Grid trading system, keeping DD to a minimum

Hello, I am a new member here, but I have been utilizing the resources for a bit so I thought I would add something (hopefully of value) to get people thinking. I am open to all constructive ideas, and welcome all people. With that said, I would ask people to be polite, not use expletives, and read the entire thread before posting any questions - I do not think that is asking too much.

There are a couple of folks recently posting systems here which have good ideas, and this idea has morphed from that - I am certain it is nothing new.

First off, I am of the mind there are no 'No loss systems' - all systems do lose. We are merely hoping that the system wins in the long run (which by my definition could be daily, or weekly, but no more than that). To do this, we need a system that survives through drawdowns. To accomplish this, the drawdowns to profit ratio cannot be excessive. My idea of excessive and your idea may be two different things, so that is subjective.

To get a profitable trading system, we have to use a positive expectancy system. Meaning our wins (in $) times the number of winning trades, must be greater than our losses (in $) times the number of losing trades. Psychologically, I am not to keen on taking large loss after loss after loss, just to get caught up all in one fell swoop by a even larger winner. Others may find this acceptable, but I do not. As such, the goal here is to take as small as loss as possible (which we all know losses will happen right?), which typically has implications on the gains you may see as well - it is a dual edged sword. If your goal is to turn $100 into $100,000, this thread is likely not for you.

That is the general overview of our primary objective for this system, so lets see if we can put our collective heads together to accomplish this goal. If you would like to add your comments, please be prepared to back up your assertions with the math behind it - I do not think it is constructive for the thread to denigrate into a he said she said style of debate. So, while dissension is welcome, please back it up with math/statistics, otherwise please use the button that says 'start a thread' to create your own topic.

Standby for system idea v.01

For myself, when trying to develop a system and determine if it will do well, I find it most productive to decide on my objective first, then develop the first round of system rules (an iterative process), then follow along actual prices on a chart clearly marking the entries and exits, and see how it did after x amounts of trades, or if I set a predetermined system stop. Did it meet my initial objective? Did it do well? Did it do poorly? While it is easy to cherry pick where you start your system after the fact, it is not so easy to do this on the hard right edge. As such, if the system doesn't fall on it face going through a few samples of data, my next step is to pick the worst place to enter, and go through the bars trade by trade. It is here that I have the opportunity to learn if the system has merit, if the system DD is too much even with this small sample, and if the system has the potential to meet the initial objective.

This is the process that I used for v.01 of this system.

If you have been following Remons 'My Trading System!' thread, there are plenty of ideas to spark your enthusiasm for developing (or fine tuning) a system that works well for you. I applaud Remon for sharing his ideas, and for being so patient even with folks who cannot muster the same patience in return. V.1 of Remons idea is a mean reverting system. On paper, it looks like a good idea. Especially considering some of the numbers people use like 'Forex is ranging 70% of the time'. Mean reversion works very well in a ranging market. I ran the code and found it did not fit my requirements for the DD (Drawdown) to gains (risk vs reward). Although the gains are impressive, the drawdowns are even more so. No disrespect to Remon or his system, everyone should choose their own system based upon their comfort level. Also, a system that takes trade after trade in a loss (shorting an uptrending market) waiting for the mean reversion to kick in can get a bit unnerving to both ones psychology and account. I hope Remons v.2 strategy is a great one, as lots of people are showing a great deal of interest. If you have not read his thread, I would encourage you to do so.

I have heard it described there are two type of traders, mean reverting and breakout traders. If you tell someone AIX manufacturing is at all all time new high, people fall into two camps, those who will short it expecting it to run out of steam, and those who will buy expecting it to go even higher. I typically am a mean reversion trader, so that is how I stumbled along Remons grid style trading system. However, there might be a way to trade with the trend and still do well, but it has to be a system limiting our DD (in my opinion).

Grid trading is not new, going with the trend is not new, most systems are not new, but I think they all require you to tweak them to meet your trading style. Going with a trend sounds great, but for markets that mean revert (Forex) they can experience significant DD in relation to the amount of money earned on a winning trade. As such, I do not think it is as easy as setting up buy orders as price goes up, and sell orders as price goes down. I think there has to be more to it, but you are welcome to have a different opinion.

Again, this cannot be a new idea, but each of us has to go on our path to find out something that might work for us.

