Multi-Timeframe Analysis: Reading Charts From the Top Down

Reading more than one chart interval before acting is standard advice, and the standard version of it is thin. Look at a long chart for the trend, drop to a short one for the entry, done.

That leaves out the things that actually go wrong. Whether the signal you read on the long chart was finished. Whether your long chart is even the same as someone else’s. What to do on the common occasion when the two charts point in opposite directions.

What follows covers the method and those failure points, which is where the difficulty in practice actually sits.

What Multi-Timeframe Analysis Actually Is

Every chart of an instrument is built from the same stream of trades. A chart interval decides only how that stream is grouped into bars.

This has a consequence worth stating early, because it removes a common confusion. Two charts of the same instrument cannot disagree about what happened. They can only differ in what a given grouping makes visible. A move that fills a short chart may be a fraction of one bar on a long one, and both descriptions are correct.

Multi-timeframe analysis is therefore not a way of collecting second opinions. It is a way of seeing both the shape of a move and its detail, which no single grouping shows at once.

Once framed that way, the method’s real purpose is clear: use the grouping that answers each question best, rather than expecting one chart to answer all of them.

Working Top Down: Context, Setup, Entry

Working from the longest chart downwards is the usual order, and there is a reason for it beyond convention. Decisions made on a short chart cannot be checked against context you have not looked at yet, whereas context established first constrains what you go on to accept.

The longest chart establishes direction and marks the levels that have mattered. It answers whether you are looking to buy, to sell, or to stand aside, and nothing more precise than that.

The middle chart is where a setup either exists or does not. This is where a pattern forms, or a level is tested, in a way that could be acted on.

The shortest chart is for timing and for placing the stop. Its job is to improve the price you get and to define the level at which the idea is wrong, not to generate the idea.

The discipline is keeping those roles separate. An entry trigger spotted on the shortest chart that has no setup behind it on the middle chart is not a top-down trade; it is a short-chart trade with two other charts open. Our guide to price action covers reading structure without indicators.

Choosing the Spacing Between Your Charts

Most guides name a specific trio of intervals. The trio matters far less than the distance between its members, and the distance is the part that transfers to any instrument or style.

Charts too close together are the more common mistake. Adjacent intervals group the same trades in nearly the same way, so they show nearly the same picture. When they agree, which is most of the time, that agreement carries almost no information, and it is easy to read it as confirmation.

Charts too far apart fail differently. If one bar on the higher chart contains a great many bars on the lower one, the levels that matter above are too coarse to locate below, and the lower chart offers no useful guidance about where to act.

A separation of roughly four to six times from one chart to the next avoids both. That is a rule of thumb about resolution rather than a discovered constant, and it works because it keeps the charts related while ensuring each shows something the other cannot. The spacing has no meaning at all on charts that have no timeframe to align, because their bars are printed by distance rather than by the clock.

Two charts is a complete method. The third is useful when your entry is precise enough to need it, and rarely otherwise.

The Higher-Timeframe Candle Is Still Forming

This is the single most consequential point in top-down work, and it is missing from most treatments of the subject.

When you read a daily chart in the middle of the session, the most recent bar has not finished. Its close is simply the current price, and its high and low are only the extremes reached so far.

So a pattern that depends on where the bar closed is not yet a pattern. A bar that looks like a rejection of a level at midday can close well beyond that level by the evening, and the chart you review later will show nothing of what you acted on. The signal did not fail; it never completed.

This is also why top-down methods appear more reliable in review than in use. A finished chart shows only the signals that survived to the close, and gives no sign of the ones that formed and dissolved during the day.

Two workable responses exist, and the choice between them should be explicit rather than drifting. Treat higher-timeframe signals as valid only on a closed bar, accepting that you act later and sometimes miss the move. Or act on the forming bar deliberately, in the knowledge that a portion of those signals will not be there at the close, and size for that.

What does not work is reading the forming bar as though it were finished, which is what happens by default when nobody has made the decision.

Why Your Daily Chart Differs From Another Broker’s

There is a hidden assumption in all of this: that “the daily chart” names one object. It does not.

A trading platform takes its time from the trade server it is connected to, and MetaQuotes states this directly in the MQL5 documentation, noting that the time value is formed on a trade server and does not depend on the time settings on your computer. Bars are stamped with the start of the period they cover.

Since a daily bar covers the span between two points on that server clock, a server running one offset closes the day at a different moment from a server running another. The trades are identical; the grouping is not.

