Gate Research Institute: BTC price hits a new swing trading high, TSI indicator Annual Percentage Rate reaches as high as 119%

2025-04-25 07:27:59

Introduction

This quantitative biweekly report (from April 10 to April 24) analyzes the market trends of Bitcoin and Ethereum, comprehensively utilizing indicators such as long-short ratio, contract open interest, and funding rate. This article delves into the principles, calculation logic, and application strategies of the true strength index (TSI) in BTC trading. Through detailed parameter optimization and backtesting validation, the results indicate that the optimized TSI indicator excels in capturing market trend reversals and momentum strength, with its backtested return and risk control metrics significantly outperforming a simple buy-and-hold strategy for BTC, providing traders with an effective quantitative trading tool.

Summary

BTC price broke through to 94,000 USDT, while ETH price broke through to around 1,800 USDT, both strengthening.
In the past two weeks, the volatility of ETH has been overall higher than that of BTC, indicating stronger price fluctuations.
The BTC long-short ratio gradually increased after the price broke through the resistance level, and quickly dropped on April 23, as market investors took short-term profits.
BTC's contract open interest has risen by about 28% from its low point, indicating a warming of bullish trading sentiment in the market.
The ETH funding rate has repeatedly fallen into negative territory, reflecting that short funds dominate during certain periods.
The TSI indicator achieved an annualized return of 119.75% under optimal parameter settings.

Market Overview

1. Analysis of Price Volatility of Bitcoin and Ethereum

Over the past two weeks, Bitcoin has mostly consolidated in the range of 81,000 to 85,000 USDT. Buoyed by a weaker dollar and easing tariff conflicts, Bitcoin broke through resistance on April 21 and then continued to strengthen and stand above the 90,000 USDT mark, reaching a phased high around 94,000 USDT on April 22. Bitcoin has risen by about 15% since April 10, with bullish momentum dominating and successfully recovering the losses of February 25. Ethereum's price performance was relatively weak, and it bottomed out several times during this period, but it also quickly rose between April 21 and April 22, breaking through and standing above the previous resistance level of 1,600 USDT, and regaining a position at 1,800 USDT. It's up about 12% cumulatively since April 10.

Figure 1: BTC price breaks through to 94,000 USDT, while ETH price breaks through to around 1,800 USDT, both strengthening.

In the past two weeks, the cryptocurrency market has maintained a consolidation trend for an extended period, with volatility significantly easing compared to early April. From April 10 to the early stages of April 13, Bitcoin's price briefly fell below 80,000 USDT, and the daily volatility of BTC quickly surged, reaching a peak of 0.0243. On the same day, Ethereum's volatility soared to nearly 0.043, reflecting that ETH's price fluctuations are much higher than BTC's, indicating that ETH is more actively participating in trading in the short term and has a stronger price sensitivity.

During the period from April 14 to April 20, the overall market volatility tended to be relatively calm, with volatility gradually decreasing to a lower range. The volatility of BTC and ETH remained around 0.005 to 0.015, indicating that the market overall was in a consolidation phase, and investor sentiment was becoming cautious and watchful.

From April 21 to April 23, during the price breakthroughs of two major cryptocurrencies, including BTC and ETH, market volatility surged sharply, indicating an increased willingness for both bullish and bearish positions, as well as a noticeable rise in trading intention. Among them, ETH volatility remained at a high level, with the highest volatility hitting 0.03 again, surpassing BTC's level during the same period, which suggests a higher enthusiasm for short-term investment participation in Ethereum and a more intense battle between bulls and bears. 【1】【2】

Figure 2: The volatility of ETH changes is overall higher than that of BTC, indicating stronger price volatility.

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

Between April 10 and April 12, BTC's long-to-short ratio climbed rapidly to 1.09, indicating a noticeable increase in traders' long-term momentum. At the same time, from April 13th to April 17th, the long-short ratio fell and moved sideways, reflecting the adjustment of the market, the long-short power tends to be balanced, and the overall market sentiment tends to be cautious. After April 18, the long-short ratio gradually increased again, and hit a phased high of 1.13 on April 21-22, reflecting the further strengthening of investors' long-term expectations. However, the subsequent rapid pullback indicates that market investors have taken profits in the short term, or that some funds have quickly taken profits after experiencing sharp fluctuations, and are wary of the persistence of price trend expansion.

