Strategy Builder
Build, customise, and manage trading strategies from composable components.
Overview
A strategy template defines your complete trading logic — which technical indicators to use, how to combine their signals into trade decisions, when to enter and exit, and during which hours to trade.
Templates are built from composable components. Each component is a technical indicator or rule that analyses market data and produces an output. You select which components to include, assign them roles, adjust their parameters, set their weights, and choose how their outputs combine.
There are two types of templates:
- System templates — Built-in strategies that are read-only. You can duplicate a system template to create an editable copy.
- User templates — Your own templates, created from scratch or duplicated from system templates. Fully editable.
Component Roles
Every component in a strategy template is assigned a role that determines how it participates in trade decisions.
Signal
Signal components are the core of your strategy. They analyse market data and produce a directional score: +1 (bullish), -1 (bearish), or 0 (neutral). These scores are combined according to your chosen combination mode to determine whether to open a long or short position. Each signal component has a configurable weight that affects how much influence it has on the final decision.
Filter
Filters confirm or reject market conditions. They evaluate the current state of the market and return a pass/fail result. A filter that fails will block any signal from triggering a trade entry. Use filters to ensure you only trade when the broader market environment supports your strategy. Several regime detectors — ADX, Choppiness Index, and Bollinger Bandwidth — expose a trend/range mode, so the same filter can require a trending market for a breakout strategy or a consolidating one for mean-reversion.
Entry Guard
Entry guards are protective checks that run before opening a position. Unlike filters, which evaluate market conditions, entry guards focus on execution quality — for example, ensuring the price is not too far from a moving average or that there is enough potential move to justify the trade. When an entry guard fails, the trade is blocked. The Counter-Trend Veto is an entry guard that blocks entering against a strong higher-timeframe trend — useful for mean-reversion strategies that only work when the higher timeframe is ranging.
Risk Gate
Risk gates are pre-entry risk checks that run after signals have been evaluated but before the trade is opened. They return a pass/fail result based on market risk conditions — for example, blocking entries when funding rates are extreme or when volume is too thin for safe execution. Unlike filters (which run before signals), risk gates let signals evaluate so you can see what the strategy would have done, while still blocking risky entries. Components that support the risk gate role include Funding Rate, Funding Rate Velocity, Open Interest Divergence, ATR, Volume Spike, and Volatility Regime Classifier.
Exit Rule
Exit rules determine when to close an open position. They include hard stop losses, take profit targets, trailing stops, break-even stops, chandelier exits, volatility exits, time-based exits, and signal flip exits. Multiple exit rules can be active simultaneously — any exit rule that triggers will close the position.
Every strategy must include at least one protective stop — a Hard Stop Loss, Break-Even Stop, or Chandelier Exit. The builder blocks saving until one is added, so a strategy can never be deployed with a position left unprotected.
Component Categories
Components are organised into categories based on the type of market analysis they perform.
Trend Following
Components that identify and follow the direction of price trends. These include moving average crossovers, ADX (Average Directional Index), Supertrend, Donchian Channels, Ichimoku Cloud, Parabolic SAR, Aroon, swing structure analysis, multi-timeframe trend alignment, and higher-timeframe support/resistance zones. They work best in trending markets and help determine the dominant market direction.
Momentum
Components that measure the speed and strength of price movements. RSI (Relative Strength Index), MACD, Stochastic Oscillator, CCI (Commodity Channel Index), ROC (Rate of Change), Micro Momentum (a fast short-EMA slope signal tuned for low-timeframe scalping entries), Micro Velocity and Tick Trend Persistence (seconds-scale regime signals), and price action patterns fall into this category. They help identify overbought/oversold conditions and momentum shifts.
Volatility
Components that measure how much the price is fluctuating. Bollinger Bands, ATR (Average True Range), Keltner Channels, the Squeeze Detector (Bollinger inside Keltner), Volatility Regime Classifier, and ATR Percentage Filter (ATRP) help assess market volatility and breakout potential. Entry guards like Min Move Guard, ATR Regime Guard, Volatility Spike Guard, and the Fast Volatility Gate (a short-window realised-volatility band that updates on the same fast timescale as scalping entries) use volatility metrics to ensure favourable trading conditions. Micro Range Compression (a fast-bar coil detector) and the Micro Volatility Regime gate extend these to the seconds-scale regime.
Volume / Market Structure
Components that analyse trading volume, order flow, and market sentiment. MFI (Money Flow Index), OBV (On-Balance Volume), VWAP (Volume Weighted Average Price), CVD (Cumulative Volume Delta), Volume Spike, Liquidity Sweep, Funding Rate, Open Interest Divergence, OI-Weighted Volume Pressure, and Funding Rate Velocity provide insight into the strength behind price movements, positioning, and potential supply/demand imbalances.
Order Flow
Components built on real taker buy/sell trade data rather than candle-body proxies. True CVD (real aggressor cumulative volume delta), Trade Imbalance, CVD/Price Divergence, and Trade Intensity read the buy/sell split of trade-synthesized sub-1m candles and are fully backtestable. A set of live tape components — Large Print (whale) Detection, Absorption, Trade Rate Burst, Print Size Distribution, and Inter-arrival Timing — read the raw trade stream directly on running bots for order-flow-aware entries and guards; these use live data and are inactive in historical backtests.
