Systems, schmystems

[barjon]

Having a read about some of the latest whiz-bang, make your fortune systems I'm reminded of the Casino operators who rub their hands with glee whenever a punter arrives at their tables with a system. So much do they love them that they advertise systems all over the walls.

Apart from one guy who operates purely from instinct (and who must have been born with a lucky streak a mile wide) all the decent traders I know have a sound strategy and are systematic in the application of their strategy. But that's a far cry from being a slave to a system.

Yes, they've got rules. Yes, they've got conditions. It's all applied with a healthy dose of discretion and it's that discretion which brings home their bacon.

good trading

jon

 

[Shakone]

The inventor of the card counting system, Ed Thorp (afterwards a very successful hedge fund manager), even announced that he had this system. The casinos as you say love people with systems, so he was welcomed with open arms. He proceeded to turn 10,000 dollars into 21,000 dollars in a weekend. According to him, he would have made more, but some of the casinos ejected him. He stuck rigidly to his strategy, because the strategy is where the edge was.

Saying that the discretion is what brings home the bacon, to me sounds like ego (no offence intended). An unwillingness to accept that the strategy is giving the edge, rather than you.

What makes you say that it is the discretion which is bringing home the bacon? How are you measuring that? What part of your discretion could not be built into the strategy? If you can't build it into the system, presumably that means you can't define it, which then means you can't compare the results of your trading with or without this discretion in any meaningful way.

Perhaps you're on a higher level of trading than I am, and can do this intuitively, but for me it is my strategy which makes money, and my discretion which tends to mess it up.

 

[barjon]

shakeone

No, I'm not boasting of being a decent trader let alone on a higher level than you.

All I am saying is that the successful traders I know have a strong discretionary element in the application of their strategies. It is they who say that the discretionary element is the bit that really adds to the bottom line. I suppose I'd accept the old Mandy Rice comment "they would say that wouldn't they".

The other interesting bit is that half of them trade a particular set-up across a range of instruments, whilst the other half trade a range of set-ups across a single instrument. The last lot are more discretionary than the first.

We're not talking a huge sample of traders here, of course.

good trading

jon

 

[Shakone]

Well I'm certainly not saying it is impossible that discretion can add an edge to your trading, but I just don't understand how it is measurable. 'Discretion' seems very imprecise. I don't have a discretionary edge, so if some people do, perhaps they could explain it in more detail. I am interested to add whatever I can to maximise gains. Also it would be good to know if those discretionary traders have actually measured the performance from discretion trading with the performance had they just followed the system. And also why the discretionary element can't be built into the system.

For me, I have a system, and the only discretionary element is that I may avoid an entry if there is an important news announcement soon. But that is now built into the system, and easily defined.

 

[barjon]

It seems to me that, by its very nature, discretion is imprecise, variable and difficult to measure. I remembered in a recent thread a long past member, Bonsai, who traded FTSE and who would often exit "because it's just sitting there looking at me". He was much more often right than wrong in timing his exits.

I suspect he would find it very difficult to give a precise definition of that phrase, albeit that he may have been subconsciously weighing a number of criteria to arrive at that judgement.

jon

 

[wackypete2]

Card counting is very different than trading. Sticking to the rigid rules of card counting without deviation or discretion can give you a pure mathematical edge, however small, which is something you can't do trading. If it was that easy then ea's would be a sure thing..easy, right?

Peter

 

[chart rider]

....Sticking to the rigid rules of card counting without deviation or discretion can give you a pure mathematical edge, however small....

Just which part is it that's not identical to trading?

CR

 

[Adamus]

Where's the similarity, apart from the bit about having an edge?

 

[chart rider]

The similarity relates to sticking to the rules without deviation or discretion, otherwise, I maintain, there can be no mathematical edge.

The market is driven by an incalculable number of inputs, of varying degrees of irrationality. Can the human mind analyse such overwhelming data and apply discretion with any consistency? How can the mind, with no mathematical system to follow, recognize the statistical character of any pattern and how can the mind even recognize varying patterns when hindered by cognitive biases.

If the system is not mathematically / statistically derived, then how can any conclusions be made about the systems edge.

