Discretionary Trading vs Automated Trading

[Soultrader 12]

With the advancement in technology and improvements in trading softwares, many intraday traders are now automating their trading. There are advantages to this... emotions can be eliminated (not completely but more than discretionary) and traders can use mechanical exit points. Example: scaling out at +10pts on every trade.

 

As a pure discretionary trader, I still believe there is an edge in discretionary trading. Intuition comes into play as well as better entries. Why better entries? If one is a good tape reader he can pinpoint his entries and take minimal heat. This can save a trader a ton of points in the long run.

 

I have a couple questions regarding this topic and could use your guys opinion. Do you think discretionary traders enter before automated traders? What are your thoughts on discretionary vs automation? What type of trader are you and which one do you prefer?

 

Thanks,

 

Soultrader

[ant 10]

Do you think discretionary traders enter before automated traders?

 

I think this depends on how aggressive a trader is. For example, if a trader is looking to fade a key support/resistance level, does the trader enter while the market is moving towards support/resistance or after the market turns? Take another example, let's say a trader is looking to play a reversal from a downtrend, does the trader wait for the higher high and higher low to be in or does the trader anticipate that the higher high/low is coming based on somthing else? I don't think one can make a blanket statement here. I believe it comes down to experience and intuition as opposed to a mechanical or discretionary style.

 

What are your thoughts on discretionary vs automation? What type of trader are you and which one do you prefer?

 

I'm a discretionary trader for two main reasons. First, there are too many variables that I look at to capture in a mechanical system. Secondly, a mechanical system tends to be too rigid. For example, let's say that the market is a tick away from an entry/exit and then turns around, the entry/exit would not trigger on a mechanical system. I don't think that trying to "program" a discretionary style into a mechanical system makes too much sense. Think of the complexity.

 

Of course, one of the advantages of a mechanical system is that it is consistent in the way it executes. Discretionary traders need to make sure that their trade decisions are made in consistent and systematic way so that they can continue to replicate in the future and determine what has been working over the longer term. Discretionary traders should not let too much subjectivity into their trading, otherwise they don't have a system. Maintaining a trading log and journal are critical to success for a discretionary trader, IMO.

[torero 10]

I agree. This is why many successful system traders advise to keep their system open and flexible, the less variables the better. This is also why higher timeframes work so well vs. intraday trading. I have 3 working systems and they do a nice job of capturing the essence of the market direction with only 1 indicator and price is the other indicator over time.

 

I discretionary trade because my mission is to understand the markets as much as possible before getting more into system trading. I want to know how the market works for myself before building more. Took me 2 yrs to come up with 3 systems so I figure my learning curve in building them would increase as my discretionary gets better.

[enlightedtrader 10]

I agree. This is why many successful system traders advise to keep their system open and flexible, the less variables the better. This is also why higher timeframes work so well vs. intraday trading. I have 3 working systems and they do a nice job of capturing the essence of the market direction with only 1 indicator and price is the other indicator over time.

 

I discretionary trade because my mission is to understand the markets as much as possible before getting more into system trading. I want to know how the market works for myself before building more. Took me 2 yrs to come up with 3 systems so I figure my learning curve in building them would increase as my discretionary gets better.

 

Hi torero,

 

I'm curious on average how many trades you get daily in your intraday trading system. From my understanding, it is difficulty to get more than 3 trades a day for an automated system. This is also one of the aspect I prefer discretionary trading over automated trading. A discretionary trader has the flexibility to trade by adjusting to different market conditions. He/she can be both a scalper and a swing trader. I'm not sure whether automated system have such a flexibility.

 

Another aspect I prefer discretionary trading over automated trading is lower drawdown. I read something about the drawdown of automated system. It usually ranges from 20% to 50% annually. It's just too high, IMO. A good discretionary trader can have much lower drawdown as long as he has a sound trading method and good money management.

[Guest scalptrader]

I'm curious on average how many trades you get daily in your intraday trading system. From my understanding, it is difficulty to get more than 3 trades a day for an automated system. This is also one of the aspect I prefer discretionary trading over automated trading. A discretionary trader has the flexibility to trade by adjusting to different market conditions. He/she can be both a scalper and a swing trader. I'm not sure whether automated system have such a flexibility.

 

See the attachment. 100% automated, even the execution.

 

I trade a portfolio of systems, no discretionary trader can come close to duplicating these results. From a conceptual view, the human mind could never find such systems, (I don't use AI). And from an execution view, any trader would burn out after a couple months trading something like this manually!

 

I'm a discretionary trader for two main reasons. First, there are too many variables that I look at to capture in a mechanical system.