As such, I was thinking that a system that goes with the trend, in a grid style system, that also limits our drawdowns (through the use of hedged orders when price reverts) may just be the ticket to limit our downside, while still allowing us to ride the trend. We set a profit target, and when our in trade equity hits the profit target, all trades are closed (winners and losers), and we start over again.

continued on next post .....

If we envision a grid that has orders configured on it at pre-determiend levels (buys and sells) according to our preference (that is distance between the orders), we setup how we approach our system initially. Fellow member JaggWaa has graciously provided code in another thread to get this idea started. He is joining us on this thread (thankfully) and will be pivotal to getting the logic behind this system coded.

One of the stumbling blocks of simply buying as price goes up (at predetermined levels in our grid) is that price does not always go up. If and when price come down, it may also trigger sell orders below our original entry point. The more this occurs during a trade, the more our DD. The more our DD, the harder it is to hit our profit target. The harder it is to hit our profit target, the longer it takes to exit our trade. The longer we remain in our trades in a loss, the more risk we are exposed to. The more risk we are exposed to, the higher our expectation of taking even a larger loss, and so on. As such, this is why the grid system that was originally coded does not meet my target objective of keeping the DD low, relative to the reward of the system.

If you do not believe this to be the case, I would ask you to present the math on actual trades likely to occur (which occurred for me today), and present the findings. It is entirely possible to get a decent sized grid going with multiple buys and sells, with none of them in profit and all of them sending our risk exposure up while we are still limited on the our reward.

continued on the next page.....

If we consider that over a large amount of trades, and looked at them on a graph showing the $ amount of each trades winners and losers, we would likely (although not for certain) have a concentrated bulge somewhere along the middle (hopefully to the right edge of the middle - or we may have a system that loses money). From a statistics perspective, if we set a profit target (like $10 or $100 or $1000) and stick to it, we effectively lop off the right edge of our graph, meaning we have no winners larger than our profit target, since we exit when we meet our profit target. However, if we do not do the same for our losers, we effectively limit our upside while accepting unlimited downside (up until the account is closed from a margin call). To me, this is not the properly designed system. It may be that I am too risk averse, so this assumption may not be true for you.

I think a system that goes along with a trend (buys on ups and sells on downs) in a mean reverting market (like Forex) must have a profit target to meet the original objective of not having loss after loss after loss, since the market is likely to go in the opposite direction at some time. As such, one of the ways we can control our losses (drawdowns) is to hedge when trading goes against us. People feel strongly about this topic of hedging. Many say it will not work - ever. Maybe they are correct. I would like to explore the other alternative. I know hedging is not allowed at U.S. brokerages, so we do not need to rehash this issue over and over again here. If this is a concern for you, you can continue reading as there are folks who believe you can mimic any hedging system using math and closing open positions and increasing lot sizes in the opposite direction.

So, I believe hedging might be a logical avenue to explore for limiting our DD exposure in a grid style system.

Onto the system.

As a base, we will start with the system already coded by JaggWaa and expand upon it (JaggWaa will be with us shortly). Upon starting the EA, it places buy orders and sell orders at predetermined points along our grid (determined by the user). As price move to one of the trigger points (lets say price moves up) a buy order is triggered. If price continues to go up, other orders are triggered and we likely hit our profit target. This is the construct of the base system - it follows the trend. However, This system has some potential for a significant drawdown if price does not move in a linear manner (like Forex).

The best way to explain this is to show a grid and do the math.

No hedge trend grid system - Price ranges up and down through various price points over a time period, and it hits and triggers the following looking grid See attached). As it currently sits, we have four buys and four sells open, we are back to our starting point when we turned on the EA, and we are looking at a 200 pip loss (or whatever your grid is set to). ALL eight orders are at a loss.

Hedge trend grid system - Price ranges up and down through various price points over a time period, and it hits and triggers the following looking grid. As it currently sits, we have eight buys and eight sells open, we are back to our starting point when we turned on the EA, and we are looking at a 80 pip loss (or whatever your grid is set to). Only the top buy order and bottom sell orders are at a loss, all the others cancel each other out.

The net of this is we had the same upside when the trade started - if price moves up or down in a trend, both of them will return the same gain. However, when price ranges around, we reduced our downside exposure by 60%.

If you have 8 lots open long and 8 lots open short and the price moves x pips how much money do you loose or make? If you want to open 1 lot long if you are short 1 lot why don't you just close the open position?