If a system is mathematically / statistically derived, then there is no choice but to follow and act as a robot.

CR

 

[tomorton]

Yes chart rider - but - a lot of traders want more to be seen as 'good traders', in their own perception and/or in other peple's, rather than make a lot of money.

These guys (I can't imagine many of them are female) are driven to deny the fact that they follow rules, they don't register or admit to their own rules, they don't spend time practicing, backtesting and refining their systems outside market hours, they like to see themselves as wheeler-dealers, surviving by their wits, always with the risk of being wiped out every day. They're outlaws, renegades, living outside the rules, effectively robbing the suckers in the market like a romantic jewel thief cat burglar, leaving only a monogrammed silk glove where they struck.

Their heads are totally clogged with misperceptions regarding money, morals, work ethic, self-worth, self determination and the fear of failure.

Some of them are very good traders. Most of them go broke very fast. The ones who are really good and who survive follow rules like a machine.

 

[chart rider]

tomorton

As has been stated in many books, successful trading is boring. To dissect a little, personally I find the development work, backtesting analysis of real world results etc to be anything but boring.

With regards to the actual trading, being a position trader and with the trades placed during the closing auctions, that's 10 minutes a day actually trading I would say it's an emotionally neutral action. Or at least that's the aim of regular mental training.

Agree with you, the trading platform is no place for action man - do all the action before logging on, then oil the machine and hang the brain up on the hook, for it has no use once logged on.

CR

 

[barjon]

.................... The ones who are really good and who survive follow rules like a machine....................

Well. that's the bit where I part company, Tom

The guys I know are rigidly disciplined in sticking to their strategy which they apply in a systematic way. However, the factors involved in their actual entries and exits are rather more loosely framed and they tend to look at what's happening in price zones rather than automatically using a specific price trigger.

An example is one who trades break outs. He wants to see a bit of oomph and sustainability in the break out rather than price sticking its head a couple of ticks above the old high for a fleeting instant while it staggers about near it. Ask him what he means by "oomph", "sustainability" and "staggering about" and his answer is much more about his feel for the price action at that level than about a set of rules that must apply. So, sometimes he'll take the trade, sometimes not. Sometimes he'll take it immediately, sometimes he'll hang on and take it a bit later. Sometimes he'll have a tight stop-loss, sometimes a looser one.

good trading

jon

 

[bedsit]

I am not in favour of discretion when using systems. Maybe a quote from Mark Douglas (my favourite author on trading psychology) may help:

"Most people like to think of themselves as risk takers, but what they really want is a guaranteed outcome with some momentary suspense to make them feel as if the outcome had been in doubt. The momentary suspense adds the thrill factor necessary to keep our lives from getting too boring. When it comes right down to it, no one trades to lose, no one puts on a trade believing it is going to be a loser, and all systems will definitely have some percentage of losing trades. So it's difficult not to be tempted into trying to guess which ones are going to be losers and not participate.

Sometimes the system will give you signals to trade in ways that are completely contrary to your logic and reasoning. Sometimes the system will defy your reasoning and be right, and sometimes you will agree with the system and it will be wrong. You need to understand that technical trading systems are not designed to be outguessed. Systems aren't designed to give you isolated signals of an opportunity to be taken when it seems right. What they do is mathematically define, quantify, and categorize past relationships in collective human behaviour to give you a statistically probable outcome of the future."

 

[tomorton]

Hi Jon - I only see that they have a system that incorporates input from price, volume and fundamentals / news - all of which could be classified and objectively quantified as to their weight in the decision process and logged as a set of rules. I suppose there is discretion as to whether a news item has quite good, very good or spectacularly good implications for future price. So one trader might go limit long, while another just takes a punt. But they should both be long.

Surely any other input, which they call 'discretion', is actually just caprice?

And why introduce caprice? To enhance risk. Why enhance risk? so the winner can inflate their perception of their skills as a trader and so the loser can console themselves it was a high-risk trade anyway.

Some of our colleagues trade like students who won't swat for an exam. if they pass, they're brilliant, they didnt even revise. if they fail, well, so what, they didn't even revise so they can't be blamed for failing, in fact they didn't really fail.

