Mechanical Trading possible ?

[new_trader]

As to performance a good mechanical system will do very well, the last 3 years i have returned over 100% each year even though at times some markets have been in drawdown.

But this is utterly meaningless and you have completely missed my point. If you started with US$5000 then it means at the end of the year you have made 100% or $5000, right?

5000/52 = US$96/ week. Is that enough to live off?

Yes there is no certainty with a mechanical system but so is there for Discretionary !,

This shows that you do no understand discretionary trading and what I mean by getting better with time and experience.

 

[Pat494]

I don't know of a single out-of-the-box system that you can plug in and make profits. They all ( that I have come across) wriggle out of this awkward question by saying the customer has to use his own discretion as well. Of course if the given trades are not doing too well it's the customer's fault !!

At the really top end of the cost scale it may be different but I haven't trusted them with my hard earned lolly.

The do-it-yourself programming lot often spout tales of mega percentages on their own systems. Makes me wonder if it is really so //////////////??

The very costly and difficult to use Market Matrix system a few years ago had admirers but blew up under pressure apparently`

Whatever happened to the Quants ? THE whizkids of the 90s ?

 

[new_trader]

The do-it-yourself programming lot often spout tales of mega percentages on their own systems. Makes me wonder if it is really so?

I would say not unless they are adding discretion in which case it completely invalidates their argument for 'mechanical' trading. ALL mechanical systems are in the same class which basically means if you have thoroughly back tested a few then you have basically back tested them all. I despair for those who think they have a good mechanical system because it has produced spectacular results over a meagre 3 years. This is absolutely possible and believable because I meticulously back tested many systems and they produced these kinds of results over a few years. It’s when you test the results over 20 years that you will be horrified and realise that mechanical trading is what I say it is - a pension scheme (IF you are lucky) and nothing more. I doubt anyone here boasting about how great mechanical trading is would gladly continue trading it every single day, taking every single signal their system gives for over 2 years just to get OUT of drawdown and back to their original equity level. I doubt they would, I doubt it very much. There would be too much temptation to tweak the original system, the very one that they boasted about in a forum because it gave 100%p.a over 3 years, yes that one!

 

[ZEPPO]

"The very costly and difficult to use Market Matrix system a few years ago had admirers but blew up under pressure apparently" - not quite; you don't have access to their forum, I do; and I can assure you, it is mighty busy and getting bigger, all that in spite of Steve Copan upping up the price of his book!
No sysem is perfect, and no trader is perfect.
A few days ago Warren Buffett apologized to his shareholders for putting his foot in it, and buying some shares (OK, $5 billions) in Goldman Sachs...on terms anybody else would simply find impossible to negotiate! Certainly not me!
This says a lot - if Mr Buffett makes mistakes, and he's a billionaire, what about the rest of us?
The truth is, successfull traders are self made; they have developped their own systems, be it mechanical or otherwise.
Some have even developped their own tools (Gann's "Square of Nine", John Bollinger's "Bollinger's Bands", etc...), but all have this one thing in common: they have their own style, developped over a number of years of trading, study of the markets and of themselves.
YOU CANNOT BUY SUCCESS!
As simple as that - otherwise, we would all be millionaires.

Eduardo.

 

[new_trader]

YOU CANNOT BUY SUCCESS!

Exactly. This is why there are so many trying to sell it instead

 

[Hammer61]

Trading is a very hard and unforgiving business, very few will make it and I would agree the big money earners are discretionary traders but they are few and far between but that doesn’t mean you cannot make a good regular income trading mechanically.

On a 1 market basis my system would average 10% to 15%, in a good year up to 30% and a bad year -5% or the odd occasion -10%, so on that basis its comparable to a pension fund, I however run it across quite few markets, this is MAYBE one advantage a mechanical system has over discretionary in that from Monday through to Friday I have orders or open positions in different markets around the globe nearly 24 hours, most discretionary traders focus or trade only 1 or 2 markets, but these markets they know inside out.

I tested over 15 years of data or from when data was available, its only the past 3 years I put my foot to the peddle, applied position sizing and traded more contracts and markets over this time I have done well but back testing shows over all the data I have this would have been the case had I traded those years as well, part of my doing well is down to money management an area I believe gets very little attention but is critical to success as a trader.

There are so many ways to trade the markets and so many people will slag off and criticize other methods or approaches that don’t fit with there beliefs or experiences, I find the most of the Gann stuff totally incomprehensible and most all classical technical analysis to be a total waste of time but I don’t doubt there are people out there who do use it effectively, each to there own.

As to commercially available mechanical systems, I don’t know of any that I would pay money for, I think those that do work and are good just aren’t available to buy and are very much kept under wraps and not disclosed.

 

[Pat494]

As to commercially available mechanical systems, I don’t know of any that I would pay money for, I think those that do work and are good just aren’t available to buy and are very much kept under wraps and not disclosed.

spot on
The under wraps lot tantalize with what may be possible. It all builds up the mystique just like theatre and probably about as tinselly if examined too closely ? If they were any good they would mostly prove it imho. The really big boys may have "invested" here ( like UBS for instance ) but are still in the doodoo !

