most people won’t have the discipline or patience to follow the rules exactly and they will likely have dismal results and toss they system.
So sorry for always replying late! But I got 2 Jobs!!
Why Would Someone Sell a Highly Profitable System?
- #22
- Jul 16, 2014 10:42pm Jul 16, 2014 10:42pm
- Joined Sep 2010 | Status: Life Time Member | 6,685 Posts
{quote} Nailed it
- #24
- Edited 1:19am Jul 17, 2014 12:34am | Edited 1:19am
- Joined Mar 2011 | Status: I should be on your ignore list | 5,597 Posts
- #26
- Edited 4:45am Jul 17, 2014 1:42am | Edited 4:45am
Can't really understand that obsession with the automated trading systems. I would say that with patience and some rudimentary prudence anyone can make profits in FX without seeking the wholy grail.
Automated trading is a completely different horse. Trying to automate the 'human' trading is non-sense (my opinion only): there are too many rules involved and you would require a supercomputer probably to equal the brain efficiency and MT4 or MQL cant cope with that.
Where automated trading has an edge is in the scenarios where you get some advantage because you operate in a way the brain can't operate: you operate too much, too frequently or with rules requiring tedious math.
An example: let's take a simple martingale. We do know that it's a negative system in the long term, a human trader should not trade with a simple martingale and I suppose that anyone trading positively such a system is (concious or incounciously) applying entry filters and breaking the basic rules quite frequently.
But an automated system can run at the same time not one but many martingales at the same time (on 20 instruments, in both directions, in new detected ranges...), at any time hundreds of operations are live and operations are opened and closed every 5 minutes... That system has to be automated.
And it is positive in the longterm (most probably
).t
What is really happening in real trading with a winning system - if it exists at all - has nothing to do with following rules.
If you trade 'discretely' you're surely using a lot of very complicated rules. The more rules, the more complicated, the better.
If you trade 'automatically' and it's a machine instead of you doing the hard work you have to use simpler and fewer rules but you take the advantages of being 24 hours a day watching tirelessly and taking decisions in a consistent way that can be analyzed statistically to eventually improve a winning system.
And yes, those systems (most probably, again) exist, here is an example: simple bi-directional martingales with increasing lineally the gap and volume in 20 or more instruments, no stop-loss, take-profit at 30 pips with a short trailing. As you see very simple rules.
This is two months (demo account) for 24 currency pairs and 6 indexes, blue line is balance, red for equity:
An this is is a 4-month on a real account, but in this case only currency:
A risky system whose goal is to get 100% in one or two months or go broke in an account of 3000 or 10000 euro. Stats showing success 7 or 8 times out of 10.
t.
- #28
- Jul 18, 2014 4:00pm Jul 18, 2014 4:00pm
- Joined Sep 2010 | Status: Life Time Member | 6,685 Posts
I Like trading manual much more exciting for me, to each it own
- #29
- Jul 18, 2014 4:28pm Jul 18, 2014 4:28pm
- Joined Jul 2011 | Status: Winning | 1,859 Posts
Please check out my blog at https://from-zero-to-hero . net
#1 I am a nice person
#2 I want to help other people, if I can
And I feel like trusting you... What if I send you an EA, a good one of course, an EA I'm running right now with real money. You try it, you validate it behaves (see my 1st post in this thread as what to ask to a well-behaved robot). Once a month you check the profits, if you're satisfied you donate what you think is fair... that's all. Deal?
A number of things can happen:
1) the EA does not deliver as expected you will lose a small amount of money or nothing at all (if you 1st try it on a demo account).
2) the EA performs but I get no benefit of this relationship, no problem, EA will eventually stop working. You will earn some money and I'll lose some time.
3) EA performs and I get the extra benefit of your donation. Perfect everybody wins and this is the start of, maybe, a long and satisfying relation.
So?
t.
- #32
- Jul 20, 2014 9:53am Jul 20, 2014 9:53am
- | Commercial User | Joined Sep 2005 | 2,784 Posts
about something like this
2years backtest, eurusd 1hr timeframe, only open bar calculation (the ea open or close positions only at the opening of a bar, so we don't need "intra-bar" backtest).
(same ea, different settings)
opening-auction price: 100k!
2years backtest, eurusd 1hr timeframe, only open bar calculation (the ea open or close positions only at the opening of a bar, so we don't need "intra-bar" backtest).
(same ea, different settings)
opening-auction price: 100k!
- #33
- Edited 10:34am Jul 20, 2014 10:06am | Edited 10:34am
about something like this 2years backtest, eurusd 1hr timeframe, only open bar calculation (the ea open or close positions only at the opening of a bar, so we don't need "intra-bar" backtest). (same ea, different settings) opening-auction price: 100k!
