and there is a certain amount of rising doubt.

In contrast, the long-short ratio of ETH fluctuated more violently. During the price breakout above 2,000 USDT and the rapid rise to 2,600 USDT, its LSR did not rise steadily but instead experienced several violent fluctuations, with a noticeable decline around May 10. This situation indicates that ETH's rise is accompanied by strong short-term trading and market games, with bears not significantly retreating, leading to a divided market sentiment.

Although BTC and ETH prices have surged in sync over the past two weeks, their long-short ratio has not shown a sustained increase. Instead, it reflects a general sense of wait-and-see and hedging in the market at high levels, with investors' sentiments being relatively cautious. The structural support behind the price increase still requires further validation.

Figure 3: The BTC long-short ratio fluctuates downward, indicating a weakening of bullish momentum at high levels.

Figure 4: The ETH long-short ratio fluctuates wildly, and market sentiment shows clear divergence.

3. Contract Position Amount Analysis

According to Coinglass data, the contract positions of BTC and ETH are showing an overall upward trend, reflecting the continuous rise in market trading enthusiasm. The open interest of BTC has gradually increased from about 60 billion USD, and although there have been fluctuations, it has remained at a high level and stabilized after early May. The open interest of ETH has risen from about 18 billion USD to nearly 24 billion USD, following a similar trend to BTC but relatively stable, especially with a noticeable jump in early May, indicating that funds were actively entering the market during this phase.

Overall, the synchronized growth of the contract positions for both has confirmed the price increase, indicating a rise in market participation and leverage usage. However, the capital inflow for BTC has stabilized after the end of April, while ETH saw a stronger rise in early May, suggesting that ETH attracted more interest in contract trading in the short term. [4]

Figure 5: The increase in BTC contract positions is relatively slow, while ETH showed a stronger rise at the beginning of May.

4. Funding Rate

The funding rates for BTC and ETH have generally fluctuated slightly around 0%, showing frequent shifts between positive and negative, indicating a relatively balanced market between bulls and bears. In mid to late April, BTC experienced several instances of negative funding rates, especially around April 20, when it dipped to -0.025%, indicating that bears dominated the market at that time, or there may have been large-scale short hedging activities. ETH had a similar trend during the same period, but the fluctuations were slightly smaller, suggesting that although the market briefly shifted to bearish sentiment, it did not lead to sustained pressure.

With the rise in prices and the increase in contract positions, both funding rates have gradually turned positive and remain between 0% and 0.01%, reflecting that bulls are gradually gaining the upper hand and the market is leaning towards active positions. However, overall, the funding rates have not continued to surge, indicating that while the sentiment for leveraged long positions has strengthened, it is not overheated, and the market sentiment remains in a mildly optimistic phase. 【 5 】【 6 】

Figure 6: The funding rates for BTC and ETH have both gradually turned positive, maintaining between 0% and 0.01%, reflecting that the bulls are gradually gaining the upper hand and the market is leaning towards aggressive positions.

5. Cryptocurrency Contract Liquidation Chart

According to Coinglass data, since mid-April, the liquidation of contracts in the cryptocurrency market has shown a mixed trend of long and short positions, with the amount of short liquidations being particularly significant in early May. Especially on May 8, the amount of short liquidations surged dramatically, reaching a daily scale of $836 million, indicating that the market price rose rapidly at that time, leading to a large number of short positions being forced to close.

On May 12, as market fluctuations intensified, the amount of long position liquidations significantly increased, reaching a daily scale of 476 million USD. This indicates that some investors who chased after high prices were unable to withstand the volatility and faced reverse liquidations. This phenomenon shows that despite an overall bullish trend, there are still significant short-term market fluctuations, with bears and bulls being repeatedly thwarted at critical junctures. The derivatives market remains highly active and concentrated with risks.

This trend corresponds with the aforementioned price increase, the rise in contract holding amounts, and the positive funding rate trend, reflecting a concentrated short squeeze in the market when breaking through key price levels, forming a short-term bullish advantage. However, even in an uptrend, long positions may still face liquidation at local highs, especially during the intensified fluctuations in mid-May, where long positions also face significant risks, indicating that market volatility remains strong, and the characteristics of high leverage and risk hedging in contract trading are still very evident. 【 7 】

Figure 7: On May 8, the amount of short position liquidation surged significantly, reaching as high as 836 million USD in a single day.

Quantitative Analysis - Dense Moving Average Breakthrough Strategy

(Disclaimer: All predictions in this article are based on historical data and market trends, and are for reference only. They should not be considered as investment advice or guarantees of future market trends. Investors should fully consider the risks and make cautious decisions when making related investments.)