Mean Reversion
Components designed for strategies that bet on prices returning to a mean. Williams %R, Z-Score of Price, RSI Divergence, and Micro Mean Reversion (a fast z-score fade tuned for the seconds-scale regime) identify when prices have deviated significantly from their average and may be due for a reversal.
Risk / Execution
Exit rules that manage position risk and determine when to close trades. Hard Stop Loss, Take Profit, Trailing Stop, Break-Even Stop, Chandelier Exit, Volatility Exit, Time Stop, and Signal Flip Exit all define different conditions under which a position should be closed to either protect capital or lock in gains.
Multi-Instance Components
Many components support more than one role. You can add the same component to your template multiple times, each with a different role and independent parameters. For example, you could use Funding Rate as a signal (to generate contrarian trade ideas from extreme funding) and simultaneously as a risk gate (to block entries when funding is dangerously one-sided) — each instance with its own threshold settings.
Each instance runs independently in the engine pipeline at the stage corresponding to its assigned role. This lets you get more value from a single indicator without needing duplicate components.
Combination Modes
When you have multiple signal components active, the combination mode determines how their individual outputs are merged into a single trade decision. Each mode has a threshold parameter that controls how strong the combined signal must be to trigger a trade.
Weighted Sum
Each signal's score (+1, -1, or 0) is multiplied by its weight, and the results are summed. If the absolute value of the total exceeds the Min Score threshold, a trade is triggered in the direction of the total (positive = long, negative = short).
Example: With three signals weighted 2, 3, and 1, if all three are bullish (+1), the weighted sum is 2 + 3 + 1 = 6. With a Min Score of 4, this exceeds the threshold and triggers a long entry.
All Agree
All enabled signal components that produce a non-zero score must agree on the same direction. If any signal disagrees, no trade is triggered. The Min Signals threshold sets the minimum number of non-zero signals required — this prevents a single signal from triggering a trade when others are neutral.
Example: With 4 signals and Min Signals set to 3, at least 3 signals must produce a non-zero score and all non-zero signals must point in the same direction.
Any
Any single signal can trigger a trade on its own, as long as its weighted score (score × weight) meets the Min Weighted Score threshold. This is the most permissive mode and generates the most trade signals.
Example: A signal with weight 3 producing a score of +1 has a weighted score of 3. With Min Weighted Score set to 2, this triggers a long entry.
Majority
A trade is triggered when more than half of the non-zero signals agree on a direction. The Min Majority threshold sets the minimum count of agreeing signals required.
Example: With 5 signals where 3 are bullish, 1 is bearish, and 1 is neutral, the majority is 3 bullish. With Min Majority set to 3, this triggers a long entry.
Anchor Component
All combination modes support an optional anchor component. When set, the anchor signal must produce a non-zero score in the trade direction for the trade to proceed, regardless of what the other signals produce. This allows you to designate one signal as a mandatory gate — for example, requiring that the trend direction (from a moving average crossover) aligns with the trade before any combination of momentum signals can trigger entry.
Weights & Scoring
Each signal component has a configurable weight (0.5 to 10, in 0.5 increments) that determines how much influence it has on trade decisions.
Weights are most impactful in Weighted Sum mode, where each signal's score is multiplied by its weight before summing. In Any mode, a higher weight means a single signal can more easily meet the threshold on its own. In All Agree and Majority modes, weights are not used — all signals count equally (one vote each).
Use weights to express confidence: give higher weights to signals you trust more or that have performed better historically, and lower weights to supplementary or experimental signals.
Trading Sessions
Trading sessions define time windows (in UTC) during which your bot is allowed to open new positions. Outside of active sessions, the bot will not enter new trades but will continue managing existing positions (exit rules remain active).
Each session has a label (e.g. "London + NY"), a start hour and end hour (0–23 UTC), and can be individually enabled or disabled.
The default session is "London + NY" (08:00–22:00 UTC), which covers the most liquid hours for most markets. You can add multiple sessions to trade during different windows — for example, a separate Asia session (00:00–08:00 UTC).
Trading sessions are important because market behaviour varies significantly by time of day. Liquidity, spread, and volatility all change depending on which global markets are open. Trading during low-liquidity hours can lead to worse fills and more slippage.
Trading sessions also control when your bot sleeps. When all sessions end and no positions are open, the bot automatically sleeps — deprovisioning its compute instance until the next session. This reduces costs for strategies that don't need 24/7 market monitoring.
Intervals
Each component declares which candle intervals (timeframes) it requires. The bot subscribes to these intervals and feeds the corresponding candlestick data to the component for analysis.
Available intervals:
Some components require multiple intervals — for example, a moving average crossover might use both 1h and 4h candles to confirm trend direction on multiple timeframes. The intervals a component needs are shown in the component reference below and in the template editor.