Between April 10 and April 12, the ETH long-short ratio rose rapidly and reached a stage high of 1.06, indicating that the market dominated the bulls during this period and the overall investor sentiment was more optimistic. However, from the 13th to the 16th, the long-short ratio fell significantly, the market sentiment entered a state of phased correction, and the bullish power subsided. From April 17th to April 19th, the long-short ratio climbed sharply to 1.08 again, and the bulls once again occupied the dominant position, and investors entered a state of bullish layout. However, from the 20th to the 21st, it fell back rapidly, reflecting the fierce game of long and short, the overall direction of the market is unclear, and investors' operations tend to be short-term, and the wait-and-see mood is strong. Entering April 22-23, with the significant increase in ETH price, the long-short ratio rose again to 1.07 after a brief pullback, indicating that the market once again had a more positive emotional reaction to the bullish direction after the price breakout.

Overall, there is a strong correlation between the trading volume ratio of the two major cryptocurrencies, BTC and ETH, and their price movements. In the short term, BTC is relatively favored by bulls, with positive market sentiment, while ETH experiences intense bull-bear competition, and the market is more cautious.

Figure 3: The BTC long-short ratio gradually increased after the price broke through the resistance level of 85,000 USDT, and quickly dropped on April 23.

Figure 4: ETH long liquidity remains relatively weak, and short sentiment is stronger compared to BTC.

3. Analysis of Contract Position Amount

According to data from Coinglass, since April 10, the BTC contract open interest rose to a peak of $58.9 billion, before falling back to a low of $52.4 billion. After entering April 21, with the strengthening of BTC prices, the contract open interest also saw a significant breakthrough, reaching a high of $67.1 billion, an increase of about 28% from the low, indicating a rebound in market investor confidence and a more active trading sentiment.

During the same period, the contract holdings of ETH remained relatively stable in the range of 17 billion to 18.5 billion USD, and rose in tandem with the increase in ETH prices, reaching a peak of 21.2 billion USD. This phenomenon indicates an increase in the risk appetite of market investors. [4]

Figure 5: The contract open interest for BTC has increased by about 28% from the low point, indicating a rising bullish trading sentiment in the market.

4. Funding Rate

From April 10 to April 12, the funding rates for BTC and ETH repeatedly fell into negative territory, usually indicating that the market is in a short-term bearish state, with market sentiment leaning towards caution.

From April 13th to April 16th, the funding rates of the two have obviously turned from negative to positive, and have repeatedly reached the high point of the positive rate range, such as BTC reached the highest funding rate of 0.0077% on April 14, and ETH also reached a maximum of 0.0062% in the same period (April 15), which shows that the market bullish sentiment is relatively high during this period, and the bulls have the advantage.

From April 17 to April 20, the funding rates fluctuated frequently, alternating between positive and negative ranges. The difference in funding rates between BTC and ETH has widened, indicating that investors are experiencing short-term divergences, with unclear market trends and intense long-short battles.

The most volatile positive and negative volatility was between April 21 and April 23, when the BTC funding rate fell rapidly to -0.0194%, -0.0186%, and reached a stage low of -0.0271% at 16:00 on April 22, and then gradually recovered. It shows that the BTC market fluctuated violently during this period, accompanied by strong bearish power, the market showed strong short-selling sentiment, and then the market gradually recovered and became rational. During the same period, the ETH funding rate also fluctuated sharply (reaching -0.0083% at 8:00 on April 21 and -0.0122% at 0:00 on April 23), which means that ETH is also facing significant selling pressure during this period.

