Why do traders still use lagging indicators?

[dartmus]

Remember: this thread is about lagging indicators, not TA.

The common ways of thinking about lag aren't helpful. It's better to think of everything with the same time stamp as centered and equal ...neither leading or lagging. A moving average based on a length of 200 historical bars neither leads nor lags behind price. Averages are simply variables. The only thing that can change whether they lead or lag price is adjusting the index of either the average or of price ...forward or backwards ...in time ...relative to their timestamps.

It's necessary to view lag in this skewed manner because the net profit or loss result within a strategy performance report is based on a correlation between a leading signal and centered price where centered simply means the index or timestamp of price is unadjusted because brokers don't allow us to tamper with the price index.

This skewed definition of lag enables adjusting the index of the independent variable within a correlation equation. Does anybody not understand this is the essence of the proper way to begin thinking about the difference between lead and lag? All that's necessary to define a leading indicator is to reference the previous value of the independent variable.

 

[bellman]

If even a single trader uses "lagging indicators," then those indicators are not lagging, but predictive indicators. IOW, there is no such thing as a lagging indicator.

 

[wizard4444]

I will post one last time in this thread my 2 cents, cause I do find it funny that there is a discussion like this. I do not know witch of you are professionals witch are newbies and witch are just trolls and wannabes but I do know 1 thing 100%.... any trader that is a professional and has traded in institutions long enough knows that there is no wrong or right way to make money in trading.... weather it is lagging, or leading.... not all traders see things the same way or interpret them or use them the same way.....as long as your account is making money it is good....even if you are using your daughters crystal ball and do not know if it is leading or lagging... but I do not care if that ball is making the money.

 

[romik]

Good comparison to averages, based on past performance, yet acts as forecasting tool.

 

[Maverick74]

I have never read a thread in 14 years that I have been on ET full of more total and utter crap. Did someone actually say moving averages are forward looking? OMFG. Go ahead and bash away. Do you guys actually read the stuff that you post?

 

[wrbtrader]

Everything is lagging unless you're using automation.

Thus, if you're not using automation and you're not using indicators...whatever you're using...its lagging due to the nature of the time elapse it takes a human brain to process information and then make a decision and then execute that decision.

That time is 5 - 7 seconds even though the initial process (the starting point) is in 13 - 80 milli-seconds. In comparison, algorithms can start and finish the process (complete a trade) in milli-seconds.

Simply, although I don't use technical indicators (I only use price on the chart), whatever I see on that chart and by the time I understand that info and then execute a decision...5 - 7 seconds has already elapsed.

In addition, whenever the human mind see an image (charts, DOM, price quotes)...its not able to process 100% of that information. We lose some of it or its mixed up with something else we saw seconds earlier and then we make trade decisions on the remainder information.

Throw in the issue of decisions about money and some emotions...everything is lagging even for people that don't use indicators or don't use technical analysis. The scary part is when neuroscience and psychology meets each other. They conclude that "decision making" for us humans is not logical. Instead, our decisions being made in trading (charts or no charts)...its an emotional response mixed in with some behavior response. The latter (emotional/behavior response) is why Wall Street is spending almost 700 million dollars per year to better understand their traders and to ensure their top traders have psychological help whenever needed if their mindset changes...resulting in poor performance.

Note: I personally believe that trading decisions is a mixture of logic and emotions. Most traders lose because the emotions and behaviors are impacting the logic too often. Therefore, most traders need to reprogram themselves so that its the other way around.

The good news is that the human brain can be reprogram to make better decisions when under pressure...in those 5 - 7 seconds. Its difficult and the reason why so few traders succeed in the business of trading. Thus, the next time you think about lagging indicators...you need to realize the human brain itself has a lag in its decision making process.

Behavior Finance
Neurosciences of Decision Making

P.S. When I say automation...I'm not talking about mechanical alert systems. I'm talking about true automation from entry to exit in which the computer trades for you after you've programmed it to do such. Too many traders in error believe mechanical alert systems is automation trading.

 

[romik]

I have never read a thread in 14 years that I have been on ET full of more total and utter crap. Did someone actually say moving averages are forward looking? OMFG. Go ahead and bash away. Do you guys actually read the stuff that you post?

They can be and you can't disprove it. They can point to strength of a trend, they can point to trend change (death crosses) and they act as anchors. They don't work all the time, nothing does. You're welcome to disprove that notion.

 

[Maverick74]

They can be and you can't disprove it. They can point to strength of a trend, they can point to trend change (death crosses) and they act as anchors. They don't work all the time, nothing does. You're welcome to disprove that notion.

Of course it can be disproved. It's called the scientific method. We have ways of measuring statistical significance for any coefficient you can dream up and test whether that coefficient can actually explain the variation in the data.

 

[romik]

OK, let's have a scientific disproval of death crosses. You would probably point to low probability hence absence of an edge leading to the conclusion death crosses don't work. Correct?

 

[Maverick74]

OK, let's have a scientific disproval of death crosses. You would probably point to low probability hence absence of an edge leading to the conclusion death crosses don't work. Correct?

No. Listen, I get it this is a little over your head so I don't want to waste a whole lot of time on something that you can google and read all about on the net on your own time. It has nothing to do with probability, win/loss ratios or even p&l. It has to do with proving that some variable in which you control for as many other factors as you deem necessary that could affect the output and show statistically that there is a significant relationship between your variable and the output say at the 95% confidence level. Put another way, you want to show that the outcome did not come by chance. The 95% confidence interval is saying there is only a 5% chance that the result you got could have been random. Honestly, there are 1000's of pages on the web that will walk you through this, some of it is quite good. OK? Give it a read.