Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Know Yourself, Know Your Setup, by Boris Schlossberg 9 replies

I dare you show your statement 13 replies

Twoblink tweaks and mods 61 replies

Changing Your Self Talk By Talking to Yourself 2 replies

Attachments: First-Strike & One-Night-Stand users, show yourself and your mods, if you dare!!!
Exit Attachments

First-Strike & One-Night-Stand users, show yourself and your mods, if you dare!!!

Hey guy's thanks for keep this thread going,

cogrove, we are all getting wrecked this week mate, but we have had some goooood weeks in the past.

asasa great to hear you have an EA working for this,


ianfOster, it is truely a curse to have to put up with returns like these for such an investment in time and effort!!! i am truely hearing your plea, lol
Ian you are also right in saying that it was too bad for the minority who where hell bent on getting this thread binned, in hind sight i should have been more in control of the direction the thread was heading, but i turely believe in the good in people not the bad.

anyway regardless of them there are plenty of us who continue to follow the simple way that joel had given us, and do our due dilegence every week and place our trades.

cheers
geoff
Quoting cosgrove
I'm getting wrecked this week so far, but still playing by the rules.

I think I've missed a few trades, because Joel now places new orders even if carrying an old trade over from last week (thought that wasn't in the rules). As of Monday I was up 20% from my start on March 1st 2009, but after this week's 3 losses so far, my balance is only up 16.5%.

Does the above EA use 100 pip stops or volatility based stops?
Hi Cosgrove,
As you know, I am not using Joel's rules. However his rules do state that you should :

5. Keep monitoring and placing your current buy and sell prices for each week. Your new buy or sell orders for the week, under some rare circumstances; may be closer than a protective stop loss order for an existing position put on the week earlier that hasn't exited profitably yet.

Even though he has a big heading before the rule set that says "When you are Flat the market ....." It is obvious that the nabove applies every week - otherwise there is no point in him saying that the new buy/sell orders may be closer than a protective stop los order on an existing position ....

I too am having a slightly negative week as of yesterday when E/U went against me. My U/J order is still not triggered, so that may put me seriiously down for the week, or it may rescue some profit.

Trading like this you have to be able to hang on in there even when it goes against you for several weeks - beause it is the few weeks that trend like crazy that produce the overall profit. My best week was commencing 6th September 2008 - I made 1,064.5 pips ( over 3 times my total profit from the previous 14 weeks). My worst week so far was 9th February 2009 when I lost 633.3 pips (but what was even worse was that was just one of four successive weeks that I lost - in fact I was in a drawdown {based upon pips} from 22nd December up to last week).
Attached File(s)
File Type: pdf FirstStrike_Plus.pdf   171 KB | 1,016 downloads
File Type: pdf OneNightStand Ruleset.pdf   728 KB | 1,704 downloads
Hi all.
It's nice to write in this forum.
I have implemented a simple enhancement in my EA and things have improved.
If we are in a condition of price range contraction, increase the position size of the orders.
Logical mod, interesting improvement in equity curve.
All of the other parameters are the same as in the previous backtest.
The results are in the picture attached to this post.
Next step (still to implement) is to increase the position size in the direction of the long-term trend.
As always, any comment is welcome.
Attached Image
And here is the effect of the addition of the concept of Narrow Range Breakout...
Pair: EUR/USD
Simulation interval: September 1999 - June 2009
Starting Capital: 10000
Ending Capital: 244976.93
Profit Factor: 1.69
Maximal Drawdown: 18.34%
Profit Trades: 58.97%
Maximum Consecutive Wins: 18
Maximum Consecutive Losses: 10
Equity: refer to enclosed picture
Attached Image
First of all, thanks to everybody helping to keep threads on mechanical systems alive.

Regarding systems popularized by Joel I have a problem: Joel mentioned this system (ONS) to be given to him by Joe Krutsinger. Checking with Krusingers books I find the system there, named ONS, albeit with 2 major differences. Krutsinger is using a shorter liikback ( 4 instead of 10 bars for longs and 8 instead of 10 for shorts) and a directional bias for currencies, expressed by 8 instead of 4. So its shorter and assymetrical.

Any explanation why the extension to 10 days?

Secondly, anybody know the roots of First-Strike?

Thanks

Forex37
Quoting Forex37
First of all, thanks to everybody helping to keep threads on mechanical systems alive.

Regarding systems popularized by Joel I have a problem: Joel mentioned this system (ONS) to be given to him by Joe Krutsinger. Checking with Krusingers books I find the system there, named ONS, albeit with 2 major differences. Krutsinger is using a shorter liikback ( 4 instead of 10 bars for longs and 8 instead of 10 for shorts) and a directional bias for currencies, expressed by 8 instead of 4. So its shorter and assymetrical.

Any explanation why the extension...