In the past two weeks, the funding rates for BTC and ETH have shown significant volatility and frequent shifts in market sentiment, indicating intense long-short competition and a lack of sustained clear market consensus. Among them, the fluctuations in market sentiment were most pronounced from April 21 to 23. Such fluctuations in funding rates reflect the current mainstream short-term speculative behavior of market investors; furthermore, the rapid switching between long and short positions indicates that short-term funds are quickly chasing trends during price fluctuations; the funding sentiment lacks stable consensus, which is particularly evident during breakout phases.

Figure 6: The funding rate for ETH has repeatedly fallen into negative territory, reflecting that short positions have dominated during certain periods.

5. Cryptocurrency Contract Liquidation Chart

According to Coinglass data, since April 10, the liquidation amount in the cryptocurrency contract market has significantly narrowed compared to early April. As of the period before April 21, the average daily liquidation amount in the overall contract market was about $216 million, indicating a decrease in market volatility and an increase in investors' risk control awareness.

However, following the strong breakout and rapid rise in cryptocurrency market prices after April 21, the severe fluctuations in market conditions led to a concentration of short positions being liquidated. On April 22, the total amount of short liquidations in the market significantly increased, reaching as high as 517 million USD, and the substantial rise in liquidated funds also indicated a phenomenon of short positions being squeezed in the short term. This further reflects that during the rapid price increase phase, market trading sentiment changes dramatically, leading to divergences in investors' judgments regarding short-term direction, which drives a sharp rise in the scale of liquidations.

Figure 7: On April 22, the total liquidation amount of short positions in the futures market was 517 million USD.

Quantitative Analysis - TSI Indicator, a Trading Tool for Accurately Capturing Trend Reversals and Momentum Strength

(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 performance. Investors should fully consider the risks and make decisions cautiously when engaging in related investments.)

1. Indicator Overview

The True Strength Index (TSI), developed by William Blau, is an oscillator-type indicator that smooths price momentum to help traders identify price trends, strength, and areas of overbought and oversold in the market. TSI has advantages in trading such as signaling trend reversal points, confirming momentum signals, and identifying divergences, making it particularly suitable for trend-following and momentum trading strategies.

2. Core Computing Logic

The basic calculation steps for the TSI indicator are as follows:

First, calculate the price momentum: Momentum = Current Closing Price - Previous Day's Closing Price
Perform double exponential smoothing (EMA) on the obtained momentum values, typically using common period settings of 25 days (slow line) and 13 days (fast line): EMA1 = EMA ( Momentum, Fast Period 13), EMA2 = EMA ( EMA1, Slow Period 25)
Calculate the absolute value of momentum with double EMA: Absolute Momentum = ∣Current Closing Price − Previous Day Closing Price∣, EMA3 = EMA ( Absolute Momentum, Fast Cycle 13), EMA4 = EMA (EMA3, Slow Cycle 25)
The final TSI indicator value is: TSI = EMA2 / EMA4 × 100

This calculation method allows the TSI indicator to smooth out price momentum fluctuations while clearly reflecting the current market trend and helping to determine the market's overbought and oversold conditions.

3. Trading Application Strategies

Trading Logic:

Buy Signal: Buy when the TSI indicator crosses above the set threshold from below.
Sell signal: Sell when the TSI indicator crosses the set threshold from above to below. The threshold refers to a critical value, which is the minimum or maximum value that an effect can produce.

Strategy Parameter Description: To improve the usability and flexibility of the TSI index calculation, we have set the following three basic parameters:

mDay (Fast Line Cycle): The EMA cycle parameter for the first exponential smoothing of price momentum, usually set to 13 days, used to quickly respond to changes in market prices. A smaller mDay value makes TSI more sensitive to price fluctuations; a larger value has the opposite effect.
nDay (Slow Line Cycle): The EMA cycle parameter for applying a second exponential smoothing to the momentum results after the initial smoothing, typically set to 25 days. nDay determines the responsiveness of the TSI to long-term trend fluctuations; a larger nDay results in smoother and more stable trend characteristics, while a smaller nDay leads to greater fluctuations in the curve.
threshold: Set the overbought and oversold critical level of the TSI indicator, usually using ±25 (plus or minus 25) as the reference threshold. When TSI exceeds the +threshold, it is considered to be overbought, and when it is below -threshold, it is considered to be oversold. This value can be adjusted according to the characteristics of different markets and individual stocks, with higher thresholds tending to fewer but higher quality trading signals, and lower thresholds producing more but more sensitive trading signals.