Fast-timeframe (scalping) support
Most components now run on sub-hourly timeframes (1m, 3m, 5m, 15m) as well as the hourly and daily candles they have always supported. Any pure-candle component — that is, every trend, momentum, volatility, and mean-reversion component that derives its output from a single interval's candlesticks — 48 in total, can run anywhere on the full ladder from 1m to 4h. This makes low-timeframe scalping strategies possible without giving up the same indicators used on higher timeframes.
In the template editor, each component exposes a supported-intervals menu, and you choose the interval that component evaluates on per instance. Because intervals are picked per component, one template can mix timeframes — for example, a fast 1m entry signal confirmed by a 15m trend component.
A few component families are anchored to their native timeframe by design and are not offered on the sub-hourly ladder: funding rate and open-interest components stay hourly (funding and OI only update on that cadence), multi-timeframe trend and structure guards keep their own interval sets, and exit rules keep their hourly risk unit.
Sub-1m (fast clock) — read before you use it
Below one minute, the platform offers a fast clock at 1s, 3s, 5s, 15s, and 30s. Exchanges do not publish candles at these resolutions, so the bot builds them by aggregating the live trade stream in real time. That has consequences you must design around:
- Data can be sparse or incomplete. A seconds bucket only holds data if trades actually occurred in it. On lower-liquidity pairs, sub-1m candles will have gaps and thin volume, and indicators computed on them are correspondingly noisy.
- Exchange support is limited. Only exchanges wired for trade aggregation can serve the fast clock — Bybit, Bitget, KuCoin, Phemex, and WOO X today. Picking a sub-1m interval on an exchange that cannot serve it means that series never arrives and the strategy will not evaluate on it.
- Not every supported component suits it. The catalog lets pure-candle components run at sub-1m, but many indicators tuned for minutes-to-hours behave very differently on seconds candles. Presence in the menu is not a recommendation.
Sub-1m algorithmic trading is a different discipline from candle-close strategies — faster, noisier, and far less forgiving of fees and slippage. Dedicated trade-level and sub-1m-specific components are on the roadmap; until then, treat the fast clock as an advanced, experimental tool and only reach for it if you know exactly what you are doing.
Evaluation Cadence & Cooldown
Three per-template settings control how often the strategy is evaluated, how quickly it may re-enter a symbol, and which timeframe's volatility sizes your stops. Together they let a single template run anywhere from a slow swing cadence to a fast scalping cadence.
Trigger Interval
The candle close that triggers a full strategy evaluation. The bot re-evaluates entries when a candle of this interval closes, so a shorter trigger interval means a faster trading cadence — set it to 1m or 3m for scalping. The default is 3m. The trigger interval is independent of the intervals your components analyse — a component reading 15m candles still updates on its own close, but new entries are only considered on the trigger cadence.
Entry Cooldown
A per-symbol re-entry cooldown — after a position on a symbol closes, the bot waits this long before opening another position on that same symbol. It prevents rapid-fire re-entries into the same instrument. The default is 45 minutes — set it low (for example a minute or two) for scalping strategies that intentionally re-enter often, or higher to space trades out.
Exit ATR Interval
Which interval's ATR (Average True Range) is used to size volatility-based exits — stop losses, take-profit targets, and trailing stops. The default is 1h, which suits multi-hour holds. For scalping, set it to a lower interval (for example 5m) so stop and target distances match the shorter hold times and smaller price ranges of fast trades.
To build a scalper, combine a short trigger interval, a low entry cooldown, and a low exit-ATR interval with low-timeframe components such as Micro Momentum and the Fast Volatility Gate. See the Low-Timeframe Scalping recipe for a worked example.
Component Reference
Complete reference of all available strategy components, organised by category. Each component shows its allowed roles, required intervals, and configurable parameters with defaults and ranges.
Trend Following
Moving Average Crossover
Short MA crosses above/below long MA to signal trend shifts
| Parameter | Default | Range | Description |
|---|---|---|---|
| Short Period | 9 | min 2, max 100 | Period for the fast moving average |
| Long Period | 21 | min 5, max 500 | Period for the slow moving average |
| MA Type | ema | ema | sma | Type of moving average to use |
ADX (Average Directional Index)
Measures trend strength (not direction); >25 = strong trend
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 14 | min 2, max 100 | ADX calculation period |
| Strength Threshold | 25 | min 10, max 60 | Minimum ADX value to consider a strong trend |
| Mode | trend | trend | range | trend: allow when ADX > threshold (strong trend); range: allow when ADX < threshold (consolidating) |
