(Bearish) CrossUnder

The Crossunder module detects the precise moment when one data series crosses below another—a classic bearish signal in technical analysis. This is the mirror image of the Crossover module and is essential for identifying trend reversals and exit points.

---

What Is a Crossunder?

A crossunder occurs when a faster-moving line drops below a slower-moving line:

graph LR
    subgraph Before["Before Crossunder"]
        A1["Fast Line: ABOVE"]
        A2["Slow Line: BELOW"]
    end
    
    subgraph Cross["⚡ Crossunder Moment"]
        B["Lines Intersect"]
    end
    
    subgraph After["After Crossunder"]
        C1["Fast Line: BELOW"]
        C2["Slow Line: ABOVE"]
    end
    
    Before --> Cross --> After

The module outputs True at the exact moment the first dataset crosses below the second (a bearish cross), False at every other bar where the comparison is possible, and null where it is not (see Missing Data below). For the bullish direction, use the Crossover module.

---

The Mathematics

The crossunder compares each bar against the last observed bar before it — call it $p$:

$$\text{CrossUnder}_t = \begin{cases} \text{True}, & \text{if } (A_{p} - B_{p}) \ge 0 \text{ and } (A_t - B_t) < 0 \\ \text{False}, & \text{otherwise} \end{cases}$$

Where:

• $A_t$ is the first dataset (typically the faster indicator) at time $t$
• $B_t$ is the second dataset (typically the slower indicator) at time $t$
• $p$ is the most recent bar before $t$ where both datasets have data

Note: This is a bearish-only signal — True only when the first dataset moves from at-or-above to below the second (standard ta.crossunder semantics). An upward cross returns False; use the Crossover module to detect it. On an uninterrupted daily series $p$ is simply $t-1$; across a weekend or data gap it is the last bar that actually traded (see Missing Data).

---

Missing Data

A missing (null) datapoint — a data gap or a closed-market slot — produces a null at the same position in the output. The signal is never faked with substitute values.

• Gaps are bridged, not broken: the cross is evaluated between the last two bars that actually have data. A Friday-close-to-Monday-open cross fires True on Monday — the weekend nulls stay null, and no signal is lost.
• The first bar is always null: a crossunder needs a previous bar to compare against, and the window's oldest bar has none. The same applies to the first bar with data when the series starts with a gap.
• Sporadic outage gaps: if data drops out mid-session, the cross is computed against the last bar before the gap, so a cross that happened during the gap surfaces on the first bar after it — delayed, never destroyed.
• Live execution returns exactly one datapoint — the newest — which is null when the current bar's data is missing.

---

The Death Cross: A Famous Crossunder

The most famous crossunder pattern in equity trading:

Fast Line	Slow Line	Signal
50-day SMA	200-day SMA	Death Cross (Bearish)

flowchart TD
    SMA50["50-Day SMA"]
    SMA200["200-Day SMA"]
    CU["Crossunder Module"]
    Signal{"Death Cross?"}
    Sell["Bearish Signal"]
    Hold["Hold Position"]
    
    SMA50 --> CU
    SMA200 --> CU
    CU --> Signal
    Signal -->|True| Sell
    Signal -->|False| Hold

Historically, the Death Cross has preceded major bear markets. In March 2020, the S&P 500 experienced a Death Cross just as the COVID crash accelerated. However, like all technical signals, it's not infallible—the market recovered rapidly afterward.