Regarding FS, the roots can be found in Toby Crabel's book.
At the moment I cannot help you with ONS strategy, since up to now I'm focusing on FS.
As soon as I have implemented and tested all my concepts in the FS EA, I will write an EA for ONS and I will be able to test the various ONS strategies/mods.
I will post the results here.
Quoting asasa
Regarding FS, the roots can be found in Toby Crabel's book.
I mean, it's a classic ORB strategy on a weekly time frame.
Here again with a stronger narrow-range filtering for drawdown reduction...
Enter only on really narrow narrow-range setups.
Less drawdown, then push on position sizing.
Results:
Profit Factor: 2.23
Profit Trades: 60.47%
Maximal Drawdown: 25.63%
Of course the equity curve is something that is not replicable on a real account.
By reducing the position size (half the size in every trade), the drawdown reduces to 12.73%.
Attached Image (click to enlarge)
Click to Enlarge Name: asasa_spec_fin_red.jpg Size: 52 KB
I was reading something somewhere that said with narrower ranges and tighter stops, the stops were hit more often. I don't remember much else, so obviously it doesn't really count for anything, but then again we don't know much else about your specific logic in your testing either.

Just something to think about!
Hindsight is 20/20
Quoting cosgrove
I was reading something somewhere that said with narrower ranges and tighter stops, the stops were hit more often. I don't remember much else, so obviously it doesn't really count for anything, but then again we don't know much else about your specific logic in your testing either.

Just something to think about!

In my case, narrow ranges do not mean tighter stops.
Simply, when I find narrow ranges, I increase the position size of the orders. The stops are left unchanged. Of course the risk for the current trade increases.
How one defines narrow range is important as well.
Quoting asasa
In my case, narrow ranges do not mean tighter stops.
Simply, when I find narrow ranges, I increase the position size of the orders. The stops are left unchanged. Of course the risk for the current trade increases.
How one defines narrow range is important as well.
Sorry I don't understand, I thought you were using Frst Strike Plus.
If you are using FSP, then narrow ranges have to mean tighter stops (assuming that we are still talking about the prevoius week's range) since the entry and stop are based on the week's range.

I have the same problem with you taking about having a larger position size when the range is narrow. You appear to consider that a modification - it is not, it is the way that Joel tells people to trade FSP because the risk is defined as a % of your equity , so if the stop is smaller then the position size MUST be larger.

Ian
Quoting ianf0ster
Sorry I don't understand, I thought you were using Frst Strike Plus.
If you are using FSP, then narrow ranges have to mean tighter stops (assuming that we are still talking about the prevoius week's range) since the entry and stop are based on the week's range.

I have the same problem with you taking about having a larger position size when the range is narrow. You appear to consider that a modification - it is not, it is the way that Joel tells people to trade FSP because the risk is defined as a % of your equity , so if the stop is smaller then...

I'm sorry if sometimes I'm not clear...
When I wrote that the stops are left unchanged, I meant that they are always calculated as the same fixed percentage of previous week range.
Of course when the price range shrinks, the stop loss in terms of pips diminishes. How much a pip is worth, it is determined by the position size.
In any case, in this thread we are referring to FSP mods, and it should be clear that I'm not following all the rules of the original strategy.
For example, at this point of the development I don't risk the same amount of equity percentage on every open trade, but I risk more when the price range shrinks.

One note of warning...
Even though my equity curves look really good, further simulations proved that this strategy works so well especially on EUR/USD. On GBP/USD after 10 years of simulation the system results in a small loss. On EUR/CHF the results are even worse. On USD/CHF the results are good, but not as good as on EUR/USD.
My suggestion is to carefully select the pairs based on simulation results and to drop the pairs with poor historical behaviour.
One more thing, in my previous posts I have misinterpreted the value of Maximal Drawdown returned by the backtester of Metatrader, since in reality it returns the maximum absolute drawdown and not percentage drawdown. Therefore, the maximal percentage drawdown is always larger.
Quoting asasa
One note of warning...
Even though my equity curves look really good, further simulations proved that this strategy works so well especially on EUR/USD. On GBP/USD after 10 years of simulation the system results in a small loss. On EUR/CHF the results are even worse. On USD/CHF the results are good, but not as good as on EUR/USD.
My suggestion is to carefully select the pairs based on simulation results and to drop the pairs with poor historical behaviour.
So are your curves representative of all pairs traded together, or just E/U? and also, this is your personal modified system, correct? I will continue to trade it unmodified and on all the majors for now, but thank you for your information!
Hindsight is 20/20
Hi asasa

I am so fortunate to read your post and your FSP mods.

I have backtest FSP about one month ago, Yes. For the original rule set, It work best for EUR, the most trendy pair.

I try different MM, such as increas the risk for the long term trend, or last week's candel direction, etc. The result is improving, but not significent.