Choppiness Index
Trend-vs-range regime via the Choppiness Index. Range mode allows entries only when consolidating; trend mode only when trending.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 14 | min 2, max 100 | Choppiness lookback |
| Threshold | 61.8 | min 10, max 90, step 0.1 | CHOP boundary — ≥ = choppy/ranging, ≤ = trending |
| Mode | range | range | trend | range: allow when choppy/consolidating; trend: allow when trending |
Donchian Channels
N-period highest high / lowest low breakout — classic Turtle Trading signal
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 20 | min 5, max 200 | Lookback period for highest high / lowest low |
Ichimoku Cloud
Multi-component trend system: TK cross, cloud position, and Chikou confirmation
| Parameter | Default | Range | Description |
|---|---|---|---|
| Tenkan Period | 9 | min 2, max 50 | Conversion line (Tenkan-sen) period |
| Kijun Period | 26 | min 5, max 100 | Base line (Kijun-sen) period |
| Senkou B Period | 52 | min 10, max 200 | Senkou Span B period (cloud) |
| Displacement | 26 | min 1, max 100 | Cloud displacement / Chikou shift |
Parabolic SAR
Trailing stop-and-reverse dots; signals trend flips via SAR direction change
| Parameter | Default | Range | Description |
|---|---|---|---|
| AF Start | 0.02 | min 0.005, max 0.1, step 0.005 | Initial acceleration factor |
| AF Step | 0.02 | min 0.005, max 0.1, step 0.005 | AF increment on new extreme |
| AF Max | 0.2 | min 0.1, max 0.5, step 0.01 | Maximum acceleration factor |
Aroon
Measures time since N-period high/low; detects trend starts and ends
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 25 | min 5, max 100 | Lookback period for Aroon calculation |
| Threshold | 70 | min 50, max 100 | Aroon level for strong trend signal |
Supertrend
ATR-based trailing stop/trend line; binary bullish/bearish signal
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 10 | min 2, max 100 | Period for ATR calculation |
| Multiplier | 3 | min 0.5, max 10, step 0.1 | ATR multiplier for band width |
Swing Structure Break
Detects breaks of swing highs/lows with RSI quality filter
| Parameter | Default | Range | Description |
|---|---|---|---|
| Swing Lookback | 20 | min 5, max 100 | Candles to scan for swing highs/lows |
| RSI Period | 14 | min 2, max 100 | RSI period for quality filter |
| Bull RSI Min | 45 | min 0, max 100 | Minimum RSI for bullish break quality filter |
| Bull RSI Max | 65 | min 0, max 100 | Maximum RSI for bullish break (avoids overbought entries) |
| Bear RSI Min | 35 | min 0, max 100 | Minimum RSI for bearish break (avoids oversold entries) |
| Bear RSI Max | 55 | min 0, max 100 | Maximum RSI for bearish break quality filter |
Counter-Trend Veto
Blocks entries against a confirmed higher-timeframe EMA-slope trend (long in a falling trend, short in a rising one); passes in consolidation.
| Parameter | Default | Range | Description |
|---|---|---|---|
| EMA Period | 21 | min 5, max 200 | EMA period for the trend read |
| Slope Lookback | 3 | min 1, max 20 | Bars to measure EMA slope over |
| Flat Threshold % | 0.001 | min 0, max 0.05, step 0.0005 | Slope magnitude below this = flat (pass both directions) |
EMA Distance Guard
Blocks entry when price is extended too far from 4h EMA in ATR units
| Parameter | Default | Range | Description |
|---|---|---|---|
| Max EMA Distance (ATR) | 2 | min 0.5, max 10, step 0.1 | Maximum distance from 4h EMA in ATR units before blocking entry |
| EMA Period | 21 | min 5, max 200 | Period for 4h EMA calculation |
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation (used as distance unit) |
| Mode | trend | trend | revert | trend: block entries extended with the move past EMA (anti-chase); revert: require price stretched against the move ≥ Max EMA Distance (mean-reversion) |
Multi-Timeframe Trend Alignment
Confirms trend alignment across multiple timeframes using configurable trend detection (EMA slope or Supertrend direction)
| Parameter | Default | Range | Description |
|---|---|---|---|
| Trend Method | ema_slope | ema_slope | supertrend | Method to detect trend on each timeframe: EMA slope or Supertrend direction |
| EMA Period | 21 | min 5, max 200 | EMA period used for slope-based trend detection |
| Slope Lookback | 3 | min 1, max 20 | Number of bars to measure EMA slope over |
| Supertrend ATR Period | 10 | min 2, max 50 | ATR period for Supertrend-based trend detection |
| Supertrend Multiplier | 3 | min 0.5, max 10, step 0.5 | ATR multiplier for Supertrend-based trend detection |
| Timeframes | 15,60,240 | string | Comma-separated intervals to check alignment across (e.g. 15,60,240) |
HTF Support/Resistance Zones
Identifies support/resistance zones from a higher timeframe using swing highs/lows, blocking entries that approach major structural levels from the wrong side
| Parameter | Default | Range | Description |
|---|---|---|---|
| HTF Interval | 240 | 60 | 240 | 1440 | 10080 | Higher timeframe to detect S/R zones on (1h, 4h, 1d, or 1w) |
| Swing Lookback | 50 | min 10, max 200, step 5 | Number of HTF bars to scan for swing highs/lows |
| Zone Width (ATR×) | 0.5 | min 0.1, max 3, step 0.1 | Width of each S/R zone as a multiple of HTF ATR |
