TdCamouflage
Tom DeMark's TD Camouflage — a one-bar reversal that looks weak (or strong) on the close but reveals hidden accumulation (or distribution) intrabar.
Quick reference
| Field | Value |
|---|---|
| Family | DeMark |
| Input type | Candle (open / high / low / close) |
| Output type | f64 (signed signal) |
| Output range | {−1, 0, +1} |
| Default parameters | None (parameter-free) |
| Warmup period | 2 |
| Interpretation | +1 hidden-strength buy; −1 hidden-weakness sell. |
Formula
Buy (+1): close < close[-1] AND close > open AND low < low[-1]
Sell (−1): close > close[-1] AND close < open AND high > high[-1]
otherwise 0A lower close looks bearish, but if the same bar closed up from its open after dipping to a new low, buyers quietly absorbed the dip — accumulation "camouflaged" by the weak headline close. The sell pattern is the mirror. Source: crates/wickra-core/src/indicators/td_camouflage.rs.
Parameters
| Name | Type | Default | Valid range | Source | Description |
|---|---|---|---|---|---|
| — | — | — | — | None. | TD Camouflage is parameter-free; TdCamouflage::new() is infallible. |
Signed ±1 encoding
+1.0 = bullish (hidden strength) signal, −1.0 = bearish (hidden weakness), 0.0 = no pattern (including during warmup the output is None, then 0.0).
Inputs / Outputs
From crates/wickra-core/src/indicators/td_camouflage.rs:
use wickra::{Candle, Indicator, TdCamouflage};
// TdCamouflage: Input = Candle, Output = f64
const _: fn(&mut TdCamouflage, Candle) -> Option<f64> = <TdCamouflage as Indicator>::update;A Candle in, an Option<f64> out. Python update(candle) / batch(open, high, low, close) → 1-D ndarray (NaN warmup); Node update(open, high, low, close) / batch(o[], h[], l[], c[]).
Warmup
warmup_period() == 2. The first candle seeds the prior bar; the first signal lands on the second (first_bar_seeds_without_signal pins this).
Edge cases
- Bullish buy. A lower close that closed up off a new low fires
+1(bullish_camouflage_buypins this). - Bearish sell. The mirror fires
−1(bearish_camouflage_sellpins this). - No pattern → 0. Otherwise
0.0(no_pattern_is_zeropins this). - Finiteness.
Candle::newrejects non-finite fields, so no in-method guard is needed. - Reset.
td.reset()clears the prior bar and the last value (reset_clears_state).
Examples
Rust
use wickra::{Candle, Indicator, TdCamouflage};
fn main() -> Result<(), Box<dyn std::error::Error>> {
let mut td = TdCamouflage::new();
td.update(Candle::new(10.0, 11.0, 8.0, 10.0, 0.0, 0)?); // seed
let sig = td.update(Candle::new(9.0, 10.0, 7.0, 9.5, 0.0, 0)?);
println!("{:?}", sig); // Some(1.0)
Ok(())
}Output:
Some(1.0)Python
import numpy as np
import wickra as ta
td = ta.TdCamouflage()
o = np.array([10, 9]); h = np.array([11, 10]); l = np.array([8, 7]); c = np.array([10, 9.5])
print(td.batch(o, h, l, c)) # [nan, 1.0]Node
const ta = require('wickra');
const td = new ta.TdCamouflage();
td.update(10, 11, 8, 10);
console.log(td.update(9, 10, 7, 9.5)); // 1Streaming
use wickra::{Candle, Indicator, TdCamouflage};
let mut td = TdCamouflage::new();
let feed: Vec<Candle> = Vec::new(); // your live stream
for candle in feed {
match td.update(candle) {
Some(1.0) => println!("camouflage buy"),
Some(-1.0) => println!("camouflage sell"),
_ => {}
}
}Streaming update and batch are equivalent tick-for-tick (batch_equals_streaming pins this).
Interpretation
- Hidden reversal. Use it to catch reversals that close-only screens miss — the bar's internals contradict its headline close.
- Context. Strongest near support (buy) or resistance (sell), or at the end of a TD Setup count.
- Confirmation. Pair with a level (pivot, prior low) so the camouflage fires where a reversal is plausible.
Common pitfalls
- One-bar, noisy. Single-bar patterns fire often; filter by trend/level.
- Definition variants. Some sources add a close-two-bars-ago term; this uses the open/low (high) internals form.
- Not a standalone system. It flags a candidate, not a guaranteed turn.
References
Perl, J. (2008), DeMark Indicators, Bloomberg Press; DeMark, T. R. (1994), The New Science of Technical Analysis.
See also
- Indicator-TdDifferential — pressure-shift reversal.
- Indicator-TdClop — open/close reversal pattern.
- Indicator-TdSequential — the exhaustion count.
- Indicators-Overview — the full taxonomy.