AlgoRover Volume Oscillator Prior-Extreme Reversal v1.1.1 — Strategy by AlgoRover
By AlgoRover
Performance Metrics
- Author: AlgoRover
- Symbol: BINANCE:HYPEUSDT.P
- Timeframe: 1 hour
Description
# Volume Oscillator Prior-Extreme Reversal v1.1.1**Reference Market / 参考交易品种：** Binance HYPE/USDT perpetual futures, TradingView symbol **BINANCE:HYPEUSDT.P**, using standard **1-hour candlesticks** with both long and short positions. All examples and backtests below refer to this instrument and timeframe. 本策略参考品种为**币安HYPE/USDT永续合约**，TradingView代码为 **BINANCE:HYPEUSDT.P**，采用**标准1小时K线**，支持多空双向交易。下文图示及回测均对应这一品种和周期。### 1. Strategy Logic / 策略逻辑Volume Oscillator Prior-Extreme Reversal uses changes in trading activity to define decision points, then compares previously recorded oscillator extremes to select a trading direction. The oscillator is calculated as **100 × [EMA(volume, 8) − EMA(volume, 14)] / EMA(volume, 14)**. Positive readings indicate that the short-term volume average exceeds the longer-term average; negative readings indicate the reverse. These readings describe relative volume expansion and contraction, rather than buying or selling pressure.成交量振荡器前极值反转策略，以成交活跃度的变化确定判断时点，再通过比较已记录的振荡器极值选择交易方向。振荡器计算公式为 **100 ×［成交量8周期EMA − 成交量14周期EMA］/ 成交量14周期EMA**。正值表示短期成交量均值高于长期均值，负值表示相反；它反映成交量相对均值的扩张与收缩，不能直接解释为买方或卖方力量。*Figure 1. Price, the oscillator and volume averages share the same time axis. Orange and aqua steps show stored positive and negative extremes; triangles mark reconstructed directional decisions, not actual fills. 图1：K线、振荡器及成交量均线采用同一时间轴。橙色与青色阶梯线分别表示已存储的正、负极值；三角形表示按规则重建的方向判断，并非实际成交。本图依据真实历史数据绘制，不是平台截图。*When the oscillator crosses zero on a confirmed bar, the strategy updates the relevant stored extreme and compares the previous positive extreme with the absolute value of the previous negative extreme. If the positive extreme is less than or equal to the negative magnitude, it selects long; otherwise, it selects short. The crossover determines when to evaluate the rule, while the extreme comparison determines direction. An upward crossover can therefore select short, and a downward crossover can select long. A qualifying opposite signal can reverse the position; repeated signals in the existing position’s direction do not add exposure.当已确认K线上的振荡器穿越零轴时，策略更新相应的已存储极值，再比较前一个正极值与前一个负极值的绝对值。正极值不大于负极值绝对值时选择多头，否则选择空头。因此，零轴穿越决定“何时判断”，前极值比较决定“选择哪个方向”；上穿零轴也可能选择空头，下穿零轴也可能选择多头。符合条件的反向信号可以触发持仓反转，同方向的重复信号不会继续加仓。*Figure 2. The downward crossover on the left selects long because 9.32 ≤ 18.62; the upward crossover on the right selects short because 9.32 > 5.80. Comparisons use stored values after the crossover updates the relevant extreme. Examples illustrate the rule and were not selected for subsequent profitability. 图2：左图下穿零轴，但因9.32≤18.62而选择多头；右图上穿零轴，但因9.32＞5.80而选择空头。比较使用穿越更新后的已存储极值，案例按逻辑特征选取，不按事后盈亏筛选。*### 2. Parameters and Visual Guide / 参数规则与图示说明The reference configuration uses volume EMA lengths of **8 and 14**, initial capital of **USD 100,000**, and a fixed target notional per new position equal to **30% of initial capital** at the default sizing leverage of **1**. Position size is based on initial capital rather than compounded as a constant percentage of current equity, and entry submission remains subject to configured exposure limits. The reference trading window runs from **26 October 2025 at 02:00 UTC to 26 October 2026 at 02:00 UTC**, excluding the ending timestamp, with a **30-calendar-day warmup**. Default costs are **0.05% commission per execution** and **2 ticks of slippage**.参考配置采用成交量EMA **8和14周期**，初始资金为 **100,000美元**，默认计算杠杆为 **1倍**，每次新持仓的目标名义金额为**初始资金的30%**。仓位以初始资金为基准，并非按当前权益的固定比例复利调整，实际入场仍需满足配置的敞口限制。参考交易窗口为 **UTC 2025年10月26日02:00至2026年10月26日02:00**，不含结束时刻，预热 **30个自然日**；默认每次成交手续费为 **0.05%**，滑点为 **2个最小价格跳动单位**。Signals are evaluated on confirmed bars, and market orders normally fill at the next bar’s open. The default **5% loss threshold** checks the completed bar’s adverse high or low relative to the average entry price, then submits a market close. It is not a resting stop order and does not guarantee a maximum realized loss of 5%.策略在K线确认后判断信号，市价订单通常于下一根K线开盘成交。默认 **5%亏损阈值**检查已完成K线中对持仓不利的最高价或最低价，相对于持仓均价是否达到阈值，再提交市价平仓。这不是提前挂出的止损订单，也不保证实际亏损不超过5%。