| ATR Period | 14 | min 2, max 100 | ATR period for zone width calculation |
Momentum
CCI (Commodity Channel Index)
Measures deviation from statistical mean; cycle-based overbought/oversold
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 20 | min 5, max 200 | CCI calculation period |
| Overbought Level | 100 | min 50, max 300, step 10 | CCI level above which asset is overbought |
| Oversold Level | -100 | min -300, max -50, step 10 | CCI level below which asset is oversold |
ROC (Rate of Change)
Pure price momentum as percentage change over N periods
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 12 | min 1, max 200 | Lookback period for rate of change |
| Threshold | 0 | min 0, max 20, step 0.5 | Minimum absolute ROC to trigger signal (0 = any direction) |
RSI (Relative Strength Index)
Overbought/oversold oscillator for directional signals
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 14 | min 2, max 100 | RSI calculation period |
| Oversold Level | 30 | min 5, max 49 | RSI level below which asset is oversold (buy) |
| Overbought Level | 70 | min 51, max 95 | RSI level above which asset is overbought (sell) |
MACD
Momentum via dual EMA difference + signal line crossover
| Parameter | Default | Range | Description |
|---|---|---|---|
| Fast Period | 12 | min 2, max 100 | Fast EMA period |
| Slow Period | 26 | min 5, max 200 | Slow EMA period |
| Signal Period | 9 | min 2, max 50 | Signal line EMA period |
Stochastic Oscillator
Compares close to recent range; good for mean-reversion in ranging markets
| Parameter | Default | Range | Description |
|---|---|---|---|
| %K Period | 14 | min 2, max 100 | Lookback period for %K |
| %D Period | 3 | min 2, max 50 | Smoothing period for %D (SMA of %K) |
| Slowing | 3 | min 1, max 10 | Slowing factor for %K smoothing |
| Oversold Level | 20 | min 5, max 40 | Level below which stochastic is oversold |
| Overbought Level | 80 | min 60, max 95 | Level above which stochastic is overbought |
Price Action Patterns
Detects engulfing candles, pin bars, and momentum continuation
No configurable parameters.
RSI Exhaustion Guard
Blocks entry when RSI is overbought (longs) or oversold (shorts)
| Parameter | Default | Range | Description |
|---|---|---|---|
| RSI Ceiling (Long) | 65 | min 50, max 95 | Maximum RSI to allow long entries (above = overbought exhaustion) |
| RSI Floor (Short) | 35 | min 5, max 50 | Minimum RSI to allow short entries (below = oversold exhaustion) |
| RSI Period | 14 | min 2, max 100 | RSI calculation period |
Micro Momentum
Fast micro-trend signal: the slope of a short EMA over a few recent candles, tuned for low-timeframe scalping entries
| Parameter | Default | Range | Description |
|---|---|---|---|
| EMA Period | 9 | min 2, max 100 | Period of the short EMA whose slope is measured |
| Slope Lookback | 3 | min 1, max 50 | How many candles back to measure the EMA slope over |
| Slope Threshold (%) | 0.05 | min 0, max 5, step 0.01 | Minimum EMA slope (percent of price) to register a directional signal |
Tick Trend Persistence
Fraction of the last N fast bars whose close moved the same direction — persistent micro-trend vs choppy tape.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (bars) | 10 | min 3, max 100 | Number of recent bars to measure persistence over |
| Min Persistence | 0.7 | min 0.5, max 1, step 0.05 | One-sided fraction required to register a micro-trend |
Micro Velocity Regime
Directional micro-regime from short-window price velocity (close ROC per bar) once it clears a threshold.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (bars) | 10 | min 2, max 100 | Bars over which to measure velocity |
| Velocity Threshold (%/bar) | 0.02 | min 0, max 2, step 0.01 | Minimum per-bar velocity to register a direction |
Volatility
Bollinger Bandwidth Regime
Volatility-compression regime via percentile-ranked Bollinger bandwidth. Squeeze mode allows entries only when compressed; expansion mode only when expanded.
| Parameter | Default | Range | Description |
|---|---|---|---|
| BB Period | 20 | min 5, max 200 | Bollinger SMA period |
| BB Std Dev | 2 | min 0.5, max 5, step 0.1 | Standard-deviation multiplier |
| Percentile Lookback | 100 | min 20, max 500 | Bars to percentile-rank bandwidth over |
| Percentile Threshold | 25 | min 5, max 50 | Squeeze if rank ≤ this; expansion if rank ≥ 100−this |
| Mode | squeeze | squeeze | expansion | squeeze: allow when compressed; expansion: allow when expanded |
Squeeze Detector
Bollinger Bands inside Keltner Channels = squeeze; release signals breakout direction
| Parameter | Default | Range | Description |
|---|---|---|---|
| BB Period | 20 | min 5, max 200 | Bollinger Bands SMA period |
| BB Std Dev | 2 | min 0.5, max 5, step 0.1 | Bollinger Bands standard deviation multiplier |
| KC EMA Period | 20 | min 5, max 200 | Keltner Channels EMA period |
| KC ATR Multiplier | 1.5 | min 0.5, max 5, step 0.1 | Keltner Channels ATR multiplier |
| KC ATR Period | 10 | min 2, max 100 | Keltner Channels ATR period |
| Momentum Period | 12 | min 1, max 50 | Lookback for momentum direction on squeeze release |
Bollinger Bands
Price envelope based on standard deviation; band squeeze = breakout setup
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 20 | min 5, max 200 | SMA period for middle band |
