Enter on the 5-minute, but only when the daily agrees

Why this is hard on a bot platform

A bot instance is configured on a timeframe. You pick 5m, and every indicator in that bot reads 5m candles. There is no second field.

So a daily filter means a second bot on the daily — which cannot gate the first, because bots do not talk to each other. The workarounds are all bad: bake the filter into a TradingView script and lose it from the platform, approximate the daily with a very long moving average on the 5-minute (288 bars is not a daily candle and behaves differently at the open), or turn the bot on and off yourself based on your own read of the daily.

That last one is what most people actually do, and it means the filter — the part protecting you from a whole category of losing trades — is the part that is not automated.

The graph

Two sources, two clocks, one gate.

Both timeframes are sources in one graph.

Neither is "the bot’s timeframe", because a graph does not have one. Nodes read whichever stream is wired to them.

The gate uses the last fully closed daily candle.

At 14:35 today, today’s daily candle has not closed. Its "close" is just the current price. If your filter reads it, your 5-minute entry is being permitted by information from later in the day — which does not exist yet.

That is lookahead bias,

and it is the single most common reason a multi-timeframe backtest looks extraordinary and the live version does not work. The gate node is set to closed candles, so it reads yesterday’s daily close until today’s actually completes.

The trigger uses closed 5-minute candles too,

for the same reason at a smaller scale. A forming candle’s RSI crosses back and forth; a closed one does not.

Getting this wrong is how a strategy backtests beautifully and loses money live without a single line of logic being wrong — which is the whole argument on backtesting.

What you see while it runs

The daily permission is visible as a value, sitting true or false, with the EMA and the close that produced it. When a 5-minute setup fires and nothing happens, you can see that the gate was false — rather than wondering whether the trigger missed.

You also see the moment permission flips, on the daily close that flipped it.

Variations on the same shape

Gate the direction, not the trade.

Instead of blocking longs, let the daily decide whether the 5-minute trigger goes long or short.

Size by the slow chart.

Full size when both agree, half when the daily is neutral rather than aligned.

Slow exits, fast entries.

Enter on the 5-minute, manage the exit on the 1-hour, so noise cannot shake you out.

Three-way alignment.

Daily for trend, 4-hour for momentum, 5-minute for the trigger — the classic top-down structure, drawn as one graph.

Keep reading