 

Yes there are so many variable, but dont forget the simple stuff always works better, especially for discretionary trader. You should be using no more than 3 non correlating technical reason reasons to enter a position. Anything more and confusion starts to creep in...

 

The main problem most trader face is they lack the technical skills required to implement automated systems. Start learning programming if you are serious!

[enlightedtrader 10]

See the attachment. 100% automated, even the execution.

 

I trade a portfolio of systems, no discretionary trader can come close to duplicating these results. From a conceptual view, the human mind could never find such systems, (I don't use AI). And from an execution view, any trader would burn out after a couple months trading something like this manually!

 

 

 

Yes there are so many variable, but dont forget the simple stuff always works better, especially for discretionary trader. You should be using no more than 3 non correlating technical reason reasons to enter a position. Anything more and confusion starts to creep in...

 

The main problem most trader face is they lack the technical skills required to implement automated systems. Start learning programming if you are serious!

 

scalptrader:

 

The chart from your system looks very interesting. It seems to me you are using some simple cross average at short time frame in your automated trading system. I'm wondering how the system performs at choppy market conditions.

 

Do you leave your system operate overnight or you just have it operated during regular trading hours? I guess sometimes the overnight spike may bring big trouble for an automated system if the system is not designed very well in this aspect.

 

Since you said that ''...no discretionary trader can come close to duplicating these results...'', I'm curious to know what's the return/drawdown of your totally automated system during its life span. Frankly speaking, I really doubt it can beat a good discretionary trader.

 

Thanks.

[Guest scalptrader]

From a conceptual view, the human mind could never find such systems, (I don't use AI). So there is no point trying to guess. The rules will not make sense to you any way either!!

[enlightedtrader 10]

From a conceptual view, the human mind could never find such systems, (I don't use AI). So there is no point trying to guess. The rules will not make sense to you any way either!!

 

Scalptrader,

 

I don't want to know the rules. I just want to know the results (return, profit/loss ratio, drawdown, etc.) if you are willing to share. No matter what approach/method/system adopted, ultimately only the results can be the final judge.

[torero 10]

Hi torero,

 

I'm curious on average how many trades you get daily in your intraday trading system. From my understanding, it is difficulty to get more than 3 trades a day for an automated system. This is also one of the aspect I prefer discretionary trading over automated trading. A discretionary trader has the flexibility to trade by adjusting to different market conditions. He/she can be both a scalper and a swing trader. I'm not sure whether automated system have such a flexibility.

 

Another aspect I prefer discretionary trading over automated trading is lower drawdown. I read something about the drawdown of automated system. It usually ranges from 20% to 50% annually. It's just too high, IMO. A good discretionary trader can have much lower drawdown as long as he has a sound trading method and good money management.

 

It's true that my systems don't trade everyday despite using intraday charts. The conditions to trade only when the direction is already confirmed. I use the same money management in these systems as my discretionary trading, so any DDs would be the same. I aim to create systems that can emulate my style and risk tolerance else I'd never trade them. This is the reason why there are many systems out there that seem profitable but the DDs are so large, I don't have the guts to withstand the DDs. I have to pass them up.

 

This is the difficult part in system trading is that no system can trade more than 5 trades a day and stay profitable. Slippage and commissions will eat it alive. The feedback from traders in TS forums have confirmed this. Higher timeframes have better chances of lasting longer and be profitable. So less trades greater profitability.

[Soultrader 12]

I think this depends on how aggressive a trader is. For example, if a trader is looking to fade a key support/resistance level, does the trader enter while the market is moving towards support/resistance or after the market turns? Take another example, let's say a trader is looking to play a reversal from a downtrend, does the trader wait for the higher high and higher low to be in or does the trader anticipate that the higher high/low is coming based on somthing else? I don't think one can make a blanket statement here. I believe it comes down to experience and intuition as opposed to a mechanical or discretionary style.

 

 

 

I'm a discretionary trader for two main reasons. First, there are too many variables that I look at to capture in a mechanical system. Secondly, a mechanical system tends to be too rigid. For example, let's say that the market is a tick away from an entry/exit and then turns around, the entry/exit would not trigger on a mechanical system. I don't think that trying to "program" a discretionary style into a mechanical system makes too much sense. Think of the complexity.

 

Of course, one of the advantages of a mechanical system is that it is consistent in the way it executes. Discretionary traders need to make sure that their trade decisions are made in consistent and systematic way so that they can continue to replicate in the future and determine what has been working over the longer term. Discretionary traders should not let too much subjectivity into their trading, otherwise they don't have a system. Maintaining a trading log and journal are critical to success for a discretionary trader, IMO.

