Gate Research: BTC and ETH Both Pull Back, Trend Strategies Become Main Source of Returns
- Key Insight: In June 2026, the crypto market continued its weak correction, with BTC and ETH both falling over 20%. Significant deleveraging occurred in the futures market. The MA cluster breakout strategy outperformed buy-and-hold, with AAVE USDT achieving a net return of 60.2% as the best practical case.
- Key Elements:
- Market Performance: BTC monthly return was -20.43%, closing at $58,632.4; ETH return was -21.67%, relatively weaker than BTC, as the overall market price center shifted downwards.
- Futures Deleveraging: Open interest in BTC and ETH perpetual contracts decreased by 25.76% and 26.31%, respectively. Long liquidation volumes were significantly higher than short liquidations. Funding rates remained neutral, indicating the decline was driven by spot selling pressure.
- Quantitative Strategy: The MA cluster breakout strategy proved effective in a weak market. It utilizes MA band convergence to wait for breakout signals and employs dynamic take-profit to control drawdowns, outperforming the buy-and-hold strategy overall.
- Best Case Study: AAVE USDT achieved a monthly net return of 60.2% with a maximum drawdown of -12.9%. It executed 4 trades with a 75% win rate. Profits primarily came from directional shifts and dynamic take-profit.
- July Outlook: Continued monitoring of the MA cluster breakout strategy is recommended, with suggestions to add volume confirmation and BTC trend filtering to reduce the risk of false breakouts during counter-trend trading.
Summary
• In June, BTC and ETH fell by 20.43% and 21.67% respectively, with the overall market continuing its weak adjustment, its price center shifting downward, and ETH continuing to underperform BTC.
• The futures market continued to deleverage, with open interest in BTC and ETH perpetual contracts declining by 25.76% and 26.31% respectively. Long liquidations were significantly higher than short liquidations, and funding rates remained generally neutral. The price decline primarily reflected a concurrent weakening of spot selling pressure and risk appetite.
• June's market was suitable for trend-following and breakout confirmation strategies. Parameter backtesting showed that the moving average convergence breakout strategy generally outperformed buy-and-hold, making it more effective for capturing directional moves.
• Based on composite metrics of net return, drawdown, and trade frequency, AAVE USDT was the best live case for June, with a strategy net return of 60.2%, a buy-and-hold return of 3.76%, and a maximum drawdown of 12.9%.
• For July, continued monitoring of the moving average convergence breakout strategy is recommended, incorporating volume confirmation and BTC trend filtering to improve signal quality and reduce the risk of false breakouts from counter-trend trading.
In June 2026, mainstream crypto assets continued their weak, volatile decline, which broadened downwards. BTC opened the month at $73,684.1 and closed at $58,632.4, a monthly return of -20.43%, with a monthly high of $74,090.8, a low of $58,106.9, and a range of 27.51%. ETH recorded a monthly return of -21.67% and a maximum drawdown of -21.88%. Structurally, BTC experienced a rapid decline in early June followed by low-level consolidation, a mid-month rebound that failed to reopen upward space, and a further pullback at the end of the month. ETH showed more pronounced relative weakness, with insufficient price elasticity and significant pressure during liquidity contraction.
On the derivatives side, open interest in major contracts did not form a stable recovery. The notional value of BTC USDT perpetual open interest decreased from $5.19B to $3.85B, a monthly change of -25.76%; ETH's open interest changed by -26.31% month-over-month. In the liquidation structure, long liquidation volumes were significantly higher than short liquidations, indicating that passive deleveraging during the downtrend remained the dominant force. Funding rates were mostly slightly positive or near neutral for most of the period, suggesting the price decline was not driven by extreme short crowding, but rather by the trend effect following the synchronous weakening of spot selling pressure and risk appetite.
From a quantitative strategy perspective, this month was suitable for trend-following and breakout confirmation. This article conducts parameter grid backtesting on 29 valid USDT spot trading pairs on the Gate exchange using 4-hour candlesticks. The screening criteria were: monthly spot trading volume on Gate exceeding $50 million, at least 2 trades executed, maximum strategy drawdown not exceeding 20%, and one-way cost and slippage combined at 0.08%. Based on composite net return, drawdown, and trade frequency, the best live case for June was the AAVE USDT moving average convergence breakout strategy: monthly net return of 60.2%, buy-and-hold return of 3.76%, maximum drawdown of -12.9%, 4 trades executed, win rate of 75%, and profit factor of 9.63.
1. Market Overview
The core characteristics of the June market were a downward shift in the price center, insufficient rebound sustainability, and trading volume concentration converging towards BTC and a few large-cap assets. BTC and ETH remain the most important bellwethers. BTC's monthly range reached 27.51%, with an annualized realized volatility of approximately 43.55%; ETH's monthly range was 34.29%, with an annualized realized volatility of about 65.43%. When major assets experience significant drawdowns simultaneously, cross-coin diversification offers limited protection for net value in the short term, necessitating strict adherence to position direction and exit discipline at the strategy level.
