Multi-Timeframe Analysis: Weekly vs Daily Charts
Use a higher time frame for direction and a lower one for timing. A real NKE case where the daily and weekly charts disagreed.
Contents
Every chart is a zoom level. The same stock can be in a downtrend on the weekly chart, bouncing on the daily, and falling again on the hourly, all at the same moment. None of these views is wrong; they answer questions about different stretches of time. Multi-timeframe analysis is simply the habit of checking more than one zoom level before you decide, and of knowing which one you let make the final call.
Choosing your time frames
A common guideline, popularised by Alexander Elder's Triple Screen method, is to space time frames by a factor of about four to six. Two or three are enough. More than that and you can always find one that agrees with whatever you already wanted to do.
| Style | Direction (higher) | Setup (middle) | Timing (lower) |
|---|---|---|---|
| Position / investing | Monthly | Weekly | Daily |
| Swing trading | Weekly | Daily | 60-minute or 4-hour |
| Day trading | Daily | 15-minute | 1- to 5-minute |
If you are just starting, the swing stack of weekly and daily is the easiest to learn. Daily candles give you time to think, weekly candles filter out a lot of noise, and neither requires you to watch the market all day. Very short charts are not a shortcut to learning faster: they show more candles, but a larger share of each move is random noise and spread.
When the charts disagree: NKE, June 2024
Nike had been falling for most of a year by mid-2024. Here is the weekly chart, built by combining each week's daily candles, with a 30-week simple moving average.
The weekly picture was clear. The week of June 10 closed at 93.39, below a 30-week average of 100.37, and that average had slipped from 102.53 in late April. Lower highs, a falling average, price underneath it: on this zoom level the burden of proof was on the buyers.
The daily chart told a more hopeful story. On June 17, NKE closed at 95.00, back above its 20-day average of 93.81, and the average itself was rising. Three sessions later it closed at 97.18. A trader looking only at the daily chart could reasonably see a base forming and a trend turning up.
Nike reported quarterly results after the close on June 27. On June 28 it opened at 77.13 against a prior close of 94.19, an 18.1% gap down, and closed at 75.37, 20.0% lower. The daily uptrend was gone in one bar.
What the weekly chart did and did not tell you
It would be easy to say the weekly chart "predicted" the drop. It did not. Nobody could know what the earnings report would say. What the weekly chart did was set the burden of proof. A daily long in a weekly downtrend is a trade against the bigger flow, and it deserves a smaller size, a tighter plan, and more caution about holding through an event like earnings. A trader who respected that would have been small or flat going into June 27.
The honest counterpoint is the same stock a year later. On June 10, 2025 NKE again closed above its 20-day average while the weekly trend was still down. Fifteen trading days later, on July 2, 2025, it was at 76.39, 19.6% higher than the 63.87 close. Trading against the higher time frame is not forbidden. It just starts with worse odds and needs a reason.
How big is the effect? A quick test
To check whether the weekly filter adds anything beyond these two stories, we ran a simple test on 16 large US stocks from 2021 to 2025. The signal: the daily close crosses above its 20-day average. We then split the signals by the prior week's trend, using the 30-week average and whether it was higher or lower than four weeks earlier.
| Weekly trend at the signal | Signals | Higher 15 days later | Median 15-day change |
|---|---|---|---|
| Up (above a rising 30-week SMA) | 505 | 55.8% | +1.03% |
| Down (below a falling 30-week SMA) | 376 | 50.3% | +0.28% |
The weekly trend tilted the odds by about five percentage points and the typical outcome by less than a percent. That is a real but modest effect, and signals in the same stock overlap, so the true difference is fuzzier than the table suggests. The lesson is not "only trade with the weekly trend". It is that a daily signal against the weekly trend is closer to a coin flip, and should be sized like one.
A top-down routine
- Start with the higher time frame. Mark the trend (higher highs and lows, or lower ones) and the one or two support and resistance zones closest to price.
- Write the bias in a sentence: "Weekly downtrend, price under the falling 30-week average, resistance near 100."
- Move to the middle time frame and look for a setup that agrees with the bias, or note clearly that you are trading against it.
- Use the lowest time frame only for the entry and the stop. It should never overrule the bias on its own.
- Check the calendar for scheduled events, especially earnings, before holding across them.
Mistakes that come from mixing zoom levels
- Taking the setup from the daily chart and the stop from the 5-minute chart. The stop ends up far too tight for the idea, and normal noise takes you out.
- Dropping to a shorter chart after a loss to "find" a better entry. That is usually impatience, not analysis.
- Reading an indicator on one time frame and price on another. A weekly moving average and a daily RSI are measuring different things.
- Treating agreement across time frames as certainty. When everything lines up, the odds are better, not guaranteed.
A good exercise is to take any daily chart, write down the weekly bias first, and only then look at the next few daily candles. It trains you to set the direction before the details pull you in.
Questions readers ask
What is the best time frame for beginners?
Daily candles, with the weekly chart for direction. They give you time to think, filter out a lot of noise and do not require watching the market all day.
How many time frames should I use?
Two or three, spaced roughly four to six times apart, for example weekly, daily and 60-minute. More than that makes it easy to find a chart that agrees with any decision.
Should I never trade against the weekly trend?
It is allowed, but the odds are worse. In our test, daily signals against the weekly trend were higher 15 days later only 50.3% of the time, versus 55.8% with it. Use a smaller size and a clear reason.
Sources
- StockCharts ChartSchool — Dow Theory
- StockCharts ChartSchool — Moving Averages (Simple and Exponential)
Charts and worked examples use daily prices from the One Candle Ahead historical dataset (split-adjusted). Past price behaviour does not predict future results.
Written and maintained by the Studio Solum Editorial Team. How we write and review guides