Then I come to a point that: technical based MM rules are actually additionally system: let's say we normally risk 1% each week, when we get an NR4(narrow range bar in four days), we risk 2%. This is actually two system: 1. Orignal FSP 1% per week, and 2. trading NR4 with the FSP entry and exit rules with 1% risk.

I believe there are more EA programmers in this forums than sucessful forex trader, could you share some of your mods, you need not give us all your ruleset, give us some indication, helping us geting on the right track is more than welcome.

Thank you very much.


Quoting asasa
I'm sorry if sometimes I'm not clear...
When I wrote that the stops are left unchanged, I meant that they are always calculated as the same fixed percentage of previous week range.
Of course when the price range shrinks, the stop loss in terms of pips diminishes. How much a pip is worth, it is determined by the position size.
In any case, in this thread we are referring to FSP mods, and it should be clear that I'm not following all the rules of the original strategy.
For example, at this point of the development I don't risk the same amount of equity...
Quoting cosgrove
So are your curves representative of all pairs traded together, or just E/U?
Just EUR/USD.


Quoting cosgrove
and also, this is your personal modified system, correct?
Yes, that's right.
The basic system leads to worse equity curves.
I thank you for your nice words.
In fact, my approach is to have one single EA with multiple independent trading systems coded in it, in order to have smoother equity curves. Money management is also applied to the trading systems themselves (how much capital to devote to the various trading systems that are part of the EA).



Quoting aidans
Hi asasa

I am so fortunate to read your post and your FSP mods.

I have backtest FSP about one month ago, Yes. For the original rule set, It work best for EUR, the most trendy pair.

I try different MM, such as increas the risk for the long term trend, or last week's candel direction, etc. The result is improving, but not significent.

Then I come to a point that: technical based MM rules are actually additionally system: let's say we normally risk 1% each week, when we get an NR4(narrow range bar in four days), we risk 2%. This is actually...
Glad to see your thread out of the bin, Geoff.
Did FF have a change of heart?
- a change of rules?
- or just delete posts containing (semi-) commercial links?

Whatever, I like being outside in the fresh air, it gets so stuffy in that bin.

Oh, I see FF have a commercial forum now. Well FirstStrike is still FREE and always has been FREE !

Ian
I think I originally followed this thread, but I'm not sure I already said this..one thing one must never, ever forget when trying to "optimize" an approach that all you will be doing if you're optimizing an EA with the DATA YOU HAVE is that you are effectively just curve fitting something onto what has already happened, not ensuring a more profitable thing for the future necessarily...imho the markets go through periods of both higher and lower volatility, and while yes, indeed there might be a "sweet spot", I am not sure how certain you are to hit it with a given set of data.
The only way to ensure you are maximizing probability of at least approaching realistic circumstance would be to have every tick ever traded on the market you're analyzing, and best I know most tick data from free providers is neither up to snuff nor as extensive as one might like for this.

Regardless, I've been looking at this and trading it on a small experimental account for a while now and been in communication with others and the one single most important lesson for me personally has been: DO NOT TOUCH!

Unless you trade the thing "stupidly" 1:1 according to rules, you WILL have occurences like closing a trade just before it tacks on 2-5% additional had you not interfered. This is as pure a lesson in "hands off" trading system as you will find imho and I am not sure "situation optimized" EAs are very much doing that, quite the opposite.

So I am highly skeptical. The "optimizations" basically say hey we can get you more profit at the same risk levels. It's not that I don't believe optimization is possible, but if at all, I'd like to see it justified from a logic based angle rather than "I optimized it via ATR and time scans over historic data" type of stuff...i.e. if you ask me why FS works, I'll tell you because it's basically putting the definition of a trend into a fire and forget order based approach. That's a logical explanation. The %s for stops and trigger distance as even Joel would say are not quite random, but interchangeable with others that would create a different curve, but also "work".

I guess what I am saying is: the "best case" for me isn't optimization to death, but actual functionality in the first place. Anything above that needs to tell me: 1. what does it change 2. why does it change it and 3. how can I be certain that it is sure to enhance the majority of cases and 4. how robust is it to extraordinary conditions

Also, "never change a running system"

This post is also a humble request to not post historical data equity curves without giving reasons, parameters and explanations about the changes so that others can actually participate in the discussion and so that it doesn't seem like "......and now after the 50th curve, I am offering this super-duper-version of the EA for just..." is coming

Dunno what's commercial about the basic system, either, it's freely available to anyone who has a throwaway email address.
I DO suspect Joel of selling his subscribers' Email addresses, since lately I've been getting a lot of spam from super-EA-trading bots with of course 0% losses and 82475987349867349876% per second made. Has anyone else been getting the classic "faperfx" and similiar spam mails to the "Joel" address?
Maybe someone can confirm my suspicion there..
Trust price. Know yourself.
0 traders viewing now