| Std Dev Multiplier | 2 | min 0.5, max 5, step 0.1 | Standard deviation multiplier for band width |
ATR (Average True Range)
Measures raw volatility; filters out low-volatility regimes
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 14 | min 2, max 100 | ATR calculation period |
| Baseline Candles | 50 | min 10, max 200 | Number of candles for baseline ATR comparison |
| Min Volatility Ratio | 0.7 | min 0.1, max 2, step 0.1 | Minimum ratio of current ATR to baseline (below = low volatility regime) |
Keltner Channels
ATR-based envelope around EMA; breakout outside = high-conviction signal
| Parameter | Default | Range | Description |
|---|---|---|---|
| EMA Period | 20 | min 5, max 200 | EMA period for middle line |
| ATR Multiplier | 1.5 | min 0.5, max 5, step 0.1 | ATR multiplier for channel width |
| ATR Period | 10 | min 2, max 100 | Period for ATR calculation |
Min Move Guard
Blocks entry when expected TP move is too small relative to price
| Parameter | Default | Range | Description |
|---|---|---|---|
| Min Move % | 0.005 | min 0.001, max 0.05, step 0.001 | Minimum TP distance as fraction of entry price to allow entry |
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| ATR Multiplier | 1.5 | min 0.5, max 5, step 0.1 | ATR multiplier for stop distance |
| TP R:R Ratio | 2 | min 0.5, max 10, step 0.1 | Take profit reward-to-risk ratio |
Volatility Spike Guard
Blocks entry during high-volatility spikes by comparing recent ATR to baseline
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| Baseline Candles | 50 | min 10, max 200 | Number of candles for baseline ATR comparison |
| Max Volatility Ratio | 2 | min 1.1, max 5, step 0.1 | Maximum ratio of recent ATR to baseline (above = volatility spike) |
ATR Regime Guard
Blocks entry in low-volatility regimes by comparing recent ATR to baseline
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| Baseline Candles | 50 | min 10, max 200 | Number of candles for baseline ATR comparison |
| Min Volatility Ratio | 0.7 | min 0.1, max 2, step 0.1 | Minimum ratio of recent ATR to baseline (below = low volatility) |
Volatility Regime Classifier
Classifies the current volatility regime (low/normal/high/extreme) using ATR percentile rank over a long lookback window
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| Percentile Lookback | 100 | min 20, max 500, step 10 | Number of ATR values to rank against for percentile calculation |
| Allowed Regimes | normal,high | string | Comma-separated regimes to pass: low (<25th), normal (25th-75th), high (75th-90th), extreme (>90th) |
ATR Percentage Filter
Filters by normalized volatility (ATR as percentage of price), enabling consistent volatility thresholds across instruments at different price levels
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| Min ATR% | 0.5 | min 0, max 20, step 0.1 | Minimum ATR% to allow entry (filters out low-volatility instruments) |
| Max ATR% | 10 | min 0.1, max 50, step 0.1 | Maximum ATR% to allow entry (filters out extremely volatile instruments) |
Fast Volatility Gate
Blocks entries when short-window realised volatility is outside a workable band — updates on the same fast timescale as scalping entries
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (candles) | 20 | min 3, max 200 | Rolling window of recent candles for realised volatility |
| Min Volatility (%) | 0.03 | min 0, max 10, step 0.01 | Block entry when realised vol per candle is below this (too quiet to scalp) |
| Max Volatility (%) | 2 | min 0.1, max 50, step 0.1 | Block entry when realised vol per candle is above this (too chaotic) |
Micro Range Compression
Recent mean bar range compressed below a longer baseline (a coil) — a breakout-entry regime on fast bars.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (bars) | 10 | min 3, max 100 | Recent window for mean bar range |
| Baseline Multiple | 3 | min 2, max 10 | Baseline window = window × this multiple |
| Compression Ratio | 0.5 | min 0.1, max 1, step 0.05 | Coil when recent range ≤ this fraction of baseline |
Micro Volatility Regime
Classifies fast-bar realised vol (quiet/normal/chaotic) from mean absolute per-bar return; gates scalps to the normal band.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (bars) | 20 | min 3, max 200 | Rolling window for realised volatility |
| Quiet Below (%/bar) | 0.02 | min 0, max 5, step 0.01 | Below this per-bar vol the tape is too quiet to scalp |
| Chaotic Above (%/bar) | 0.5 | min 0.05, max 20, step 0.05 | Above this per-bar vol the tape is too chaotic |
Volume / Market Structure
Funding Rate
Contrarian signal from extreme perpetual funding rates; fade when funding is one-sided
| Parameter | Default | Range | Description |
|---|---|---|---|
| Threshold | 0.01 | min 0.001, max 0.1, step 0.001 | Absolute funding rate threshold to trigger signal (e.g. 0.01 = 1%) |
Open Interest Divergence
Detects divergence between price direction and open interest changes
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback | 14 | min 2, max 100 | Number of candles to measure price/OI change over |
MFI (Money Flow Index)
Volume-weighted RSI; uses price × volume to measure buying/selling pressure
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 14 | min 2, max 100 | MFI calculation period |