1. Strategy Overview

The "Moving Average Convergence Breakout Strategy" is a momentum strategy that combines technical trend analysis. The strategy identifies potential directional volatility in the market by observing the convergence of multiple short to medium-term moving averages (such as the 5-day, 10-day, 20-day, etc.) over a specific period. When multiple moving averages trend together and converge, it usually indicates that the market is in a consolidation phase, waiting for a breakout. At this point, if the price clearly breaks upwards through the moving average region, it is considered a bullish signal; conversely, if the price breaks down through the moving average band, it is seen as a bearish signal.

To enhance the practicality of the strategy and the effectiveness of risk control, this strategy also includes a fixed ratio for profit-taking and stop-loss mechanisms, ensuring timely entry and exit when trends appear, balancing reward and risk control. The overall strategy is suitable for capturing medium to short-term trend markets and possesses a certain degree of discipline and operability.

2. Core Parameter Settings

3. Strategy Logic and Operational Mechanism

Entry Conditions

• Moving Average Density Judgment: Calculate the maximum and minimum differences of the six moving averages SMA 20, SMA 60, SMA 120, EMA 20, EMA 60, and EMA 120 (referred to as moving average distance). When the distance falls below the set threshold (for example, 1.5% of the price), it is considered a dense moving average.

Threshold refers to the critical value, which is the minimum or maximum value at which an effect can occur.

• Price breakout judgment:

￮ When the current price crosses above the highest value of the six moving averages, it is considered a bullish breakout signal, triggering a buy operation.

￮ When the current price breaks below the lowest value of the six moving averages, it is considered a bearish breakout signal, triggering a sell operation.

Entry Conditions: Dynamic Take Profit and Stop Loss Mechanism

• Long position exit:

￮ If the price falls below the lowest moving average at the time of opening, trigger the stop loss;

￮ Or the price rises beyond "the distance between the opening price and the lowest moving average × the profit and loss ratio", triggering take profit.

• Short Position Exit:

￮ If the price rises above the highest moving average at the time of opening, trigger the stop loss;

￮ If the price drops more than "the distance between the opening price and the highest moving average × the profit-loss ratio," the take profit will be triggered.

Practical Example Diagram

• Trade signal triggered

The following chart shows the ETH/USDT 2-hour candlestick chart at the most recent trigger for entry on May 8, 2025. It can be seen that the price broke upwards after being densely packed with six moving averages, meeting the entry conditions set by the strategy. The system executed a buy operation at the current price upon the breakout, successfully capturing the starting point of the subsequent upward trend.

Figure 8: Actual entry position diagram when the strategy conditions for ETH/USDT were triggered (May 8, 2025)

• Trading actions and results

The system automatically exits based on a dynamic profit-taking mechanism once the preset profit-loss ratio is reached, effectively locking in the main segment gains. Although there is still room for further increases that are not included, the overall operation adheres to strategic discipline, demonstrating good risk control and execution stability. In the future, if combined with a trailing stop or trend-following mechanism, it may further extend the profit potential in a strong market.

Figure 9: ETH/USDT Strategy Exit Position Schematic (May 8, 2025)

Through the above practical example, we intuitively present the entry logic and dynamic profit-taking mechanism of the strategy when the conditions of dense moving averages and price breakouts are triggered. The strategy accurately captures the trend initiation points by linking price and moving average structures, and automatically exits during subsequent fluctuations, locking in the main profit range while controlling risk. This case not only verifies the practicality and execution discipline of the strategy but also reflects its stability and risk control capability in real markets, laying the foundation for subsequent parameter optimization and strategy summarization.

4. Practical Application Examples

Parameter Backtesting Settings

To find the best combination of parameters, we conduct a systematic grid search over the following range:

• tp_sl_ratio: 3 to 14 (step size of 1)

• threshold: 1 to 19.9 (step size 0.1)

Taking ETH/USDT as an example, in the backtest data of the 2-hour K-line over the past year, the system tested a total of 23,826 parameter combinations and selected the top five with the best cumulative return performance. Evaluation criteria include annualized return rate, Sharpe ratio, maximum drawdown, and ROMAD (return to maximum drawdown ratio), used to comprehensively measure strategy performance.

Figure 10: Comparison Table of Performance of Five Optimal Strategies

Strategy Logic Description

When the system detects that the distance between the six moving averages converges to within 1.4%, and the price breaks above the upper edge of the moving averages from below, a buy signal is triggered. This structure aims to capture the moment when the price is about to initiate a breakout, entering at the current price, and using the highest moving average at the time of the breakout as a reference benchmark for dynamic profit-taking, enhancing the ability to control returns.

The settings used in this strategy are as follows:

• percentage_threshold = 1.4 (maximum distance limit of six moving averages)

• tp_sl_ratio = 10 (Dynamic Take Profit Margin Setting)

• short_period = 6, long_period = 14 (moving average observation period)

Performance and Results Analysis

The backtesting period is from May 1, 2024, to May 12, 2025. This set of parameters performed exceptionally well during this period, with an annualized return of 127.59%, a maximum drawdown of less than 15%, and a ROMAD of up to 8.61%, indicating that the strategy not only has a stable capital appreciation capability but also effectively compresses downside risk.