The system,

We start with a grid, buys above, sells below.

As price moves up to touch the grid level we buy. At this same time - all the other orders remain, but in addition we also setup a corresponding sell order at the grid level below if price touches that level in the future. If price moves up further to the next grid level, another buy order is triggered. At this time, an additional sell order is placed at the grid level below if price touches that level in the future. If price continues to move up one level at a time, none of the sell orders are triggered. However, if price moves back down to the previous grid level, a sell order is triggered, effectively allowing us to have only the top buy order as our exposure if price continues to fall.

We are allowed to have no more than one sell order and one buy order at a specific grid level. So, depending upon which way price moves, we could have a buy and a sell at one grid level, but no more than that, and each buy order is the same size as the sell order, and none of them increase or decrease in size throughout.

Hey

im a new member here

i would love to join this new thread and share some grid ideas that have resonable risk

first idea


at every grid level we put pending buysand sells lot sizes are the same everywhere
the picture shows the main idea's strategy

You bring up a great point, I have seen many people here on the forum suggest that all hedges can be replicated without taking a trade in the opposite direction, and I agree that in many cases it can be done (this likely being one of those cases), so I am not against discussing the idea of doing this same type of thing here, as you suggest. However, after looking at Remon's v.2 strategy, I would have a hard time identifying how a non-hedged trading system (one where you do not have opposing orders opened on the same pair) is possible while still mimicking that system. It may be possible, but I would think it would be more advanced in mimicking how he takes his profits at a time much later than the entry of the order was placed. The system proposed here is likely not that difficult of a system to mimic, so your concept is likely to prove worthy of further discussion.

At this point, I am wanting to bring up the concept, and get other peoples opinion as to how best to achieve this, and if the end result will meet the objective or not. Part of the reason that I bring up the losing trade examples right off the bat is it tends to attract people who are more realistic in their trading - people who know they will be taking losses, so they understand the importance of minimizing the DD in relation to their reward.

Thanks for your comment.

On the vein of discussing the drawdown on the grid trend system. Someone may think that the scenario I showed in the spreadsheet was cherry picked to show price back at the exact same level that the grid system was started, thereby maximizing the loss on either the hedged or non-hedged system.

What if we took a look at the math of the drawdowns at each of the price levels within our grid, what would the drawdown be then?

As you can see, no matter where price is inside of the grid system, the drawdown never decreases (or increases for that matter) on the hedged and non-hedged grid system. This is shown on the attachments. The non-hedged loss remains at 200 pips, and the hedged system remains at 80 pips. A big detractor of this type of system is that you have to get out of the grid and pick up a trend that leaves the price levels of the grid to turn the drawdown back into profit (if you did not already close all positions at a predefined DD level).

Or, the math would look like this if we used Killians example of closing the open orders, instead of hedging each of them *without opening any new trades from within the grid after the first one was closed.

In this case of using actual hedges (where we have both open sells and buys on the same pair), one thing the hedge does allow us to do is to defer our loss in hopes of getting the DD back to profit (and some may argue this is not a good thing...). Without using hedging, I do not see any way other than taking the loss as the top buy trade or bottom sell trade is closed out prior to entering a trade in the opposite direction. Your opinion is welcome on this matter.

Here is another example of where the hedging helps. Here is the no-hedged 5 pip grid on display. It has entered 6 total trades over a spread of about 38 pips. Currently, this system losing $26. If this were conducted the same way as described in the hedge strategy above, not only would it not be in a loss, it would actually be in a gain of $1 (since the $27.30 buy trade which it now in a loss, would have been cancelled out by a sell trade at the same level). So, not only are the drawdowns reduced (as explained in the posts above), but in this case the system returns to profit quicker.

Just trying to get an idea of how it would work in a worst condition scenario. I dont know if at the start of grid you place orders or have a gap. I tried to make a pic to see how it would go. I assumed a gap ....start point is 155.00....grid step is 10 pips. So grid starts buys at 155.10 and sells at 154.90

Dont think it would make much difference if there were orders on the start level. Anyway have a look.

Forexmeoff, Thanks for your reply....

You are pretty close in your logic, but you missed some of the entry points in between on your grid (explained below). I would think not opening any order at the start would be the best way, as you let price movement open trades on the grid. But you could do it either way.