 

[GCC]

A famous example...

I was hoping to hear someone that was doing 100% mechanical trading to know if there was something clear to aspire to.

Perhaps the following example will be some encouragement for those mechanical traders:

Richard Dennis Trader Profile: Back from the Bottom of the Pit Article at StreetStories.com

In 1994, however, he [Richard Dennis] was back on the scene, with a strategy based solely on mechanical trading signals from computer systems. With this system, he executes an order every time the computer gives him a signal. "When I started in this business there were no computer, and it was hard to do research to determine what made the market tick," he says. "Now, with technology racing ahead, new computers will never tire of being able to perform research that allows me to improve the rules of the system to make my trading more successful."

...Dennis credits the success of his trading system on his extensive research. "I am no longer a trader, I am now a researcher," he says. "I am constantly performing research to determine how I can tweak the system and create new parameters to allow it to adapt to the changes that are constantly occurring in the market."​

I guess the key to mechanical trading success is mentioned above: you have to "tweak the system and create new parameters to allow it to adapt to the changes that are constantly occurring in the market."

 

[new_trader]

"the changes that are constantly occurring in the market"

Utter BS!

 

[GCC]

"the changes that are constantly occurring in the market"

Utter BS!

Well, it's not impossible that it's BS, since "ome critics believe that Dennis’ strict mechanical trading formula is a marketing ploy to put investors at ease over his past record" (quoted from the same article).

But I guess, even if he does use a computer, he would use it with discretion. I don't think he'll forget his big mistake in 1984.

Of course, then, it is not a 100% mechanical system. But who knows, perhaps Dennis has really figured out a holy grail system to trade?

 

[Pat494]

I suppose to be fair one ought to consider that Automated Trading competition. As far as I remember it was over a 3 month period and the winner did produce good results. But it was on stocks. as I recall, so a good chance it was pure luck that a stock or 2 went through the roof ?

 

[donaldduke]

100% pure mechanical trading is possible.
There are methods out there that work. You have to work hard to find them. I mean lots of research, real world trading and feed back and more research... Eventually (years later) you will find something really good you can trade.
And after all that hard work you wont want to put it on Collective2 or sell it, you will want to trade it for yourself.
Mechanical trading is not easy either, multi month drawdowns and long losing streaks, most people cannot trade that way. Buts its no impossibly hard and many people do trade that way.

The only time i need to change a good system is when the underlying market changes. For example say it trades the CL during pit hours, if the pit hours change i would need to change the parameters in the system to reflect this. Or if previously i was trading QM but volume in QM has died and everyone now trades CL instead. Otherwise i let the system get on with it.

 

[mechanicaldaytrader]

I assume the following as facts:
A) the purpose of day trading is to protect your equity account from overnight drawdowns - have tight stop losses to protect the equity from large losses
B) capture the maximum upside to intraday swings in the market

Good discussion gang...considering the Mini Dow easily moves an average of 300+ percent per day of day trading margin (between intraday cycle highs and cycle lows) making 100% per year of day trading margin would not be acceptable to me. With 10% to 20% discretion - and if you miss 66% of the intraday cycles - your account would still be up 50% per day - assuming you only traded the same # of contracts each day and you withdrew your daily winnings each day (the daytrader would have only double the daily margin required by the broker in the account each day).

Mechanical systems suffer drawdowns, so why bother using them in a daytrading environment? IMO you're trying to put a round peg in a square hole.

The Mini Dow typically has 2 or 3 trending trades per day, 80% of all days. As an average, each Mini Dow trend returns 200% to 400% of daytrading margin. There is one institutional stop in the Mini Dow that is touched 80% of the time within a certain number of minutes each day, for instance. A discretionary trader can hold a trade until that stop is touched or initiate a trade FROM that stop - to figure out to program that mechanically, frankly it's not worth the effort imo - just trade the darn stop for 9 months ....and retire...

In the 12 years I've been in this biz, I have not met a single person that understands 3 concepts required (imo) for developing a mechanical system (if one or more exist, I haven't met them);

1] is an expert in measuring when the OVERSOLD and OVERBOUGHT conditions end in a intraday market
2] is an expert in using the proper technical indicator(s) to measure the above conditions AND program a trading platform to compute those conditions
3] is an expert in money management, day trading schemes and knows how the market he trades reverses money flow on a intraday, daily and multi-day time periods.


Until and unless you do ALL 3 things, you're in the weeds and you will stay in the weeds, imo.

I know the creator of Easy Language and probably 3 or 4 of the top Tradestation programmers in the World (that are publicly known)..and NONE of them know how to do #1 (above). The might know a great deal about #2 or #3 (separately) but they can't put it together in a day trading environment, in my experience. The tend to over-analyze and ignore how the "money-flow" works in the market.