As shown in the figure, the strategy has significantly outperformed the ETH Buy and Hold strategy (-46.05%) over the past year, especially during periods of increased market volatility or trend reversals, demonstrating a good take-profit and re-entry mechanism, with drawdown control clearly superior to passive holding.

We also conduct a horizontal comparison of the five sets of best-performing parameters. Currently, the combination achieves the best balance between return and stability, demonstrating strong practical application value. In the future, we can further integrate a dynamic threshold adjustment mechanism or incorporate trading volume and volatility screening logic to enhance adaptability in volatile markets, and expand to multi-currency and multi-period strategy deployment.

Figure 11: Comparison of the cumulative return rates of five optimal parameter strategies and the ETH holding strategy over the past year.

5. Summary of Trading Strategies

The "Converging Moving Average Breakout Strategy" is a trend momentum strategy designed based on the dynamic aggregation state of multiple medium to short-term moving averages. It detects the convergence of moving averages and price breakout behaviors to capture key turning points before the market starts. This strategy integrates price structure judgment and a dynamic profit-taking mechanism, allowing effective participation in medium to short-term trend waves while controlling drawdowns.

In this backtest, we used ETH/USDT as the underlying asset and conducted systematic grid parameter optimization using 2-hour K-line data, covering 23,826 sets of parameter combinations. The backtest period was from May 1, 2024, to May 12, 2025. Ultimately, we selected the top five parameter sets with the best performance in terms of return and risk control, and analyzed their performance based on annualized return rate, maximum drawdown, Sharpe ratio, and ROMAD. The best strategy combination is: percentage_threshold = 1.4, tp_sl_ratio = 10, with an annualized return rate of up to 127.59%, a maximum drawdown controlled below 15%, and ROMAD reaching 8.61%. The risk-adjusted performance far exceeds the ETH Buy and Hold benchmark (which was -46.05% during the same period).

From the parameter distribution observation, the best performance is often concentrated in the range of low threshold values and medium-high tp_sl_ratio, indicating that detecting dense moving average structures during the early stage of market brewing, along with moderately relaxing the profit-taking space, helps to capture the complete wave market. In contrast, when the threshold is set too high or the profit-taking ratio is too low, the strategy is more likely to fall into the problems of frequent entries and exits and premature exits, thereby lowering the overall return rate.

Overall, this strategy demonstrates high reward and risk control efficiency in the mid-term volatility structure of ETH. The strategy logic is stable and has parameter adaptability, showing significant practical potential. Based on the characteristics of this backtest parameter distribution, combinations with a threshold between 1.3 and 1.5 and a tp_sl_ratio falling within the range of 9 to 11 exhibit more stable performance in terms of reward and risk control across various performance indicators, reflecting the strategy's strong ability to capture initial momentum in trends and sustain segment profits. Additionally, the integration of volume screening and oscillation filtering mechanisms is expected to further enhance the strategy's adaptability and robustness under different market conditions, expanding its cross-market deployment potential.

Summary

From April 25 to May 12, the cryptocurrency market exhibited a structural characteristic of "strong price increase, while sentiment remains cautious." BTC and ETH rose simultaneously, with ETH experiencing a greater increase and more volatility. The long-short ratio and funding rates did not show significant bullish bias, indicating limited market willingness to chase prices. Open interest continued to rise, with shorts concentrated on liquidating at the beginning of May, followed by longs facing reverse liquidation on May 12, reflecting increased divergence in the market under high leverage. Overall, while prices strengthened, market sentiment and capital momentum have not aligned, making risk control and timing key to operations.

The quantitative analysis employs a "moving average dense breakthrough strategy" for systematic parameter optimization and performance evaluation. In the 2-hour level data of ETH/USDT, the strategy's annualized return reaches 127.59%, far superior to the -46.05% of the ETH Buy and Hold strategy during the same period. This strategy demonstrates good trend-following ability and drawdown control through momentum structure and trend filtering. However, in actual operation, it may still be affected by market volatility, extreme conditions, or signal failure. It is recommended to combine other quantitative factors and risk control mechanisms to enhance the strategy's stability and adaptability, while making rational judgments and responding cautiously.

Reference Materials:

1. Gate,

2.Gate,

3.Coinglass,

4.Coinglass,

5.Gate,

6.Gate,

7.Coinglass,

8.Glassnode, f 4-41 fe-5606-798 a 2 f 3001 3a? s= 1679144783&u= 1742303183

Gate Research Institute is a comprehensive blockchain and cryptocurrency research platform that provides readers with in-depth content, including technical analysis, hot insights, market reviews, industry research, trend forecasts, and macroeconomic policy analysis.

Disclaimer

Investing in the cryptocurrency market involves high risks, and it is recommended that users conduct independent research and fully understand the nature of the assets and products being purchased before making any investment decisions. Gate.io is not responsible for any losses or damages caused by such investment decisions.