Your example is about the worst case scenario for any grid system - a trend trading grid system, or a mean reversion grid system. If you had a sell only trend system, your account would be at a loss. If you had a buy only trend system, you would be at a loss. If you had a mean reverting buy or sell system or a mean reverting buy and sell system you would be at a loss. So, at a grid spacing of 10 pips, all these systems will be losing (given that you have not yet exited your trade). This is what I have seen when using grid systems. My thoughts were to try to minimize this loss.

In looking at your entry point, if were were to say we will stick with your grid placement levels (155, 155.1, 155.2, etc) and if we imagined we would enter just a bit below where you show us entering, this would exactly mimic the grid I described in my attachments above, where would have ranged through the four grid bars above the entry and the four bars below the entry causing us to have hedged orders (if we are not taking the path Killian recommended) on all but the very top of the grid where price traveled (155.30), and the very bottom of the grid where price traveled (154.60). We would have both buys and sells on all the levels in between these extreme points (cancelling each other out), and be left with a short at the 154.60 and a long at 155.30 - all the rest of the grid levels would either be fully hedged, or have not trades at all (using the method Killian suggested - the resulting gain/loss would be the same).

If we started the system just below your arrow, price first went down so we sold at 154.9, price went up to the next grid level so we bought at 155.0, price went back down to 154.9 but we already had a sell so do nothing, price went down to 154.80 so we sell, then down to 154.7 so we sell again. We are in profit at this point. However, price starts to rise, and since this is a hedge system, and we do not know what will come next, we trigger buy orders at 154.8, 154.9 - nothing at 155 since we already have a buy there. Price goes all the way back down to 154.6 and we enter a sell order there. Price starts to come back up and we enter a buy order at 154.7 - Leaving the only un-hedged sell order down at 154.6 as you point out. Price steadily climbs all the way up to 155.3, so we have buys at 155.1, 155.2, and 155.3. Price starts to go down and we enter shorts at 155.2, 155.1, and 155.0. At this point, we have one solo buy at 155.3 and one solo sell at 154.6, All the other orders are fully hedged (again, using Killians method we would just close out the trade instead of hedging, and it would be the same result in this scenario). So, we are at a losing position of -80 pips. If we used the other method of buy above the entry line, sell below it, we are in the hole -200 pips at this same time.

Grid systems and ranging markets typically do not go well with each other. Counter trend (mean reversion) grid systems can do ok (like break even) in a perfectly ranging market, but if you have a downward or upward skew to your range, they lose money. If the market trends, you lose your account as you never get back to where you shorted over and over in an uptrending market (or reverse for downward trends).

I mention Remons thread a couple of times above. Put his strategy #1 EA into a demo account and see what happens - it makes good money for few days and then loses all of the profits and half of your account on the following day. I was in negative margin two days in a row in my demo account, with pretty low leverage and no increasing the lot sizes. I applaud Remon for his ideas and think he is a great guy - and hope his v.2 is much better at limiting the drawdowns. His v.1 idea is a mean reverting system. Around page 90 of that thread, there is the same system in reverse - a trend following system - which does great in trends. However, this system starts to have significant drawdowns (in relation to its profit targets), in any type of ranging conditions. After turning it on, it made $10 twice, and then had a drawdown of $50. So an 80% win rate is required using this small sample. If it was opened on your example above, it would be in a loss right now of $200 (unless you guessed ahead of time where and when to time your exit perfectly using a crystal ball with a $30 profit at 154.70). That is why I wanted to try the hedged system (not new, but I want to address the idea here). Having a drawdown to profit ratio of almost 7:1 (using your example here today) is not going to end well. And the end would come soon.

Based upon the low interest of the thread so far, I think I should have only discussed how the system always wins and never has a drawdown, but that simply is not the case in any trading. I was hoping to get the ideas flowing for getting a system to keep the winning side aspect without blowing up the account on the downside. Lets see if that can happen.

The problem with systems is that you NEED a statistical edge to generate money. Quirks with random order or money management can shift around profits and looses but you won't gain money overall. Fact 1: No matter what type of money management you apply it alone will never turn a loosing strategy into a winning one. If you value more likely positions higher and trade them bigger this is great but through this you have your edge which is not generated by MM.
It's the same concept as martingale. A simple grid isn't better than coin tossing entries you just delay your orders. In terms of money you have made or lost unrealized and realized losses are the same. If your net exposure is flat in the market it simply does not matter if you have your positions open or not. The trade statement may look different at the end but there is no reason to hope that a position comes back into profit if you are hedged.
Besides this part I think Hanover once said that every single nedged position can be single handedly replicated using a single order. If you want to partially take profit you simply have to keep track of the correct numbers and do a little bit of math. With todays computers this is an easy thing to do.
Nedging (http://www.forexfactory.com/showthread.php?t=568015).