---

Building a Complete Trading System

Crossunder works best when paired with Crossover for a complete entry/exit system:

flowchart TD
    subgraph Data["Price Data"]
        Price["Close Prices"]
    end
    
    subgraph Indicators["Moving Averages"]
        EMA12["EMA(12)"]
        EMA26["EMA(26)"]
    end
    
    subgraph Detection["Signal Detection"]
        CO["Crossover Module"]
        CU["Crossunder Module"]
    end
    
    subgraph Actions["Trading Actions"]
        Buy["Enter Long"]
        Sell["Exit / Short"]
    end
    
    Price --> EMA12 & EMA26
    EMA12 & EMA26 --> CO & CU
    CO --> Buy
    CU --> Sell

---

Use Cases Beyond Moving Averages

1. RSI Overbought Exit

Detect when RSI crosses below the 70 overbought threshold:

First Dataset	Second Dataset	Signal Meaning
RSI	Constant 70	Exit overbought zone

2. MACD Signal Line Cross

The MACD crossing below its signal line is a momentum sell signal:

First Dataset	Second Dataset	Signal Meaning
MACD Line	Signal Line	Momentum turning bearish

3. Price Breaking Support

Detect when price crosses below a key moving average:

First Dataset	Second Dataset	Signal Meaning
Close Price	200-day SMA	Major support broken

---

Cryptocurrency Considerations

Crypto markets are particularly volatile, making crossunder signals both:

• More frequent — Higher volatility means more crossings
• More significant — Large moves can follow confirmed breakdowns

Consider using:

• Longer timeframes (4H, Daily) to filter noise
• Volume confirmation — Breakdowns with high volume are more reliable
• Multiple confirmations — Wait for 2-3 candles below the cross point

---

Advantages

• Clear exit signals — Unambiguous moment to close positions
• Pairs naturally with Crossover — Complete entry/exit system
• Works across all markets — Equities, crypto, forex, commodities
• Objective — No interpretation needed

Disadvantages

• Lagging indicator — Signals come after the move begins
• Whipsaws — Sideways markets generate false signals
• No magnitude — Can't distinguish small crosses from major breakdowns

---

---

Data Alignment in Systemic

When connecting inputs to the Crossunder module, Systemic automatically aligns your data:

Scenario	First Dataset	Second Dataset	Behavior
Both from modules	BTC 1H	ETH 15m	Both upsampled to 15m frequency
Module + Constant	RSI output	70	Constant padded to match RSI length
Different lengths	100 points	150 points	Shorter dataset padded by repeating last value

Important Considerations

• Upsampling uses forward-fill: When a 1-hour series is aligned to 15-minute data, intermediate points inherit the previous hour's value
• No interpolation: The system does not interpolate between values—this ensures crossunders are detected based on actual data points
• First output is always null: Since crossunder requires a previous bar to compare against, the first output value cannot be computed and is null — never a fabricated False

---

Indicator Assessment

%%{init: {"themeVariables": {"cScale0": "#4F46E5", "cScale1": "#059669"}}}%%
quadrantChart
    title Crossunder Indicator Assessment
    x-axis Low --> High
    y-axis Low --> High
    quadrant-1 High Popularity & Reliability
    quadrant-2 High Reliability & Low Popularity
    quadrant-3 Low Reliability & Popularity
    quadrant-4 High Popularity & Low Reliability
    "Traditional Markets": [0.85, 0.75]
    "Cryptocurrency": [0.75, 0.60]

Dimension	Stocks/Macro	Crypto	Explanation
Popularity	⭐⭐⭐⭐⭐	⭐⭐⭐⭐	The Death Cross is among the most cited technical signals in financial media. Crypto traders also widely use it, though newer indicators sometimes take precedence.
Reliability	⭐⭐⭐⭐	⭐⭐⭐	In traditional markets, crossunders on major timeframes (daily/weekly) have historically preceded significant moves. Crypto's higher volatility leads to more false signals.
Complexity	⭐	⭐	One of the simplest indicators to understand and implement—just compare two values at consecutive time points.
Versatility	⭐⭐⭐⭐⭐	⭐⭐⭐⭐⭐	Applicable to any two time series: prices, indicators, on-chain metrics, or custom data. Works across all timeframes and asset classes.
Recent Interest	⭐⭐⭐	⭐⭐⭐⭐	A classic indicator with steady interest. The 2022 crypto bear market renewed focus on bearish crossunder signals for risk management.

---

Further Reading

• Death Cross - Investopedia
• Moving Average Crossover Strategies