 

 

Thank you Ant. I have yet to go on a mission to designing a pure mechanical system. The methods I have learned and apply in trading are far too complicated to program. How would one go about programming tape reading? Pretty impossible I'de say. At the same time discretionary traders need to approach the markets in a very systematic way: Trading off setups, controlling losses, target points, etc...

[torero 10]

Believe or not, soul, I've seen some code in the TS forum and has been shown by other traders using internals to signal and time their trades. Nothing is impossible that cannot be done in code. Just a matter of how skillful or skilless the coder is. Check around the forum and you can find some nice nuggets to get your ideas going.

[Soultrader 12]

Very interesting. I do not come from a comp, engineering background and I have absolutely zero knowledge when it comes to coding. I find Easy Language extremely hard.

 

How would one go about learning coding? Is this something that will take years to become skilled at? Is learning Easy Language enough? Thanks

[walterw 10]

Well... systems is like Formula 1 races... if its raining you need certain types of tires if not other types... thats also true with systems... you design a system for a market climate... if you are good analist you even design an alternative system for a diferent climate.... the big problem comes in discerning wich climate are we having and wich sistem should be implemented... thats where discretionary traders have an edge... they do know and feel the climate in order to adapt to it... systems dont discern market conditions.... some naive traders put conditions to their systems to filter market climate ,believe me there is no rule , no specific hours, no nothing to market conditions... market can go wild on lunch hour as it can get sleepy on the first 30 minutes before a news... or be wild without news , there is no parameters as to market climate, only a seasoned trader in front of his screen can tell you the market conditions an wich of his alternative techniches can be applied in that conditions...

[Anna-Maria 10]

Very pertinent points Walter. I agree, unless the trader or fund can bankroll their automated system with deep pockets, the drawdowns will freak most folks out.

 

As you say, the ever changing climate dilutes the vast majority of these programmed systems, well certainly in the FX environment anyway.

 

Obviously, there are well capitalized firms (excl the large corporates) & individuals out there who can & do run successful forays into the automated streams - but they're few & far between. For the average market participant, total automation is a tricky prospect.

[torero 10]

I got a bachelors in international marketing. Just nights studying to become a webmaster when webmasters belonged in the marketing department. From there it went on. It took a few months of reading, doing tutorials etc. Matter of time, just like how one would start on trading I guess. I got to know EL with a tutorial and manual in no time.

 

Again the less variables and rules the system is the more robust it is. That means with different conditions, it manages to survive and thrive in it because there are little rules to thrive in one and die in another. So, in truth, the good strategy doesn't have to be so sophisticated or difficult to code. Know the market and code should come pretty well placed.

[torero 10]

Most system traders (successful ones) admit no system will last forever. Hence there is always continuing search for the new systems. Even robust ones need tweaking, revisions, etc from time to time. Same as discretionary traders, they need tweaking (bullish to bearish mindset etc) as new conditions arise. This also explains why these systems for sale out there don't make money. Either too many people trading it making the market absorbing the strategy into failed ones or just don't last long enough when changing conditions become evident.

[Soultrader 12]

Thanks for the comments Walter, Texxas, and Torero. Some very interesting points. Im very new to this system world... which is why I am still very skeptical with trading systems. I have been taught by discretionary traders only and seeing how successful they are, I always believed discretionary traders make the best traders.

 

Very good point on the funding part Texxas. Deep pockets is definitely a consideration especially with some of the drawdowns I keep hearing with system based traders. One of the biggest concern I have as a discretionary trader is that more and more traders are creating systems and being automated. Will the edge of a pure discretionary trader ever fade away? The market needs to be full of emotion for traders like myself to survive.... without emotions, I wouldnt know what to exploit. (of course I dont believe this will ever happen but could affect price movement?)

[Anna-Maria 10]

On the subject of "mechanical trading systems" & the pro's-con's etc:

 

One of the guys in our office zapped this missive across to me late last night.

 

I won't reveal the blog/trader, it's immaterial - however, it raises the complexities of this type of trading when the template/mechanics are not built & managed within a solid psychological and/or physical framework.

 

I think the (abridged) commentary say's it all:-

 

 

Firstly to report, results which are rather shitty, and second to suspend / close the blog, possibly forever. Reasons to follow.

 

First of all lets talk turkey about the results.

 

The month was terrible - in fact the worst that we have experienced in the 14 months trading it, however not the worst on record from our tests.

 

We suffered this month a 48% drawdown. We suffered this drawdown at a point when i was supposed to be enjoying a holiday in New York - it was 30% when I left the UK and 45% when i got back. Needless to say due to things beyond my control, much of my holiday was totally spoilt.