Grids can be valid but first of all think about what the basic idea behind them is. Is there a higher probability for a pair to range in a certain price band than to trend? If so, thats your edge you can abuse. If they are more trending than ranging great you can inverse the grid positions and profit through this. How about you combine different currencies baskets? Do they tend to trade? These are the questions you have to look at but simply mashing a grid on a chart and hoping that it will generate money won't do you any good. In my opinion grids are just some exotic way of using a breakout pending order strategy which will burn your account much much faster if you are wrong one time. It's comparable to martingale and combine those 2 and you got the nightmare of trading. If people are not prepared to take losses on their account they simply should not trade!

Killian,

Thanks for the analytical thoughts in putting this response together. I would agree with much of what you have said at face value. However, I have a few thoughts that may stretch the limits on a couple of these ideas. Everything you mention (minus the grid trading) is based upon well founded trading ideology and gaming theory. I will not refute those aspects as I believe they are sound. However, I am questioning what an edge actually is. For some, they have a clear definition of what they 'believe' their edge to be. For others, they may not have an edge, or have an edge that comes and goes, as the markets changes from one style to another. These people may question if they really do have an edge. I do agree that simple money management will not turn a losing system into a winning one. However, I am of the 'belief' that a person could change something as simple as their trade exit for either or both their stop (assuming one is in place) or their profit target, and have a statistical edge. When many people think of an edge, they think of a system strategy that has > 50% win rate. However, as we peel back the layers more and more, you could have an edge on a system that had a 20% win rate, but had a positive expectancy over time. In that case, it would simply require that the winners far exceed the losing trades (in regards to $). This could be accomplished by reducing the average $ amount of the losing trades, or by increasing the $ amount of the winning trades, or both. The turtle traders had such a system. Lose a little a whole bunch of times, and gain a lot a little bit of the time. Does this system have an edge? It is up the the reader to decide for themselves, but they made a lot of money with this system. Whether someone considers this an edge or not, the money mgmt was a key aspect for them. I agree with you in that it did not make their system a profitable one just through money mgmt alone. But, with proper money mgmt, it meant they could stay in the game over the long run and realize the 'edge' in their system (or maybe the term should not be 'edge' if one does not think of it as having an edge, but realizing the positive expectancy of this system over the long run).

I tested a breakout (non-mean reverting) strategy awhile back where I traded stocks that had a overnight gap. At 5 minutes after open the morning after the gap, I placed a buy limit order at the high price of the first 5 minute bar, and placed a sell limit order at the bottom price of the first 5 minute bar. I did not care which way the trade was entered - long or short. 99% of the time, price always left the confines of the first 5 minute bar. Once the trade was triggered, the opposite limit order was cancelled, and a stop order was placed 1 cent above (for shorts) the high of the 5 minute candle, or one cent below the first 5 minute bat for longs. If I went long, and price moved up, I reset the stop the same percentage as the price movement in my direction relative to the size of the first 5 minute candle. In other words, if the opening 5 minute candle was 32 cents in height (high to low) and price moved in the direction of my trade 16 cents, I reset the stop 16 cents up (so now my risk is half of what I started with - .5R) If price moved up (this movement did not have to be linear) another 17 cents, I would now be break even
(32 cent bar height - 1 cent below this bar for initial stop = 33 cents upward movement required to BE). This part of the system was fixed. I did not have specific profits targets, and I used a discretionary exit for my profit target (this only mattered for the stocks that were not stopped out).

I traded this system until I had one thousand trades (more than enough for running statistical functions on), and looked at the results. Most days there were between 5 to 13 trades (above this the account ran out of margin for placing the trades). All trade orders were placed 5 minutes after market open. Most days, the trading would only last around 30 to 50 minutes (many were stopped out within minutes, some of the winners could go on much longer throughout the day if I did not exit). That system set the risk level at 1% (1R) of account size for any one trade, fixed at $1k per entry of loss. At the end of the 1000 trade period (which took a bit over 6 months to run), it generated 180R in profit (minus slippage and trade costs). It had a win rate of around 40%.