In terms of trading volume, the samples with the highest spot trading volume on Gate in June were concentrated in high-liquidity assets such as BTC, ETH, SOL, XRP, and DOGE. High trading volume has two implications. First, backtesting signals are closer to a real-world executable environment. Second, during periods of amplified volatility within the month, rising volume is often accompanied by both passive stop-losses and active position adjustments, making it easier for trend strategies to capture consecutive price ranges.
2. Structural Observations on BTC and ETH
BTC's price action in June can be divided into three phases. The first phase, from June 1st to June 6th, saw prices rapidly decline from the early-month area, with daily candles weakening consecutively and long liquidations in futures expanding simultaneously. The second phase, from June 7th to June 18th, saw BTC consolidate within a low range, with a local rebound triggering short covering, but prices failed to reclaim early-month highs. The third phase, in late June, saw BTC lose mid-month support again, closing near the low of the month, indicating a continued preference among capital to reduce risk exposure.
ETH's performance was weaker than BTC. ETH's monthly return in June was -21.67%, a relative underperformance of -1.25% against BTC. During weak months, ETH often requires on-chain activity, ecosystem capital inflows, or expanding risk appetite to provide additional support. This month, these factors failed to offset the pressures from macro risks and market deleveraging. Strategically, ETH is suitable as a risk thermometer rather than a standalone offensive asset: when ETH cannot strengthen relative to BTC, the Beta risk for altcoin portfolios needs to be reduced.
The relationship between volume and volatility is also noteworthy. BTC's trading volume notably increased during the initial decline and the late-month pullback, suggesting the price decline was not merely a low-liquidity slide but accompanied by genuine turnover. If BTC subsequently enters a low-volatility consolidation phase, the moving average convergence strategy will wait for the moving average band to contract before judging the breakout direction; if prices continue along a downtrend channel, short-cycle trend models may still outperform mean-reversion strategies.
3. Futures Market: Open Interest, Liquidations, and Funding Rates
The signals from futures data are relatively consistent. It represents passive risk reduction following a decline. Total long liquidations for BTC amounted to $329.4M, while total short liquidations were $144.9M. For ETH, total long liquidations were $314.8M, and total short liquidations were $193.4M. The high proportion of long liquidations implies that leveraged longs were forced to exit during the price decline, which also transmitted sentiment pressure to the spot market.
Funding rates did not show extreme negative values, indicating the market was not overheated in a one-sided short direction. For most of the time, funding rates were close to neutral or slightly positive, suggesting that some capital attempted to buy the dip or maintain long positions even as prices weakened. If funding rates were to turn significantly negative while prices stopped making new lows, that would be closer to conditions for a short-term rebound. Such a strong reflexive structure did not form this month.
A ratio of long-to-short accounts above 1 is not necessarily bullish. In weak market conditions, an increase in this ratio can sometimes result from retail traders going against the trend. Without expanding Open Interest and price appreciation, this can actually become a source of future liquidation pressure. The long-to-short account ratio for BTC and DOGE was high on certain trading days, yet prices failed to recover sustainably. Such divergences need to be incorporated into risk control.
4. Quantitative Analysis: Moving Average Convergence Breakout Strategy
4.1 Strategy Logic
This report adopts the core concept of the moving average convergence breakout strategy. When multiple short-to-medium-term moving averages gradually converge, prices are in a compressed state before a directional decision. When prices break upwards above the upper band of the moving averages, it suggests bulls are regaining control. When prices break downwards below the lower band, it indicates a higher probability of the downtrend continuing. This strategy does not predict turning points but waits for price to provide a direction after the moving average band converges.
This article uses six moving averages to form the moving average band, consisting of three pairs of SMAs and EMAs. The parameter grid includes period groups (6,18,54), (8,24,72), (12,36,108), and (20,60,120), thresholds of 1.2%, 1.8%, 2.2%, 3%, and 4%, and dynamic take-profit multiples of 3, 4, 6, and 8. Using 4-hour candlesticks, data from May 1st to May 31st is used for indicator warm-up, and performance is calculated from June 1st to June 30th.
Entry rules are as follows:
• Moving Average Band Width = (Highest value of six MAs - Lowest value of six MAs) / Closing Price;
• When the MA Band Width is below the threshold, MAs are considered to be converging;
• When the closing price breaks above the upper band of the MA band from below, go long at the open of the next 4H candle;
• When the closing price breaks below the lower band of the MA band from above, go short at the open of the next 4H candle;
• Stop loss: for long positions, if price breaks below the lower band; for short positions, if price breaks above the upper band;
• Take profit: when profit reaches ("MA Band Width at entry" x "Take Profit Multiple"), close the position at the open of the next 4H candle;
• Any open positions at month-end are forcibly closed at the closing price of the last 4H candle.
The backtesting cost assumption is a 0.08% deduction for each position change, covering trading costs and slippage. This assumption does not represent Gate's actual fee structure and is only used for uniform comparison across different trading pairs and parameter combinations. The strategy does not use leverage, and capital utilization is calculated at 100%. Buy-and-hold return is calculated using the opening price of the first daily candle in June to the closing price of the last daily candle for the same trading pair.