*Figure 3. The native short position opened at 36.212 USDT, with a 5% adverse-price reference near 38.023. The recorded closing price was 38.717, producing a gross price return of approximately −6.92% before commissions. This native backtest example illustrates why a threshold does not guarantee execution at that price; it is not a TradingView fill record. 图3：这笔原生空头交易入场价为36.212 USDT，5%不利价格参考约38.023，记录中的平仓价为38.717，未计手续费的价格收益约−6.92%。观察到阈值不等于能在阈值价格成交。本例来自原生回测，不是TradingView成交记录，两种引擎的执行时序存在差异。*In the visual layout, teal and red distinguish positive and negative oscillator readings. Orange and aqua step lines show the stored prior positive and negative extremes. Upward and downward triangles show long and short decisions at zero crossovers; a decision does not necessarily produce an order. L/S labels on the price chart identify submitted entry orders, while actual fills are recorded separately by TradingView’s strategy engine. Yellow and red price lines show the average entry price and completed-bar loss reference. The status panel displays the oscillator, stored extremes, current comparison, position and reference-window status. The comparison is recalculated when a crossover updates the stored values. Visual settings do not change trading rules.图示中，青绿色与红色分别区分振荡器正值和负值，橙色与青色阶梯线显示前正极值和前负极值。向上、向下三角形分别表示零轴穿越时形成的多头、空头判断，方向判断不一定产生订单。K线上的L/S标签表示入场订单提交位置，实际成交由TradingView策略引擎另行记录；黄色与红色价格线分别表示持仓均价和亏损阈值参考。状态面板显示振荡器、已存储极值、当前比较结果、持仓及参考窗口状态，穿越零轴更新极值后会重新计算比较结果。显示设置不改变交易规则。### 3. Reference Backtest / 参考回测**TradingView：** The visual edition was verified using the reference instrument, timeframe and settings above: **54.47% cumulative return, 9.93% maximum equity drawdown, 499 closed trades, 49.10% profitable trades and a profit factor of 1.284**. Closed-trade count, return, drawdown and profit factor matched the version before the visualization changes. Return represents cumulative strategy-account performance under the stated capital and sizing assumptions.**TradingView：** 本可视化版本已按上述参考品种、周期和参数核验：**累计收益54.47%、最大权益回撤9.93%、已平仓499笔、盈利交易占比49.10%、盈利因子1.284**。已平仓交易数、收益、回撤和盈利因子均与加入可视化前一致。上述收益为指定初始资金与仓位规则下的策略账户累计收益。**AlgoRover native research / AlgoRover原生研究：** A separately saved native backtest for the reference historical window, with initial capital of **100,000 USDT**, recorded **63.65% cumulative return, 10.87% maximum drawdown, a Sharpe ratio of 1.46 and 495 closed positions**. The following chart belongs exclusively to that native run and must not be interpreted as the TradingView equity curve.**AlgoRover原生研究：** 另一份对应参考历史窗口的原生回测，以 **100,000 USDT** 为初始资金，记录的**累计收益为63.65%、最大回撤为10.87%、夏普比率为1.46、已平仓495笔**。下图仅对应这份原生回测，是独立披露的研究证据，不能视为TradingView权益曲线。*Figure 4. Equity and drawdown from saved native run 20261026-103133-735878. The upper panel shows account equity; the lower panel shows drawdown. This is an AlgoRover/EdgePilot native result, not the TradingView backtest. 图4：已保存原生回测20261026-103133-735878的权益与回撤，上图显示账户权益，下图显示回撤。本图属于AlgoRover／EdgePilot原生结果，并非TradingView回测。*These results describe one instrument and one historical window. They do not establish equivalent performance across other instruments, timeframes or future market conditions. Python/AlgoRover and TradingView use different execution engines, so their backtest results are reported separately.这些结果对应单一交易品种和特定历史窗口，不能据此认定其他品种、周期或未来市场具有相同表现。Python／AlgoRover与TradingView采用不同执行引擎，两者回测结果分别披露。### Platform Background / 平台背景This strategy was generated through **AlgoRover, formerly EdgePilot**, and adapted into Pine Script for visual inspection and backtesting on TradingView. It illustrates a workflow in which a trading idea becomes an explicit rule set that can be examined, modified and tested. According to its platform description, AlgoRover supports **more than 4,000 strategies** across different categories, custom strategy generation, modification of existing strategy parameters and logic, extensive market data, and execution on local or cloud servers. Its positioning centers on making quantitative strategy research and implementation more accessible—from generating a strategy to understanding its rules and evaluating its historical behavior.该策略通过 **AlgoRover（原EdgePilot）** 生成，并改写为Pine Script，以便在TradingView中进行图示观察与回测，展示了从交易想法到明确规则，再到理解、修改和验证的研究流程。根据平台介绍，AlgoRover支持 **4,000余种不同类别的策略**，支持策略自定义生成、现有策略参数与逻辑改写，并提供海量市场数据及本地、云端服务器运行能力。其定位贯穿策略生成、规则理解、逻辑改写和历史验证，降低量化策略研究与实现的门槛，让量化交易变得更简单。