One could say this trading system has no edge. One could say it is a simple system that uses no intelligence to perform and cannot have an edge as a result. One could say it is unwise to ever have a system where we move our stops (this is echoed over and over and over in various books, and web sites). But, at the end of the day, the system generated profits. Not a lot of profits (it has .18 expectancy minus slippage and trading costs), but on a 100k account, it would generate 180k of gross profit every six months (Net would look more like 180K minus $10k in brokers fees ($5 in, $5 out), minus slippage (depends on how fast moving PA was when going through the entry target, the speed of execution at your broker)). So taking all of this. After deducting trading cost, we would have a Net of $170k. If you factor in slippage you could expect another $30-40k off this amount, so actual Net would be something more like $130K. How does 13% increase in account value every 6.5 months sound on a system with a questionable edge (some would say no edge at all)?

So, back to your main point. Does this system described above have an edge? I think so based upon its ability to generate profits over 1000 trades, but many would argue that it does not have any edge. I think the real reason this system was profitable was the stop placement logic. Most of the losing trades were stopped out at < 1R. The system had < 50% winners, so keeping the drawdown in check was critical to the systems success.

Can a simple grid system have an edge? Possibly. If it does have an edge, the common theme from my perspective is minimizing the drawdowns, just as the trading system above had. Maybe neither system (grid or the one I describe) has an edge, but they may both be able to generate profits over the long run IF the losing trades are kept under control.

Just my two cents...

My 2c FWIW:

 Randomness and edges in myth #22 here.
 Same pair hedging here and here.
 MM in myth #2 here.
 Types of system and their effectiveness here.
 Grids use entries that are effectively random. You would need to either overlay/filter grid entries with choice of instrument, direction, market conditions, time of day, etc -- or otherwise exploit proven market truisms -- to gain any kind of significant edge.
 Example, description and ideas for automated trend following trading in this thread. The best edge attainable from a trend following system is choice of instrument, i.e. use some kind of nuanced TA/FA to determine which instrument has the highest probability of a long, strong trend.

All merely IMHO. Feel welcome to agree or disagree, and use the info however you wish. Hope it gives you some more ideas to work with.

Good luck with your project.

Hanover,

Thank you for your response. I read through all the posts you recommended and I appreciate your effort in helping people along their path. I can say that you have thought through a LOT of the topics people are faced with. I am not certain that I agree with everything written in your myths, but they appear sound and I am open to hearing them and making a determination for myself some time down the road. I have read so many different things from so many different people, that the only conclusion that I have come to is there are likely 10,000 different ways people can and do make money in the market. Some use stops, some never use stops, some are discretionary, some are totally automated. Some average down, some are convinced no one should ever average down. Some scale in on winning trades, some do not. Some scale out, some are all or nothing. Some never ever move stops, some do. Some swear hedging is not effective in any circumstance, others say it gives them a winning edge. Some people are convinced they have an edge, and suddenly that edge dries up - which makes me wonder if they ever had an edge - or if it was imaginary. Some spend thousands of hours researching, testing, and coding to get a 'winning' strategy. And some spend 5 minutes and their gut and put on trades - and each can be successful. At this point, I am willing to listen to pretty much anything that people do with their trading, as long as they are being honest. And by being honest, I mean with themselves and others.

I have seen people that swear it takes 10,000 hours of constant high level study to master the markets, and I have heard of teenagers take some lessons from their father and do better than the fellow who has 10k hours under his belt. I think the only correlation that I may even consider using to determine success is that success comes to those who have a positive relationship with this game and the money challenges that come along with it. The rest of us somehow short circuit and self sabotage our own path and get very mixed, or discouraging results. Does grid trading have and edge? I am not sure. Can you make money grid trading? My answer is absolutely. Will it lose more money than it wins in the long run? I do not know. I know I fired up a grid trading EA last week that made 32% in 36 hours. If my goal was 20% annual return for the year, I would have shut it down after about 24 hours and called it a year. Does that mean the grid had no edge? Not sure. Assuming for the sake of argument, it has no edge at all, then my next question is do I really need an edge if I just achieved my annual profit target in 24 hours trading the grid? If I felt that much conviction I would put all my funds in a live account and let it run for 24-36 hours. I do not have that much conviction as of yet.