Once you know how the Institutional Traders complete their trades and/or reverse their positions then you're on to something. Without that, you will likely be spinning your wheels "hoping" a certain trade works, but constantly plagued by self doubt.

The supremacy of a "methodology" over a "system" is that it only considers "money-flow" and how the OVERSOLD or OVERBOUGHT condition ends...if you trade knowing that a fair percentage of the time, you're account will be up every single day, no sweat.

Example: the Mini Dow DOES NOT reverse on price about 50% of the time....so how are you going to program a mechanical system if you are trying to trade the price? Another 25% of the time the Mini Dow DOES NOT reverse on a futures price/condition but rather another factor...so why bother using a mechanical system that is doomed to fail on the Mini Dow, since you're trading the futures but other prices/factors cause the Mini Dow futures to reverse?

There are NO mechanical systems that are successful that work intraday to my knowledge. There is one Tradestation programmer out there that writes a number of systems but all of his systems suffer about a 70% or more drawdown per month - thus, all his systems violate 1 of the 2 reasons one even considers daytrading! duh!! Adding a little discretion and using a daytrading methodology instead of a inflexible system should guarantee no daily drawdowns and no monthly drawdowns ever occur.

SK - The Mechanical Day Trader

 

[new_trader]

Well, it's not impossible that it's BS, since "ome critics believe that Dennis’ strict mechanical trading formula is a marketing ploy to put investors at ease over his past record" (quoted from the same article).

But I guess, even if he does use a computer, he would use it with discretion. I don't think he'll forget his big mistake in 1984.

Of course, then, it is not a 100% mechanical system. But who knows, perhaps Dennis has really figured out a holy grail system to trade?

"Changing market" is a term mechanical traders use to account for inefficiencies in their systems and with the concept of mechanical trading. You don’t back test 20 years of data and come up with a system to trade only then to change it 2 or 3 or 5 years later because “the market has changed”, it absolutely just doesn’t make any sense whatsoever even from a logical perspective. You don’t even have to be a trader to see this.

Get your favourite trading book and have a look when it was published. Was it published yesterday? No? Has the market ‘changed’ since it was published? Will the book still be good in 5 years time?

If I gave Dennis two sets of stock data and providing he doesn’t compare them to a database but just simply analyses and compares the data itself I bet with all the computing power at his disposal he wouldn’t be able to tell me if they came from “different” markets or the “same” markets. It is utter utter utter crap.

 

[dcraig1]

"Changing market" is a term mechanical traders use to account for inefficiencies in their systems and with the concept of mechanical trading. You don’t back test 20 years of data and come up with a system to trade only then to change it 2 or 3 or 5 years later because “the market has changed”, it absolutely just doesn’t make any sense whatsoever even from a logical perspective. You don’t even have to be a trader to see this.

Really ? If you go back 18 months or so, in the bull market up to that time you could have done pretty well just by buying the stocks with high relative strength in sectors/industries with high relative strength. Even better, buy the dip. Largely a mechanical strategy. That game is over for the time being. Looks like changed market conditions to me.

 

[new_trader]

Adding a little discretion and using a daytrading methodology instead of a inflexible system should guarantee no daily drawdowns and no monthly drawdowns ever occur.

Whatever a "little discretion " means..

I started trading with a 100% mechanical system and after poor performance I added a little discretion to it and then back tested it and kept doing it until I eventually ended up with a 100% discretionary methodology!

If there were no difference in price between First Class and Economy would people still choose to fly Economy?

If there were no difference in price between a 5 star hotel and a 2 star bed and breakfast would people still choose to stay at a 2 star bed and breakfast?

If you had to go from point A to point B and there were two paths, 1) A long and treacherous mountainous rocky terrain and 2) A nice pleasant safe flat footpath, would anyone choose the long and treacherous mountainous rocky terrain?

Why do people "choose" the discretionary trading route?

 

[new_trader]

Really ? If you go back 18 months or so, in the bull market up to that time you could have done pretty well just by buying the stocks with high relative strength in sectors/industries with high relative strength. Even better, buy the dip. Largely a mechanical strategy. That game is over for the time being. Looks like changed market conditions to me.

That is because you aren't looking properly.

 

[dcraig1]

That is because you aren't looking properly.

You're the one making some pretty strong claims with some implied fundamental knowledge of the "essence" of markets. I think any reasonable person will define a bear market as changed market conditions.

 

[new_trader]

You're the one making some pretty strong claims with some implied fundamental knowledge of the "essence" of markets. I think any reasonable person will define a bear market as changed market conditions.

This is the claim "the changes that are constantly occurring in the market"

Have a look at Wasps "WOT" series, over 50 posts from people trying to figure out what happens next and NOBODY asked for the date of the chart

 

[nine]

Nt desperately wants to assume Socrates' mantle - making seemingly intelligent pronouncements.

But the aspiring emperor, like he who went before, has no clothes.