MatHold

Five-bar bullish continuation. A long white candle is followed by a brief three-bar pullback that gaps up and then drifts on small bodies without surrendering much ground, after which a white candle breaks to a new high and the uptrend resumes.

Quick reference

Field	Value
Family	Candlestick Patterns
Input type	Candle
Output type	f64 — +1.0 bullish, 0.0 otherwise (never -1.0)
Output range	{0.0, +1.0}
Default parameters	penetration = 0.5 (TA-Lib default); bindings use the default
Warmup period	5 (first four bars always 0.0)
Interpretation	Bullish continuation after a shallow rest

Formula

long body = |close − open| >= 0.5 · (high − low)
bar1 white & long
bar2 small body gapping up above bar1        (min(o2,c2) > close1)
bar2, bar3, bar4 each small                   (|body| <= 0.5 · body1)
the pullback holds                            (min low of bars 2..4 > close1 − penetration · body1)
bar5 white, closing at a new high             (close5 > max high of bars 1..4)

Bullish-only (never −1.0). A close-cousin of RisingThreeMethods where the rest gaps up and merely holds rather than drifting back through the first body. See crates/wickra-core/src/indicators/mat_hold.rs.

Parameters

Name	Type	Default	Valid range	Source
penetration	f64	0.5	[0.0, 1.0)	MatHold::with_penetration (mat_hold.rs)

with_penetration outside [0, 1) errors (rejects_invalid_penetration, accepts_valid_penetration). Python/Node construct with the default.

Signed ±1 encoding

Single-direction shape: +1.0 bullish, 0.0 no pattern — one feature-matrix dimension.

Inputs / Outputs

use wickra::{Indicator, MatHold, Candle};
// MatHold: Input = Candle, Output = f64
const _: fn(&mut MatHold, Candle) -> Option<f64> = <MatHold as Indicator>::update;

• Always emits a value. Never None; warmup and no-match bars return 0.0.
• Node. update(open, high, low, close) → number; batch(open, high, low, close) → Array<number>.
• Python. update(candle) → float; batch(open, high, low, close) → 1-D numpy.ndarray (0.0 on warmup / no-match).

Warmup

warmup_period() == 5. The first four bars return 0.0 (first_four_bars_return_zero, accessors_and_metadata).

Edge cases

• Pullback must hold. A retracement deeper than penetration into the first body yields 0.0 (pullback_breaks_hold_yields_zero).
• Fifth bar must break out. Without a new high the result is 0.0 (no_new_high_yields_zero).
• Reset. reset() clears the four-bar cache (reset_clears_state).

Examples

Rust

use wickra::{Candle, Indicator, MatHold};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let mut t = MatHold::new();
    println!("{:?}", t.update(Candle::new(10.0, 15.1, 9.9, 15.0, 1.0, 0)?));  // long white
    println!("{:?}", t.update(Candle::new(16.0, 16.1, 15.4, 15.5, 1.0, 1)?)); // small, gaps up
    println!("{:?}", t.update(Candle::new(15.5, 15.6, 14.9, 15.0, 1.0, 2)?)); // small drift
    println!("{:?}", t.update(Candle::new(15.0, 15.1, 14.4, 14.5, 1.0, 3)?)); // small drift, holds
    println!("{:?}", t.update(Candle::new(14.5, 17.1, 14.4, 17.0, 1.0, 4)?)); // white breakout
    Ok(())
}

Output:

Some(0.0)
Some(0.0)
Some(0.0)
Some(0.0)
Some(1.0)

The three-bar rest gaps up and holds above the penetration floor, then bar 5 closes at 17.0, a new high — a mat hold. This matches mat_hold_is_plus_one.

Python

import numpy as np
import wickra as ta

o = np.array([10.0, 16.0, 15.5, 15.0, 14.5])
h = np.array([15.1, 16.1, 15.6, 15.1, 17.1])
l = np.array([9.9,  15.4, 14.9, 14.4, 14.4])
c = np.array([15.0, 15.5, 15.0, 14.5, 17.0])

print(ta.MatHold().batch(o, h, l, c))  # [0. 0. 0. 0. 1.]

Node

const ta = require('wickra');
const t = new ta.MatHold();
t.update(10, 15.1, 9.9, 15);
t.update(16, 16.1, 15.4, 15.5);
t.update(15.5, 15.6, 14.9, 15);
t.update(15, 15.1, 14.4, 14.5);
console.log(t.update(14.5, 17.1, 14.4, 17)); // 1

Streaming

use wickra::{Candle, Indicator, MatHold};

let mut t = MatHold::new();
let candle_stream: Vec<wickra::Candle> = Vec::new(); // your live OHLCV candle feed
for bar in candle_stream {
    if t.update(bar) == Some(1.0) { /* uptrend resumes after a shallow rest */ }
}

Interpretation

1. Shallow rest = strong trend. A pullback that gaps up and barely gives ground before a breakout is a high-quality continuation signal.
2. Penetration tunes strictness. A smaller penetration demands a shallower rest (stronger trend) to qualify.
3. Confirm with the trend. Continuation pattern; use within an uptrend.

Common pitfalls

• Deep pullback. A rest that digs too far into the first body fails the hold test — that is a deeper correction, not a mat hold.
• No breakout. The fifth candle must make a new high; a flat fifth bar does not confirm.

References

• Steve Nison, Japanese Candlestick Charting Techniques (1991).

See also

• RisingThreeMethods — the non-gap sibling.
• UpsideGapThreeMethods — gap continuation.
• Indicators-Overview — the full taxonomy.