Which timeframe is better to use for the development of our trading systems: weekly, daily, or intraday?
I’m sure every systematic trader wondered this at least once in life.
Let’s see what are the strengths and limitations of each of them.
Today let’s talk about which timeframe we should use for the development of our trading systems.
Well, I’m Francesco Placci, I’m in charge of the research and development of systematic trading strategies for the Unger Academy.
I said “systematic” because what we do is to teach traders a way to become profitable, without spending hours in front of a monitor, but let’s talk about today’s topic.
Which timeframe is better to use to develop a trading system?
Daily, weekly, or intraday?
And what are the differences that derive from the use of these timeframes?
Weekly Time Frame: what systems is it good for?
Starting from the weekly timeframe, we can say that it is generally too slow for systematic trading, for our type of approach.
Obviously, there are exceptions, for example, the seasonal systems that work over long time horizons or asset allocation models that can use the weekly timeframe, also rotational trading systems on stocks often use this type of timeframe.
These are all systems that keep the positions open for several days, while generally speaking, our approach is more a short-term approach.
For this reason, in this video, we will mainly focus on intraday and daily timeframe.
Let’s focus on the intraday and daily time frame
The first thing that you notice when working in intraday, is more background noise.
This means that it is more difficult to find market inefficiencies because market movements are erratic.
For this reason, even when working in intraday, sensitive market points such as open, high, low, close, or volatility, are generally identified using the higher timeframe, which is the daily one.
In fact, the daily timeframe is cleaner because the background noise drops and it is easier to spot market inefficiencies.
As for the programming aspect, it is generally more difficult to program on an intraday timeframe.
This depends on the fact that, by referring to a superior timeframe, it is necessary to insert additional time series or to make use of more complicated programming constructs than a simple variable, such as, for example, arrays or lists.
There is, therefore, a greater difficulty of programming, while using the daily timeframe, the coding is much simpler and faster.
For this reason, we can say that the daily timeframe is more suitable for a quick view of the market to verify the presence of an edge or to deepen the knowledge of a specific market.
The other side of the coin is that the intraday timeframe, allows us to have more precise, more reliable backtests.
Why are there inaccuracies in using the daily timeframe?
This generally depends on the fact that the platform is not able to understand how the prices have moved within a single bar.
Did a stop-loss or a take-profit come first?
In these cases, it is possible to use a Multicharts feature, called Bar Magnifier.
This feature allows Multicharts to check how prices have moved within the single daily bar and this allows you to increase the accuracy of the backtest, even using the daily timeframe.
This is a very important aspect because it is a very common mistake among novice traders.
A practical example
Let’s take an example.
This is a simple trading system that enters the market when prices break the five-day high or low.
The positions are closed at the end of the day or using a take profit or a stop loss.
This is the beautiful equity line of the system which seems to be very effective, but let’s see what happens if we activate the bar magnifier feature.
Now we are telling Multicharts to check how prices have moved within a single bar.
Let’s take a look at the equity line.
As you can see it is a very different picture.
So please, do not underestimate these aspects, especially if you work on a daily timeframe and you don’t know the suitable stop-loss or take profit for the underlying.
Another type of error that can occur when using the daily timeframe, refers to the close of the bar.
Last market price or settlement price?
Does this correspond to the last market price or to the settlement price?
When we build trading systems, we have to rely on real prices.
We know instead, that the settlement price is unreal, it’s a sort of mathematical calculation and therefore we cannot use it.
A backtest based on the settlement price will return unreliable results that do not correspond to reality.
A positive aspect of using daily bars is the speed of code execution.
The platform has to process fewer data and therefore all processes will be faster, such as the optimization process.
So obviously, if we work on intraday, we have to expect longer processing times.
However, by working in intraday, we have the possibility to test conditions that will not otherwise be possible to test if working on a longer timeframe.
I refer for example to intraday biases.
The prices tend to rise or fall in certain time bands.
In the same way, by using daily timeframe is not possible to use patterns that use conditions deriving from the current prices, such as the high of the current day.
It is not possible to test the multiple entries or exits and it is also more difficult to use custom sessions.
So, in this case, the intraday timeframe gives us a big advantage allowing us to test and use conditions that often bring important benefits to our trading systems.
Generally speaking, we can say that the systems that use daily bars are generally swing trading systems, that stay in position more than a day, that works mainly on stocks, but also sometimes on futures.
The intraday timeframe is instead mainly used on futures, due to the leverage effect which makes it preferable to stay in position for the shortest time possible.
For this reason, it is suitable for low capitalized traders because by staying in position for a short time, allows you to reduce drawdowns.
In fact, we know that both the average trade and the drawdown increase in the proportion to the increase of the timeframe.
Therefore, daily systems have a greater average trade and, at the same time, a greater drawdown.
For this reason, they are generally used by well-capitalized traders.
So having said that, what are the conclusions?
I personally believe that if you don’t have a good knowledge of the market on which you want to develop a trading system, starting from the daily timeframe, it can be very useful to understand the dynamics of the market and find usable edges.
Anyway, once you find something interesting, you need to move to the intraday timeframe to have a more reliable backtest and at the same time, to test those conditions that cannot be verified using the daily timeframe, especially if your intention is to keep the positions opened for the shortest time possible because you are not building a swing trading system.
I hope you enjoyed this video and if you have any questions don’t hesitate to leave a comment and we will reply as soon as possible.
Thanks for watching