But I also see plenty of people who say they have and edge, are convinced they have an edge, and then disappear very shortly after like the Bermuda triangle swallowed them up (I am hoping this was not the fate for Mim2005). He was convinced he had an edge, and he seems like a very intelligent person. I hope he achieved his success and suddenly retired in the south pacific after his four EA's hit the jackpot - I really do. But, I think being the altruistic and helpful person he seems to be he would have continued posting if the EA's did as expected. So, if a very well educated, top of class, statistically oriented, math savvy, structured minded person is convinced that he has found an edge, and took most if not all the proper steps necessary to confirm it with all the computational horsepower and a high performance brain for this type of thing, and if the results were not as expected, then it is getting harder for me to believe that this elusive 'edge' actually exists.

One could say that grid trading has no edge, but does that mean it cannot make money? Does it mean a grid system cannot make money in the long run? Meaning that executing the system will always result in losing in the long run? Does any system really have an 'edge', or is that something we make up to give ourselves the confidence to trade the system, and maybe that is all we need. Maybe that is our 'edge'. That keeps me thinking there just might be something more to it. Something that comes from our individual relationship with money, and how this affects our decision making. I know from a statistics perspective, one can prove or disprove if a system has an 'edge' by properly backtesting. But that may or may not have anything to do with the results going forward, as pointed out on these forums time after time. In order for one to have an 'edge' does one have to pick a trade direction (long or short) first to have this 'edge'? Or, can something like grid trading - which lets the market decide the entries it makes - also have en edge merely through the use of the exits chosen by the system operator and good money mgmt? I know many people have strong opinions about this, I am just not sure at this point which opinion is correct, or better stated, which opinion is correct for me (since this all may be tied to my belief system anyway).

So here is an idea that I came up with, that may or may not have an 'edge' - but only if you trade it. This idea requires about a second grade education (so a 7 year old should be very capable in developing and following this system). I trade CL Oil futures. If you take a look at a daily chart over the last two years on a continuous contract, and draw a line from lowest low to lower low, this system of mine - the second grader system, which I am convinced that works and may have an edge (or maybe not) is to buy every time CL touches that line. It worked the first time you could have traded it in March of 2015, it worked in August of 2015, it worked in Dec 2015 (although less of a up spike after), then price finally broke down below this line then started a series of new lower lows, and it worked again using the new lower low line, the most recent time, on Feb 11, 2016. So, if you traded it every time it touched the line, it worked 4 out of 5 times - which gives us a 80% win rate, the MAE was next to nothing (with the exception in January) the return was 200% (or over) in March, and again in Aug, and again in Feb, with Decembers bounce only amounting to a 60% gain. The gains all came in a week of trading (or less) after entry.

Is this an edge? I am not sure. I do know if you would have played each of them you could have had more than 660% non compounded return, and only one loss (that you could have actually gotten out at break even as I did (thankfully)). If you had a system objective of 20% annual return, you could have shut the whole system down after your first trade back in March 2015. No huge EA required, no super computers, no brain surgery, just ask any 7 year old to draw a line at the bottom of the daily CL chart. I have no idea if this is an edge, but if a simple person like me can come up with such a trivial system, that may not even have an edge, then is an edge really necessary? This may not even qualify as a system (or in simplified terms merely an entry signal for a system), if it is not, then it certainly has no edge. And it will not be repeatable unless Oil (CL) makes a new lower low and touches this same line once again. So, maybe for a fleeting moment (spanning 5 trades over 1.5 years) it may have worked, but if this worked why won't it work on others, and do I really need to be a market wizard with thousands of hours of experience and expertise, or do I just need to find something where I draw some lines, that works a few times, and use proper MM so I do not go broke in case it does not go my way (like it did not go my way in Jan 2016 for a week or so when CL went below this line to make a new Lower Low), and then live to trade another day?

Basically, that is why I am looking at reducing my drawdown on new ideas that other people who are much smarter than I come up with, just to give it a chance that they may know something that I don't - which is entirely possible.

I appreciate your drawing my attention to your words of wisdom. Thank you - I will use them wisely.

genasea,

Thanks for your comprehensive reply. I won't respond in great detail as I'd merely end up repeating material that I've written many times before, some of which is covered in the links that I gave.