The practical effects are larger than they sound. Two brokers can show different daily opens, closes, highs and lows for the same instrument on the same date. A pattern that depends on the close, and most candle patterns do, can exist on one and be absent on the other. The number of weekly bars in a year can differ, because where the week is cut differs.

None of this is a fault, and no broker is wrong. But it does mean a level read from one platform is not automatically a level on another, and that a setup someone else describes may not be visible on your charts even though you are both looking at the same instrument. Check what offset your platform runs on before assuming a discrepancy is an error.

Session boundaries interact with the same issue, which our guide to forex market trading hours sets out in detail.

When the Timeframes Disagree

Guides describe the case where the charts agree, then stop. Disagreement is the normal condition, and it is where the method is actually tested.

Take the usual form: the higher chart is still rising while the lower one has turned down. That describes two different situations. It is a pullback within a move that will continue, or it is the beginning of the move ending.

At the time, nothing on either chart tells you which. They look the same until one of them resolves, and any method claiming to separate them reliably in advance is claiming more than the charts contain.

That leaves a small number of honest responses. Wait for the lower chart to turn back in line with the higher one, which is the ordinary way of buying a pullback and is covered in our guide to pullback trading strategies. Or stand aside until they agree, accepting you will miss some moves.

The response to avoid is the one that feels most natural: adopting whichever chart supports the position you already wanted. Having several charts open makes this easy, because on any given day one of them usually agrees with you. Deciding in advance which chart governs which decision is what prevents it.

How Many Charts Before It Stops Helping

Each chart added is supposed to add information. Past a point they mostly add scope for the problem just described.

Two is a complete method: one for context, one for execution. Three is reasonable when timing needs to be precise, with the middle chart carrying the setup.

Beyond three, the additions tend to sit close to charts you already have, so they repeat information rather than adding it, while multiplying the chances that one of them endorses whatever you were inclined to do. More charts also means more time between the decision and the action.

A useful test: if a chart has never once stopped you taking a trade, it is not contributing to the decision. Turning that judgement into something measurable means writing the roles as rules, which is where how to backtest a trading strategy becomes the next step.

Who Should Not Use This Method

Top-down analysis is a poor fit for anyone who has not written down which chart governs which decision. Without that, extra charts function as a supply of justifications rather than as a filter, and the method makes outcomes worse rather than better.

It is also a poor fit where the trading style does not have room for it. If positions are held for minutes, a weekly chart contributes nothing that will change within the life of the trade.

Anyone drawn to it because more charts feel like more certainty should be cautious. The charts contain one set of trades however many ways they are grouped, so no arrangement of them produces information the market has not yet generated.

What the method genuinely offers is context for a decision, not confidence in it, and position sizing still carries the risk, as our guide to risk management sets out. For stacking intervals in one window, see the multi-timeframe sync indicator.

Frequently Asked Questions

What is multi-timeframe analysis in forex?

It is the practice of reading the same instrument on more than one chart interval before acting, usually working from a longer interval down to a shorter one. The longer chart supplies context, such as the direction of the prevailing move and where the significant levels sit. The shorter chart is used to time the entry and place the stop. The charts are different summaries of the same underlying trades, not different markets.

Which timeframes should you use together?

The useful question is the spacing rather than the specific pair. Charts that sit close together largely repeat each other, so agreement between them means very little. Charts that sit very far apart leave the shorter one unable to show the levels the longer one is built on. A separation of roughly four to six times between one chart and the next keeps them related while still showing you different information.

Can a higher-timeframe signal disappear before the candle closes?

Yes, and this is the most common source of error in top-down work. Until a bar closes, its high, low and close are all provisional, so a pattern visible on a daily chart at midday may be absent from the same chart that evening. Either treat higher-timeframe signals as valid only once the bar has closed, or accept explicitly that you are acting on an unfinished bar.

Why does my daily chart look different from another broker’s?

Because the platform takes its time from the trade server rather than from your computer, and different brokers run servers set to different offsets. A daily bar covers the period between two points on the server clock, so a server on one offset cuts the day at a different moment from a server on another. The result is different daily opens, closes, highs and lows, and therefore different patterns on what appears to be the same chart.

What should you do when two timeframes disagree?

Treat it as information rather than a problem to resolve. The usual case, a longer chart still pointing one way while a shorter one turns against it, is either a pullback within the larger move or the start of its end, and nothing on the charts separates those at the time. Wait for the shorter chart to realign, or stand aside. Adopting whichever chart agrees with the position you already wanted is the response that reliably costs money.