 

[tomorton]

Hi bedsit - I was thinking of precisely Mark Douglas when I wrote my earlier post / rant. I'm happy you picked up the unspoken reference.

 

[wackypete2]

Well. that's the bit where I part company, Tom

The guys I know are rigidly disciplined in sticking to their strategy which they apply in a systematic way. However, the factors involved in their actual entries and exits are rather more loosely framed and they tend to look at what's happening in price zones rather than automatically using a specific price trigger.

An example is one who trades break outs. He wants to see a bit of oomph and sustainability in the break out rather than price sticking its head a couple of ticks above the old high for a fleeting instant while it staggers about near it. Ask him what he means by "oomph", "sustainability" and "staggering about" and his answer is much more about his feel for the price action at that level than about a set of rules that must apply. So, sometimes he'll take the trade, sometimes not. Sometimes he'll take it immediately, sometimes he'll hang on and take it a bit later. Sometimes he'll have a tight stop-loss, sometimes a looser one.

good trading

jon

Good Post.

If you are going to trade like a robot (trading without ever bending rules or never using discretion) then why not just use a robot and sit back and make millions the easy way? Program your ea and look out world. There are plenty of pre-canned ones out there for $97

If only it was that easy. Again, card counting and trading are very different. In a 52 card deck, if all 4 aces are already out then you know EXACTLY your probability of getting blackjack on your next hand - zero. That will affect how much you bet or how you play. Every card played has an effect on what can happen next because it is no longer available. It's just isn't the same in trading .

Peter

 

[lbranjord]

Good Post.

If you are going to trade like a robot (trading without ever bending rules or never using discretion) then why not just use a robot and sit back and make millions the easy way? Program your ea and look out world. There are plenty of pre-canned ones out there for $97

If only it was that easy. Again, card counting and trading are very different. In a 52 card deck, if all 4 aces are already out then you know EXACTLY your probability of getting blackjack on your next hand - zero. That will affect how much you bet or how you play. Every card played has an effect on what can happen next because it is no longer available. It's just isn't the same in trading .

Peter

I'm finding this out the hard way right now. I've learned MQL4 and Programmed a dozen eas over the course of one weekend (breaking for 9holes today).

I've used the SAME EXACT ENTRY CRITERIA as I used for my live trades. I have a few tricks in my bag that I profited consistently from. In an Expert Advisor I cannot bring in profit over the long term. No way, no how. I'm not the best EA coder yet, but I'm experienced in both coding and trading so I know the robot is not failing, just the systems. Hedges, retracements, all types of price action, I've made a trend follower w3 slopes and an MA (kind of like that sniper system)...all failed.

Now I'm wondering this: Was I fooling myself and headed for inevitable failure with my live trading? or Is some process taking place behind the scenes that allows me to trade successfully live with a system that does not work in EA form? I feel like a lunatic.

 

[Adamus]

If you have coded exactly what you do as a discretionary trader, you could prove it by running the system on one machine and trading discretionary yourself on another machine.

If you coded what you really trade, you would have exactly the same results from your mechanical system as you've got from discretionary trading. And in that case, then yes your discretionary trading would be doomed, but of course you would adapt. That's what a system won't do on its own.

 

[lbranjord]

Yeah it opened my eyes. Certainly not happy about it, but happy to know the truth of it. I think trading from your system+your gut is the best way to go, but unsustainable for me. If I can't get an EA built that trades successfully, I'll probably place my forex capital back into the stocks I currently hold. From there I'll just keep learning/coding until I find the - holy grail || holy fail.

 

[Adamus]

holy fail

I always figured that any trading system ever invented is going to do "the Holy Fail" at some point, no matter how good the rest of the time. There's no way that a piece of code will always always always adhere to your definition of 'what works'.

And if your type of discretionary trading is actually something that can be realistically coded into an EA, then it'll be subject to the same tendency to fail once in a while.

What a system can do that the trader who invented it can't do manually is to be traded simultaneously with 10 or 50 other such systems - whose PnL together will buffer the time when the Holy Fail hits the fan on one of the systems.

My idea of successful discretionary trading is a mix of technical and fundamental that can't be coded.