4.2 Sample and Screening
The candidate pool includes 29 valid USDT trading pairs on Gate: BTC, ETH, SOL, XRP, DOGE, BNB, ADA, TRX, LINK, AVAX, BCH, LTC, DOT, NEAR, UNI, AAVE, ICP, ETC, ATOM, FIL, OP, ARB, SUI, WLD, INJ, PEPE, SHIB, ONDO, and HBAR.
To avoid a single coincidental signal becoming the best sample, this article restricts the live case criteria to: monthly spot trading volume on Gate exceeding $50 million, at least 2 trades executed in June, maximum strategy drawdown not exceeding 20%, and position exposure not exceeding 95%. The goal of this rule is not to pursue theoretically maximum returns, but to find a strategy combination executable in a live June environment.
4.3 Best Live Case for June: AAVE USDT
According to the screening rules above, the best case for June is AAVE USDT. This pair had a monthly spot trading volume of $108.2M, a monthly buy-and-hold return of 3.76%, a monthly range of 72.28%, and a maximum drawdown of -24.02%. The optimal strategy parameters were: MA period group (8, 24, 72), MA convergence threshold of 4%, and dynamic take-profit multiple of 8.
Backtesting results show that the equity curve for AAVE USDT exhibited a step-like change in June. The strategy did not predict the direction at the start of the month but waited for breakout signals after the MA band converged. This characteristic allowed it to avoid some ineffective choppy moves and retain positions when prices showed a consecutive direction. Compared to buy-and-hold, the strategy's return differential was 56.44%, with maximum drawdown controlled at -12.9%, indicating that the month's returns mainly came from directional shifts and dynamic profit-taking. This sample is not just a replay of spot prices but also provides the foundational conditions for expressing long/short direction via perpetual contracts.
Looking at the trade details, the strategy performed best during phases when prices rapidly diverged from the MA band. Short signals contributed more during the declining month, while long signals functioned more as rebound confirmations. If only spot long positions were allowed, the strategy's returns would have been significantly lower this month. If using perpetual contracts for execution, additional attention is needed for funding rates, liquidation prices, and position limits.
4.4 Sources of Strategy Returns
The moving average convergence breakout strategy was effective this month, primarily due to three types of market structures.
First, prices transitioned multiple times from narrow-range consolidation to directional expansion. The MA convergence condition divides the market into "waiting" and "executing" states, reducing frequent trading during erratic fluctuations. The strategy only assumes directional risk when prices break away from the MA band.
Second, in a weak market, the downward phases were more continuous. Many high-beta trading pairs in June did not immediately recover after a single day of decline but instead experienced consecutive down moves over several 4H candles. Trend-following strategies have a higher probability of positive expectancy in such environments compared to mean-reversion.
Third, dynamic take-profit reduced the risk of giving back gains. Fixed take-profit can lead to premature exit during expanding volatility, while pure MA-based stop-losses might return realized profits to the market. The strategy in this article uses the "MA Band Width at entry x Multiple" method, making the take-profit target dynamic based on the degree of compression at entry. The tighter the MA convergence, the smaller the initial take-profit distance; the wider the MA band, the more room the strategy allows for a trend.
The weaknesses of the strategy are also clear. MA confirmation is inherently lagging and cannot capture the very beginning of a trend. When prices rapidly reverse, short positions may stop out near the upper MA band. If the market enters a wide, directionless range, the MA band will repeatedly converge and expand, and trading costs will erode returns. Therefore, this strategy is suitable as a trend enhancement module, not as a standalone all-weather allocation.
5. Portfolio Perspective: Combining Trend Enhancement with Neutral Strategies
The samples from June show that trend strategies can play both a defensive and offensive role during down months. Short signals can hedge spot Beta, and long signals can capture low-level rebounds. However, their return distribution is not smooth. If used for portfolio management, the moving average convergence breakout strategy is better suited as an enhancement module, paired with low-correlation strategies.
One feasible portfolio framework is as follows:
• Core holdings use BTC, ETH, or stablecoin yield strategies as a low-turnover base layer.
• The trend enhancement module activates only after MA convergence and a breakout occurs; otherwise, it remains in cash.
• Risk budget for any single trading pair is limited to 10%-15% of total portfolio equity.
• Set lower per-trade loss limits for high-beta altcoins.
• If both BTC and ETH break below their daily short-to-medium-term MAs, reduce the weight of long signals.
• When funding rates are continuously and significantly positive while prices fail to make new highs, avoid chasing longs.
• When funding rates turn negative, prices stop making new lows, and OI begins to stabilize and rise, then increase the weight of rebound signals.
The key point of this framework is to place strategy signals within a risk budget, rather than directly extrapolating a single backtest result. The best June case is representative, but it does not mean the same returns can be replicated in July. The vitality of a trend strategy comes from discipline: not trading when there is no convergence breakout, exiting when stop-loss conditions are triggered, and taking profits when the dynamic take-profit target is reached.
6. Risk Warnings and Subsequent Observations
Going forward, three types of indicators need close monitoring.