Trading Example: Taking BTC as an example, parameter settings (mDay=13, nDay=25, threshold=25):

Trading signal triggered: On April 20, 2025, at 22:00 (UTC+8), the TSI (True Strength Index) indicator for Bitcoin broke above the -25 threshold in the 15-minute K-line, which according to our strategy settings, is a clear buy signal.

Trading actions and results: After confirming the trend, investors execute a buy operation on the next K-line. Subsequently, the price experiences a surge and is sold off when the TSI indicator breaks below the +25 threshold from above. This wave of increase reached 3.36%, validating the effectiveness of the TSI indicator in this operation.

Through the above example, we have demonstrated in detail how to construct and apply the TSI indicator strategy using the three parameters mDay, nDay, and threshold. Among them, mDay and nDay represent the short-term and long-term moving average periods, respectively. By adjusting the combination of these two parameters, the sensitivity of the TSI indicator can be flexibly controlled to adapt to different market volatilities. The threshold parameter is used to set the critical entry and exit signal threshold values, such as -25 and +25, helping us to more clearly identify the overbought and oversold ranges.

By reasonably setting these parameters, the strategy can not only capture the timing of trend reversals but also effectively avoid false signals in volatile markets, thereby enhancing the overall robustness and win rate of the trading system. The above-tested case is a typical application that fully demonstrates the importance of parameter settings for strategy performance.

In the next step, we will test the performance of different parameter combinations in the market environment to find the combination that yields the best cumulative returns.

4. Parameter Optimization and Backtesting Validation

This strategy is mainly based on the True Strength Index (TSI) indicator, which grasps the turning point of the price trend through the change of momentum. TSI combines a double exponential smoothed average of price movements to provide a smoother and more representative kinetic signal while filtering out noise. We systematically optimize and backtest its three core parameters, including:

mDay: The number of days for the short-term EMA, used to capture recent price momentum;
nDay: The number of days for the long-term EMA, used to smooth the overall trend;
threshold: The threshold used for entry and exit judgment, typically set at ±25 to capture overbought and oversold ranges.

This backtest also uses the 15-minute K-line data of BTC_USDT, with the testing period from April 22, 2024, to April 22, 2025, without including transaction costs such as fees. To maintain the consistency of the trading logic, the strategy executes operations on the next K-line after each signal appears and closes the original position while opening a reverse position when the opposite signal is triggered.

Parameter Backtesting Settings To find the best combination of parameters, we conducted a systematic grid search over the following range:

mDay: 1 to 50 (step size of 50)
nDay: 1 to 50 (step size of 50) (but mDay must be less than nDay)
threshold: 10 to 30 (step size of 20)

A total of 50 × 50 × 20 = 50,000 parameter combinations were tested. We selected the five parameter sets with the best cumulative return from these combinations, based on the overall annualized return, Sharpe ratio, maximum drawdown, and Calmar ratio for performance evaluation.

Figure: Comparison of five sets of parameters for Bitcoin's cumulative return performance over a 15-minute period since April 22, 2024.

Figure: Distribution of Annualized Yield

Figure: Distribution of Sharpe Ratio

The parameters for the medium to short term ( mDay and nDay in the range of 10 to 30 ), with a lower trading threshold (approximately in the range of 10 to 20), show a significantly higher return performance.

A long period ( mDay and nDay exceeding 40 ), or a threshold that is too high (threshold exceeding 40 or 50), the strategy generally performs poorly, indicating that the combination of overly long and high parameters may not keep up with the market pace.

5. Summary of Trading Strategies

This strategy uses the True Strength Index (TSI) for trading decisions, accurately grasping price trend reversals through changes in momentum. After initial testing and optimization, it has demonstrated excellent performance. TSI effectively filters short-term market noise through double exponential smoothing, helping to capture clear trend signals.