As you say, there are many potential edges, including some that seem to contradict each other. It reminds me of a comment by Jack Schwager in the preface of one of his Market Wizards books: "There are a million ways to make money in markets. The irony is that they are all very difficult to find." There many folk whose amazing performance appears to be the result of some kind of opportunistic knack, rather than through the systematic application of math, logic or 'conventional' trading wisdom.

Re mim2005, his primary focus has always been his investment portfolio, although he does trade (but not using MT4). The MT4 EAs that he wrote were something of a diversion for him. I still keep in touch with him on a regular basis.

I'm not familiar with CL oil futures, so I won't attempt to comment.

Anyway, best wishes, I hope you attain the success that you're seeking.

David

Hello genasea, I think everything is simple. You have one base trend strategy you fully rely upon (whatever that might be, it does not matter, it just have to be really good, unfortunately it will cause some, even big DDs during ranges , but that is not a big problem) and one slave range strategy to support it (which could be a reversed based or just another strategy which delivers quite opposite results). You do trade both one against the other. The trend is always the leader and everything you do is wrapped around it.

I always hedge. In between 50% and 100%. I do close the range strategy first. Then let the trend one finish the job. Here are some of examples, which I manage to screenshot (by pure luck or chance), some 5-10 mins ago:

The Aggressive approach:
30 mins before the screenshot was taken, the range system locked down $5k of profit and stopped trading, leaving the leading trend strategy to finish the rest:


The trend system locks down $15k of cumulative profit (the extra $5k closed on beforehand is included) and the trading is all over (restarted).


The Mild approach (I closed manually and everything had happened so quickly I could not take any screenshots). Here you can see BOTH strategies at the time of closure are profitable at the same time. I have said the trend must be the leader. You can see that sometimes I miss to close the range strategy, therefore the whole trading ends up at profit, but the range strategy being at loss at that very same moment of time. The last two closures happen to catch those rare moments when both strategies are still profitable at the same moment of closure:


Here is only one level traded with the Aggressive approach. Take a look at the dashboards. -$78 for the trend and +$45 for the range:


I do decide to close down the range, cause I think there is gonna be movement soon:


The trend's max profit moment (+$175). Here I have screwed it up with the range, cause I just have found it out, the range did restart, when it should not (my bad in the set file), but it is too late already. Unfortunately, this unnecessary restart lowered down, a little, my trend's profit:


Decided to close down everything at a net of $146 of which the range was $29 (used to be $44, but I have screwed it up with that trading session):


The Aggressive and the Mild approaches are really very complicated to explain. However, you can see the simple 1 level Aggressive trading. At the time when the DD of the whole session was -$78 USD of which +$44 was range's profit. Without the range strategy, the DD would have been -$122 instead. Something like 1/3 LESS DD and the same amount of MORE profit (when traded properly).

With practice, all this can be dramatically improved...

P.S. Too all that might disagree with me, please spare me your opinions. I am not interested in going into any kind of arguments. Be kind to me, please, and just think of me as someone being very stupid... That would perfectly work for both sides and definitely suits me best. Thank you!

That is absolutely correct, genasea. That is why you must take the best out of ANY losing moments your system generates. Do not watch how the banksters are taking your money out of your account, but take their side instead. Do not trust the charts. The banksters are magicians. Top magicians. The best one around, in my opinion. They do want you to be looking at the charts. They want all your attention just there. So they can do their focus-bogus thing, without you noticing it.

The focus-bogus happens within THE PRICE itself, but almost nobody is looking in there... because everyone is staring at the stupid charts, which are meaningless in about 80% of the time. That is why the losers in forex are about 99.99% of all... because they are looking at the charts, when they should have been looking at the price, instead.

Of course, I am not telling it all. I am just generalizing. All charts are crap. Shortly put. Price is the what it is important. And there is A LOT to be SEEN IN THE PRICE. You just cant imagine... Anyways.

Here are two examples where the trend and the range strategies can be both at profit. I am trading 30+ demos at a time and most of times I am unable to close down at the max P/L. Here the peak P/L was +$12.6k, while the peak DD was -$3.7k. I closed it all at +$8.1k. The range was still maintaining some profit (+$0.8k) at the time of closure:


Here the range strategy was useless and could not contribute a dime to the trend one, but still, everything ends up in profit. That is how important the trend strategy is. That is why one should wrap around everything around it...