Skip to content

Double Top / Bottom

Twin-peak / twin-trough reversal detected from confirmed swing pivots: two matching highs around a trough → bearish -1; two matching lows around a peak → bullish +1.

Quick reference

ItemValue
FamilyChart Patterns
Input typeCandle (uses high, low)
Output typef64 (+1 / -1 / 0)
Output range{-1.0, 0.0, +1.0}
Default parametersnone (swing threshold 5%, level tolerance 3%, baked)
Warmup period5
InterpretationReversal — confirmed on the second matching extreme

Formula

swing pivots are confirmed by a 5% non-repainting zig-zag (pattern_swing)
on each new pivot, look at the last three pivots p1, p2, p3:
  double top    : High , Low , High   with  High₁ ≈ High₃ (±3%)  → -1
  double bottom : Low  , High , Low    with  Low₁  ≈ Low₃  (±3%)  → +1
otherwise → 0

Two extremes count as the same level when they are within 3% of the larger magnitude. Because pivots strictly alternate high/low, the trough between twin tops (or the peak between twin bottoms) is beyond both by construction. See crates/wickra-core/src/indicators/double_top_bottom.rs.

Parameters

None. The swing threshold (SWING_THRESHOLD = 0.05) and level tolerance (LEVEL_TOLERANCE = 0.03) are baked-in constants shared across the family (crates/wickra-core/src/indicators/pattern_swing.rs), mirroring how the candlestick patterns bake in their geometric thresholds. DoubleTopBottom::new is infallible.

Inputs / Outputs

rust
const _: fn(&mut wickra::DoubleTopBottom, wickra::Candle) -> Option<f64> =
    <wickra::DoubleTopBottom as wickra::Indicator>::update;
  • Python. update((o,h,l,c,v,ts))float (0.0 until a pattern, never None); batch(open, high, low, close) → 1-D ndarray.
  • Node. update(open, high, low, close)number; batch(open, high, low, close)number[].
  • WASM. update(open, high, low, close)number.

Warmup

warmup_period() == 5. Three confirmed pivots are required and the earliest bar that can confirm a third pivot is the fifth; before that the detector returns 0.0. The metadata is pinned by test accessors_and_metadata.

Edge cases

  • Double top reports -1 (test double_top_is_minus_one).
  • Double bottom reports +1 (test double_bottom_is_plus_one).
  • Mismatched second peak reports 0.0 (test unequal_tops_do_not_trigger).
  • reset clears the swing state and history (test reset_clears_state).
  • Streaming equals batch (test batch_equals_streaming).

Examples

Rust

rust
use wickra::{Candle, DoubleTopBottom, Indicator};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let bars = [
        (119.88, 120.0, 119.88, 119.88),
        (100.0, 118.8, 100.0, 100.0),  // confirms the first top at 120
        (101.0, 120.0, 101.0, 101.0),  // confirms the trough at 100
        (108.0, 118.8, 108.0, 108.0),  // confirms the second top → double top
    ];
    let mut pat = DoubleTopBottom::new();
    let mut last = 0.0;
    for (ts, (o, h, l, c)) in bars.iter().enumerate() {
        last = pat.update(Candle::new(*o, *h, *l, *c, 1.0, ts as i64)?).unwrap();
    }
    println!("{last}"); // -1
    Ok(())
}

Python

python
import wickra as ta

bars = [
    (119.88, 120.0, 119.88, 119.88, 1.0, 0),
    (100.0, 118.8, 100.0, 100.0, 1.0, 1),
    (101.0, 120.0, 101.0, 101.0, 1.0, 2),
    (108.0, 118.8, 108.0, 108.0, 1.0, 3),
]
pat = ta.DoubleTopBottom()
print([pat.update(b) for b in bars][-1])  # -1.0

Node

javascript
const wickra = require('wickra');
const bars = [
  [119.88, 120.0, 119.88, 119.88],
  [100.0, 118.8, 100.0, 100.0],
  [101.0, 120.0, 101.0, 101.0],
  [108.0, 118.8, 108.0, 108.0],
];
const pat = new wickra.DoubleTopBottom();
let last = 0;
for (const [o, h, l, c] of bars) last = pat.update(o, h, l, c);
console.log(last); // -1

Streaming

python
pat = ta.DoubleTopBottom()
for o, h, l, c, v, ts in candle_feed:
    signal = pat.update((o, h, l, c, v, ts))
    if signal < 0:
        pass  # double top confirmed — potential short / exit longs
    elif signal > 0:
        pass  # double bottom confirmed — potential long / exit shorts

Interpretation

  1. Confirmation lag. The pattern fires on the bar that confirms the second matching extreme via the zig-zag, so it is non-repainting but lags the visual peak/trough by the threshold move. Treat it as confirmation, not prediction.
  2. Reversal strength. A double top/bottom is a classic reversal; combine the signal with the broader trend and the neckline (the intervening extreme) for an entry/stop framework.

Common pitfalls

  • Threshold sensitivity. The 5% swing threshold suits swing/daily data; on very low-volatility series few pivots confirm and the pattern rarely fires.
  • Not a target. The detector reports the reversal, not a price target — the classic measured move (height of the pattern projected from the neckline) is left to the caller.

See also