 

The month of December ended up at -45% on the account - from £100,502 down to £55,000 The system still works and is within parameters but the money management, which at the end of the day is the bit that creates the severity of drawdown, was at fault and has since been rectified.

 

We were running the account at 100% efficiency - which meant that at a max drawdown there would just be enough in the account to sustain it. what i didnt take into account was the psychological pressure of actually getting close to reaching this level and seeing the account decimated.

 

Even after five years of trading, the market can still humble you, and this drawdown certainly has. So if you are reading this and you're a trader - dont ever think you have this thing licked.

 

To this point, i hope that you've enjoyed the journey and it's with sadness that I am to close this blog.

 

There are a few reasons for my decision, some of which I cannot talk about at the moment, however the main one is that because of the blog I am getting emotionally involved when the system has a losing month. The fact that it's in a drawdown at the end of a month shouldnt matter because it peaks and troughs no matter what the date is - the fact it's at the end of the month is immaterial.

 

Getting emotionally involved in the trading of a mechanical system is not good - you start to analyse every facet of the trades and it can drive you nuts.

[torero 10]

I totally agree. Trading mechanical doesn't really take away the fact that you still can pull the plug or manipulate it during the trades, etc. This intervention is what makes the system worst and not stay with the performance as expected.

 

As I said before, my system were built on money management first. They must fit my trading style and tolerance level. All the systems I've put it on trading has then 20% maximum drawdown. So I know I'll pull the plug on it when it reaches there. It's a benchmark point where the system will have stopped working.

 

Most people don't understand that systems can handle drawdowns on its own, it's the owners or traders who trade them that can't handle them. This is the other reason why renting them can be an emotional panic when you don't know in full details what the system is capable of doing and not doing. I always judge the systems' soundness by seeing if any money management is in place, then see if it's within my emotional tolerance.

 

I learned this very quickly after deciding to go mechanical with the original belief that no more emotions will be involved. Nothing can be further from the truth.

[momentom 10]

The biggest issue in mechanical systems for me is EXECUTION. Slippage is an enormous problem as is the system getting out of sync with the actual fills.

 

My systems trade about 20 times a day.

[Guest scalptrader]

Who is your broker? You tried just using limit orders?

[momentom 10]

Problem with limit orders is that the market trades there but you don't get filled. This means one of two things:

Your system thinks you are filled and you are not, or,

You miss many profitable trades and get all the losers whick makes your back testing useless.

[torero 10]

The biggest issue in mechanical systems for me is EXECUTION. Slippage is an enormous problem as is the system getting out of sync with the actual fills.

 

My systems trade about 20 times a day.

 

From testing out many systems on my own. Slippages and commissions must be set as realistic as possible. Limit orders will hide the fills in backtesting but in real trading, you'll get bad trades and good trades will be missed (everyone is out to get that fill too). 20 trades/day is alot of trades, comms and slippages will eventually eat up your profits long term.

[ezduzzit 10]

Hrrummpff! A very interesting thread, to be sure. I think Ant, Torero and Walter nailed it all down pretty well. I am probably older than most of you and I seem to recall LTCM (Long Term Capital Management) putting together billions of dollars, the best tacticians, the most powerful computers and coders who could probably program rockets to Mars and beyond.. in an attempt to use their edge in automated system trading and clean house.

 

Well, house was cleaned all right. Despite all the best resources available, not only were they unable to win more than they lost, they collapsed in spectacular style soon thereafter.

 

In other words, boys and girls, it ain't worth it, but I know that won't discourage most of you from attempting it nevertheless. Of course marketers like Modus and others will pronounce that the powers that be have taken a factual survey and 75% or more of the pros and big institutional players succeed only by using an automated system. Well, lest ye forget, they can afford some very significant drawdowns, which are precisely what kill most of the retail traders in their valiant attempts to play like the big boys. Additionally, they have a full cadre of analysts, computer whizzes, etc.. on hand at all times running up to date scenarios with informational access that you can still only dream about. Additionally, until they collapse, you never hear about their whopping losses or their real ROI, etc.. so best to stick to your own knitting and for most, that is going to still come down to being a discretionary trader. When the computers can do it all without effort on your part, the game will be over or they will change all the rules anyway.

 

Happy Trading

[Pismo10 10]

You have to design these backtest limit orders so they only get filled when the market moves one tick past your limit. You rarely sell at the ask or buy at the bid. Certainly nothing you can depend on. This has to be taken into account when programming. Tradestation/Easy Language does not take this into account, you need to program it yourself. Tradestations ProftTarget strategy will fill limit orders on a touch which is not realistic at all.