| Oversold Level | 20 | min 5, max 40 | MFI level below which asset is oversold (buy) |
| Overbought Level | 80 | min 60, max 95 | MFI level above which asset is overbought (sell) |
OBV (On-Balance Volume)
Cumulative volume direction; divergence from price = potential reversal
| Parameter | Default | Range | Description |
|---|---|---|---|
| OBV EMA Period | 21 | min 5, max 100 | EMA period applied to OBV for trend confirmation |
VWAP
Institutional benchmark price; price above/below VWAP as intraday bias filter
| Parameter | Default | Range | Description |
|---|---|---|---|
| Required Bias | above | above | below | any | Filter direction: above VWAP passes for longs, below for shorts, any passes always |
CVD (Cumulative Volume Delta)
Tracks buy vs sell aggression; divergence exposes trapped liquidity
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback | 20 | min 5, max 100 | Number of candles to compare CVD direction over |
Volume Spike
Z-score volume spike with directional body confirmation
| Parameter | Default | Range | Description |
|---|---|---|---|
| Baseline Candles | 50 | min 10, max 200 | Number of candles for volume baseline |
| Z-Score Threshold | 1.5 | min 0.5, max 5, step 0.1 | Z-score required to trigger spike signal |
Liquidity Sweep
Wick pierces local high/low then closes back — sweep of resting liquidity
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback | 14 | min 5, max 100 | Number of candles for local high/low detection |
OI-Weighted Volume Pressure
Classifies positioning pressure using OI change + volume into 4 regimes: strong conviction, short covering, new shorts, or liquidation cascade
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback Period | 5 | min 2, max 50 | Number of bars to measure OI change and volume trend over |
| Volume Threshold | 1.2 | min 0.5, max 5, step 0.1 | Minimum volume ratio vs lookback average to confirm signal |
Funding Rate Velocity
Measures rate of change of the funding rate to detect accelerating crowding or unwinding positioning, generating contrarian signals at velocity extremes
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback Period | 8 | min 2, max 50 | Number of funding rate observations to measure velocity over |
| Velocity Threshold | 0.005 | min 0.001, max 0.1, step 0.001 | Minimum absolute funding rate change over the lookback to trigger signal |
Mean Reversion
Williams %R
Fast overbought/oversold oscillator; inverse of Stochastic %K for aggressive mean-reversion
| Parameter | Default | Range | Description |
|---|---|---|---|
| Period | 14 | min 2, max 100 | Lookback period |
| Oversold Level | -80 | min -100, max -50 | Level below which %R is oversold (buy signal) |
| Overbought Level | -20 | min -50, max 0 | Level above which %R is overbought (sell signal) |
Z-Score of Price
Measures standard deviations from rolling mean; classic stat-arb input
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback Period | 50 | min 10, max 500 | Rolling window for mean and std dev calculation |
| Entry Threshold | 2 | min 0.5, max 5, step 0.1 | Z-score magnitude required to trigger entry signal |
| Exit Threshold | 0.5 | min 0, max 2, step 0.1 | Z-score magnitude below which to signal exit (mean reversion) |
RSI Divergence
Price/RSI divergence detection; leading reversal signal
| Parameter | Default | Range | Description |
|---|---|---|---|
| RSI Period | 14 | min 2, max 100 | Period for RSI calculation |
| Oversold Level | 30 | min 5, max 49 | RSI level below which regular bullish divergence is valid |
| Overbought Level | 70 | min 51, max 95 | RSI level above which regular bearish divergence is valid |
| Divergence Lookback | 5 | min 2, max 50 | Candles per window for comparing price/RSI extremes |
Micro Mean Reversion
Fades fast overextensions: z-score of the latest close vs a short rolling mean beyond ±threshold.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (bars) | 20 | min 3, max 200 | Rolling window for the close mean/stddev |
| Z Threshold | 2 | min 0.5, max 6, step 0.1 | Fade when |z| exceeds this |
Order Flow
True CVD (Aggressor Delta)
Cumulative REAL taker buy−sell volume delta from trade-synthesized candles (unlike the candle-body `cvd` proxy). Sub-1m only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Lookback | 20 | min 5, max 100 | Candles to compare cumulative aggressor delta over |
Trade Imbalance
Taker buy share buy/(buy+sell) over a window of trade-synthesized candles versus a neutral 0.5 ± band. Sub-1m only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (candles) | 20 | min 5, max 200 | Rolling window of candles to sum aggressor volume over |
| Imbalance Band | 0.1 | min 0, max 0.49, step 0.01 | Distance from 0.5 buy-share to register a signal |
CVD / Price Divergence
Fades an unsupported move: price direction disagreeing with real aggressor delta over a window. Sub-1m only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (candles) | 14 | min 3, max 200 | Window over which to compare price change vs aggressor delta |
Trade Intensity
Z-score of a candle's print count vs a rolling baseline; blocks entries during abnormal tape bursts. Sub-1m only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Baseline (candles) | 50 | min 5, max 500 | Rolling baseline window for the print-count z-score |
