Timeframes and Multi-Timeframe Analysis
The timeframe decides what is signal and what is noise. The same price series produces contradictory readings at different aggregations, and both readings are correct.
MadStockAlerts Research · Updated August 28, 2026
What to take away
- A timeframe is how much time each bar represents, and it is a choice, not a property of the stock.
- Higher timeframes carry more weight because more participation formed each bar.
- Contradictory readings across timeframes are normal, not an error.
- Levels drawn on a higher timeframe tend to matter on the lower ones too.
- Timeframe shopping always succeeds and always proves nothing.
- Higher and lower timeframes disagree routinely, and nothing in the chart resolves it: the holding period does.
- Bar boundaries are a convention, so the same trades can produce different candles on two platforms.
MAD Academy Training Video · 0:46
Three Charts, Three Verdicts
The same security can be in an uptrend, a downtrend and a range at once, depending only on which timeframe you opened.
This lesson is part of a Stock Alerts + Tools plan.
The same data, aggregated differently
A day of trading is one daily candlestick, sixty-five five-minute candles, or a fraction of a weekly one. Nothing about the underlying trades changed; only the size of the bucket did.
That choice determines what is visible. A three percent intraday swing dominates a five-minute chart and is invisible on a weekly one. Neither view is hiding anything; each is answering a question at its own scale.
It also determines what every indicator says. A 14-period RSI is fourteen days on a daily chart and seventy minutes on a five-minute one, and the two numbers have nothing to do with each other despite carrying the same name.
Scroll the chart sideways to see all of it.
Why higher timeframes carry more weight
A weekly bar aggregates five days of participation from every kind of participant. A one-minute bar can be formed by a handful of orders. A support level that has held for months on a weekly chart has been tested by far more decisions than one that has held for two hours.
The practical consequence is a hierarchy rather than a preference. When a daily and a weekly reading conflict, the weekly is describing a larger and slower structure that the daily sits inside. Neither is wrong; one contains the other.
A workable convention
| Purpose | Timeframe | What it is used for |
|---|---|---|
| Context | Weekly or monthly | The prevailing structure and the major levels |
| Structure | Daily | The pattern currently forming and its boundaries |
| Detail | Hourly or 15-minute | Where within the daily structure price currently sits |
| Execution | 5-minute or 1-minute | The immediate mechanics of getting filled |
The usual guidance is to keep the ratio between adjacent timeframes at roughly four to six. Jumping from monthly to one-minute skips the intermediate structure entirely, and the intermediate structure is where most of the useful levels live.
The trap
Timeframe shopping is looking through timeframes until one agrees with a position already held. It always succeeds, because on any given day some timeframe supports any conclusion. Choosing the timeframe before forming the view is the only defence.
It is a specific instance of confirmation bias with an unusually convenient supply of evidence, which is what makes it so common. The countermeasure is procedural rather than analytical: fix the timeframes in a saved layout and do not add one mid-analysis.
Multi-timeframe analysis, and what it resolves
The common convention is to read a higher timeframe for context and a lower one for timing: the weekly chart says whether the security has been making higher highs for a year, the daily says where it is within that, and the hourly says what happened this morning. Each answers a question the others cannot, and the reason the practice is widespread is that a decision made on one timeframe alone is a decision made without knowing what it sits inside.
The difficulty is that timeframes disagree constantly, and there is no rule that resolves the disagreement. A daily chart in a clean uptrend can sit inside a weekly chart that has gone nowhere for two years. Neither reading is wrong. They are descriptions of different windows, and the question of which one matters is a question about the intended holding period rather than about the chart.
The failure mode is adding timeframes until one of them agrees with the position. There are always more: 4-hour, 2-hour, 78-minute, 30-minute. Given enough of them, something is always in an uptrend, and a conclusion reached that way was reached before the charts were opened.
The usual discipline is to fix the set in advance, typically to three, and to fix which one is the context and which is the trigger. Three timeframes roughly a factor of four or five apart cover a wide range without overlapping much: weekly, daily and hourly is the standard triple for a swing horizon, and monthly, weekly and daily for a longer one.
One structural point is worth stating: a signal on a lower timeframe is not confirmation of a signal on a higher one, because both are computed from the same trades. Two views of one series agreeing is arithmetic, not evidence. What the higher timeframe genuinely adds is the size of the structure the decision sits inside, which is information the lower one does not contain.
Where the bar boundaries fall
A bar is defined by where its window starts and stops, and those boundaries are a convention rather than a property of the market. A daily bar on a US equity conventionally runs from the opening auction to the closing auction, so a trade at 4:15 in the afternoon belongs to that day only if extended hours are switched on. Change that setting and every daily high, low and close on the chart can change with it.
Intraday aggregations are less standardised than they look. A four-hour bar has to start somewhere, and the regular session is six and a half hours long, which does not divide by four. Platforms resolve that differently: some anchor to the session open and leave a short final bar, some anchor to midnight and produce bars that straddle the open. Two four-hour charts of the same security from two providers can therefore show genuinely different candles.
| Aggregation | Where it conventionally starts | What changes it |
|---|---|---|
| Daily | The opening auction | Including extended hours |
| Weekly | Monday's open, closing Friday | Holiday weeks, and providers that roll Sunday to Sunday |
| Monthly | The first trading day | Nothing, which is why monthly charts agree across providers |
| 4-hour | Session open, or midnight | The provider's anchor choice |
This matters mainly because a pattern defined by the shape of individual bars, a reversal candle above all, can exist on one provider's chart and not on another's. A structure defined by levels rather than by bar shapes survives the change, which is one practical argument for reading levels rather than candles when the two disagree.