Through systematic backtesting of BTC_USDT (15-minute period) data, we optimized the three core parameters of TSI—short-term EMA period (mDay), long-term EMA period (nDay), and threshold—over the testing period from April 22, 2024, to April 22, 2025, delineating 50,000 parameter combinations and selecting the best performing five sets for more in-depth performance analysis.

The top five combinations that performed exceptionally well mostly fall within a short-term EMA period of about 5-7, a long-term EMA period of about 16-21, and a threshold range of about 10-16. The cumulative return rate of these five parameter strategies averages around 118% to 120%, significantly higher than the BTC holding strategy during the same period (approximately 43.58%). Additionally, the overall risk indicators also show robust performance, reflected in lower maximum drawdowns (19.19% to 27.12%), high Sharpe ratios (2.25 to 2.30), and Calmar ratios (4.36 to 6.22), indicating that strategies within this parameter range also exhibit good performance in risk control, far superior to the BTC benchmark of buying and holding during the same period.

In addition, the parameter trend clearly shows that a combination of short-cycle parameters and lower trading thresholds can capture more significant market trends, creating higher and more stable excess profits for the strategy; when the mDay, nDay cycles or trading threshold parameters are set too high, the strategy's effectiveness significantly declines. This means that the effectiveness of the TSI strategy depends to some extent on the market momentum conditions and the frequency of trend reversals; excessively large cycles or high thresholds may easily miss a large number of market movements, reducing performance.

Overall, through this complete systematic empirical verification and three-dimensional parameter space visualization analysis, we confirm that the TSI strategy performs excellently in capturing short-term trends in BTC, showing significant practical potential. After parameter adjustments and optimizations, it now possesses better robustness and stable profit potential. In the future, if you wish to engage in actual trading or further optimizations, it is recommended to focus on the short-term EMA periods of 5-7, long-term EMA periods of 16-21, and trading thresholds within the range of 10-16, prioritizing attention on and leveraging the robust profit advantages in this area.

Summary

This article provides an in-depth analysis of the market performance of BTC and ETH from April 10 to April 24. After breaking through key price levels, the bullish momentum for Bitcoin and Ethereum has clearly increased, leading to a recovery in long positions. However, this has also been accompanied by significant short-term market fluctuations and intense long-short sentiment battles. Among them, indicators such as open interest, long-short trading volume ratio, and funding rates can effectively track changes in market liquidity sentiment and risk appetite.

In addition, the performance testing and parameter optimization of the quantitative indicator (TSI True Strength Index) have shown outstanding effectiveness in capturing market trend reversals and momentum strength. Combining historical backtesting with systematic parameter research, the optimized parameter range for the indicator has been determined (mDay: 5-7, nDay: 16-21, threshold: 10-16). The strategy significantly outperforms a simple buy-and-hold BTC strategy, with a stable annual return and exceptional risk control performance.

Reference:

Gate.io, https://www.gate.io/trade/BTC_USDT
Gate.io, https://www.gate.io/trade/ETH_USDT
Coinglass, https://www.coinglass.com/LongShortRatio
Coinglass, https://www.coinglass.com/BitcoinOpenInterest?utm_source=chatgpt.com
Gate.io, https://www.gate.io/futures_market_info/BTC_USD/capital_rate_history
Gate.io, https://www.gate.io/futures/introduction/funding-rate-history?from=USDT-M&contract=ETH_USDT
Coinglass, https://www.coinglass.com/pro/futures/Liquidations

Gate Research Institute 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 users are advised to conduct independent research and fully understand the nature of the assets and products they purchase before making any investment decisions. Gate.io does not take responsibility for any losses or damages resulting from such investment decisions.

BTC-0.41%

ETH-2.37%

View Original

Last edited on 2025-04-25 07:42:11

The content is for reference only, not a solicitation or offer. No investment, tax, or legal advice provided. See Disclaimer for more risks disclosure.