| Z Threshold | 2 | min 0.5, max 6, step 0.1 | Block when print-count z-score exceeds this |
Trade Rate Burst
Ratio of recent trade arrival rate to a longer baseline; blocks entries during activity bursts. Live only — neutral in backtest.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Recent Window (ms) | 5000 | min 500, max 60000, step 500 | Short window for the recent arrival rate |
| Baseline Window (ms) | 60000 | min 5000, max 600000, step 1000 | Longer window for the baseline arrival rate |
| Burst Multiple | 3 | min 1, max 20, step 0.5 | Block when recent rate exceeds this multiple of baseline |
Large Print Detection
Net aggressor side of top-percentile (whale) prints among recent trades. Live only — neutral in backtest.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Size Percentile | 90 | min 50, max 99 | Prints at/above this size percentile count as large |
| Lookback Prints | 200 | min 20, max 2000, step 10 | How many recent prints to consider |
Absorption
Heavy one-sided aggression that fails to move price (passive absorption) → reversal against the aggressor. Live only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (ms) | 10000 | min 1000, max 60000, step 500 | Trade window to assess absorption over |
| Max Price Move (%) | 0.05 | min 0, max 2, step 0.01 | Absorption requires price range below this percent |
| Min Imbalance | 0.65 | min 0.5, max 1, step 0.01 | Minimum one-sided buy (or sell) share to call absorption |
Print Size Distribution
Share of window volume from whale-sized prints (≥ multiple × median); blocks when whale-dominated. Live only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (ms) | 60000 | min 5000, max 300000, step 1000 | Trade window for the size distribution |
| Whale Multiple | 5 | min 2, max 50 | A print is a whale at ≥ this multiple of the median size |
| Max Whale Share | 0.5 | min 0, max 1, step 0.05 | Block when whale volume share exceeds this |
Inter-arrival Timing
Clustering (CV of inter-print gaps) and mean gap of trade arrivals; requires a liquid, non-sparse tape. Live only.
| Parameter | Default | Range | Description |
|---|---|---|---|
| Window (ms) | 30000 | min 1000, max 120000, step 1000 | Trade window for inter-arrival statistics |
| Max Mean Gap (ms) | 2000 | min 50, max 30000, step 50 | Block when the mean gap between prints exceeds this |
Risk / Execution
Hard Stop Loss
ATR-based hard stop loss
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| ATR Multiplier | 2 | min 0.5, max 5, step 0.1 | Stop-loss distance as multiple of ATR |
Take Profit
R:R ratio-based take profit
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| ATR Multiplier | 2 | min 0.5, max 5, step 0.1 | Stop-loss distance as multiple of ATR (used for R:R base) |
| TP R:R Ratio | 3 | min 0.5, max 10, step 0.1 | Take profit as a multiple of the stop-loss distance |
Trailing Stop
ATR-based trailing stop with activation threshold
| Parameter | Default | Range | Description |
|---|---|---|---|
| Activation % | 0.008 | min 0.001, max 0.05, step 0.001 | Profit percentage to activate trailing stop |
| Trail ATR Multiple | 1.5 | min 0.5, max 5, step 0.1 | Trail distance as multiple of ATR |
| Min Trail % | 0.005 | min 0.001, max 0.05, step 0.001 | Minimum trailing distance as fraction of price |
Time Stop
Time-based exit with dynamic hold periods
| Parameter | Default | Range | Description |
|---|---|---|---|
| Base Duration (s) | 21600 | min 60, max 604800, step 60 | Base maximum hold time in seconds |
| Short Duration (s) | 28800 | min 60, max 604800, step 60 | Hold time for short positions (optional) |
| Extended Duration (s) | 36000 | min 60, max 604800, step 60 | Extended hold time for profitable trend-aligned positions |
| Early Kill Time (s) | 14400 | min 60, max 604800, step 60 | Kill losing positions after this time |
| Early Kill Loss % | 0.005 | min 0.001, max 0.05, step 0.001 | Loss threshold for early kill |
Break-Even Stop
Moves stop to entry + buffer after position reaches configurable R:R profit
| Parameter | Default | Range | Description |
|---|---|---|---|
| Activation R:R | 1 | min 0.5, max 5, step 0.1 | R:R threshold to activate break-even (e.g., 1.0 = 1R profit) |
| Buffer (ATR) | 0.1 | min 0, max 1, step 0.05 | Buffer above entry price in ATR units (small margin for noise) |
Chandelier Exit
ATR-based trailing stop anchored to highest high (longs) or lowest low (shorts)
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 22 | min 2, max 100 | Period for ATR calculation |
| ATR Multiplier | 3 | min 0.5, max 10, step 0.1 | ATR multiplier for chandelier distance |
| Lookback | 22 | min 5, max 100 | Period for highest high / lowest low anchor |
Volatility Exit
Exit when volatility regime collapses (ATR drops below baseline ratio)
| Parameter | Default | Range | Description |
|---|---|---|---|
| ATR Period | 14 | min 2, max 100 | Period for ATR calculation |
| Baseline Candles | 50 | min 10, max 200 | Number of candles for baseline ATR comparison |
| Min Volatility Ratio | 0.5 | min 0.1, max 1.5, step 0.1 | Exit when ATR / baseline drops below this ratio (regime collapse) |
Signal Flip Exit
Exit when engine direction flips against position
No configurable parameters.