But you already knew that.
t
- #35
- Jul 20, 2014 10:46am Jul 20, 2014 10:46am
- Joined Feb 2009 | Status: Half in the Bag | 17,827 Posts
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
- #37
- Jul 20, 2014 11:16am Jul 20, 2014 11:16am
- | Commercial User | Joined Sep 2005 | 2,784 Posts
{quote} Sorry, I was not clear about the MM. Of course any (manual or auto) system or strategy has to consider MM. What I meant is that in order to evaluate a system we need to understand how it operates, MM makes that difficult to understand. Any well-behaved system should be positive with no MM. I don't think that MM can turn a bad system into a profitable one. But maybe someone can produce a proof against that. I need to analyze the operations done, to verify that the risk is controlled. For my taste any system should be based in simple assumptions...
As per my system, it is "pip" positive with or without MM, of course the "magic" of 10-500k% returns is given thanks to the MM, and i can share the base concept as well:
We have around 70% of profitable trades, this mean that every 10 trades 7 are profitables, right..
The profitable/loss pips is good too (around 1:1), so here comes the idea that, when we have a loosing trade, the next one will use a slightly more aggressive MM/lotsize (this doesn't mean martiangle, just slightly more aggressive).
The aggressivity of each trade is dictated from the difference between the actual equity and the "ideal" equity (an ideal equity line that tick up of 0,xx% on each new candle, always going up).
The more far the current equity is far from the ideal equity, the more aggressive the ea become.
Doing this (and thanks to the winners/loosers % ratio) the ea is able to close each month with the same roi%, and as the capital grow, the profit grow too, exponentially.
The only real danger would come from an exceptional loosing streak, we know that even if it didn't happened in the past 10 years, it can happen next month (the same principle apply to any EA and manual-trader).
But i guess the only way for achieve the perfection in trading is having a time-machine, and literally know the future.
I just shared the MM base concept of my EA, it is a really powerful concept, i'm sure that this will be enlightening for someone.
- #38
- Jul 20, 2014 12:14pm Jul 20, 2014 12:14pm
{quote} We have around 70% of profitable trades, this mean that every 10 trades 7 are profitables, right.. The profitable/loss pips is good too (around 1:1)
1) if you have a 70% expectancy of win with a 1:1 R/R you have the holy grail. I don't need the excel or matlab to know that I'd be rich in a few months with a system like that.
2) if you had that system then why you would use the MM system that increases the risk when you are getting some bad results (that can be expected, of course, within the EA specs)???
The decision to increase the volume of the trades based on the past history has to be done when we have some info about how positive and negative trades distribute with time. It involves no obvious math but analyzing the historical trades (and assuming that the behavior will not change!) you eventually get 'relevant' statistics showing that the probability of a win the next operation can be higher (or lower) depending in the result of the n last operations.
Theoretically at least we might say the system is 'trendy' in its performance (when after some wins we can expect another win) or 'compensator' (if it's the opposite, after some wins you should expect a loss). If you get relevant statistical data that supports your system is 'trendy' then the right MM requires you to increase the volume after a win and decrease after a loss. If you decide your system is 'compensator', and only in that instance, you increase volume after loosing.
I've found during this years systems doing the trendy and the compensator thing. But, I've NEVER found any evidence, with enough stats support, to do one thing or the other. And that's assuming the markets will not change in a way that a trending system will become compensator or viceversa. So.... I don't usually use that approach. If I had to use that type of MM my gut feeling is to do exactly the opposite: after an unusual series of losses I would go on 'test' mode, operating at minimum volume or using a demo account till the system gets back on specs.
But, anyway, why would you do anything like this with the holy grail????
t.
- #39
- Jul 20, 2014 1:36pm Jul 20, 2014 1:36pm
- | Commercial User | Joined Sep 2005 | 2,784 Posts
{quote} But, anyway, why would you do anything like this with the holy grail????
To answer your question, the purpose of my MM is to produce a normalized constant steady roi% every month (same %return everymonth), regardless of the market volatility changes.
As for the system itself, it is nothing complex, just a buylow sellhigh ranging system, buy when price go down of xx pips, sell when go up of xx pips, and the number of pips is dynamic too, strictly linked to the daily average true range (atr).
The eurusd is a trendless ranging market most of the time, so this concept is giving good results.
Simple things are always the best!
Of course looking the weekly/monthly chart it seems that the market is getting more and more slow, more and more on a tight range, ready for a powerful breakout and perhaps a powerful trend, if that will happen, all range-based systems will start to fail miserably.
That's the dark side of EAs, they work untill they stop working.
Trending systems have good profits when market trend, and give them back all when market range.
Range system have good profits when market range, and give them back all when market trend.
On a long scale enough every EA is doomed to fail (the holygrail long-term EA simply cannot exist).
The first Buddha teaching: "Everything changes, nothing last forever."
- #40
- Jul 20, 2014 2:59pm Jul 20, 2014 2:59pm
- Joined Feb 2009 | Status: Half in the Bag | 17,827 Posts
The market volatility is dynamic...On a long scale enough every EA is doomed to fail (the holygrail long-term EA simply cannot exist). The first Buddha teaching: "Everything changes, nothing last forever."
TA DA!
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!