Tool
Momentum Confluence
calc.momentum_confluence
Inputs
The first yes/no condition that contributes to the tally (e.g. trend is up).
The second yes/no condition (e.g. momentum is positive).
The third yes/no condition (e.g. price above VWAP).
The fourth yes/no condition (e.g. volume confirms the move). Leave unwired if you use fewer votes.
The fifth yes/no condition (e.g. higher-timeframe bias agrees). Leave unwired if you use fewer votes.
Outputs
A confidence line from 0 to 1: the fraction of wired votes that are true on this bar. Threshold it (e.g. ≥ 0.6) to require a majority before trading.
Momentum Confluence counts how many of up to five yes/no ‘votes’ you wire in are currently true, then turns that tally into a confidence score from 0 to 1. It is a simple way to require several signals to agree before acting.
How it works
Each input `vote_1` through `vote_5` is a boolean condition; the node counts how many are true and divides by `num_votes` to get a 0–1 confidence. Set `num_votes` to how many votes you actually wired so the score reflects the real denominator — three of three is full confidence, three of five is partial.
Configuration
How many of the five vote inputs are actually in use — the denominator for the confidence score. If you wire three conditions, set this to 3 so ‘all three true’ reads as 1.0 rather than 0.6.
Connects to
Examples
- Entry gate: wire trend, momentum, and location checks into three votes, set num_votes to 3, and only enter when the score reaches 1.0 (all agree).
- Confidence sizing: feed the 0–1 score into the order’s confidence so position size scales with how many votes line up.
Common mistakes
- Leaving num_votes at 5 while wiring only three conditions, so the score tops out at 0.6 and a ‘≥ 0.8’ gate can never fire.
- Wiring votes that all measure the same thing — five flavours of one trend filter give false confidence; pick genuinely independent signals.