Journal related to: Spud's MTF Stochastics

First look at one of the 'NEW Escalator to PIPs' subversion:


StartDate: 04/17/2009 EndDate: 04/23/2009


TOTAL ACCOUNT REPORT: START Balance:100000 --> END Balance: 106875.04

Trades: 35
Profit: 31 | Loss: 4
WIN/LOSS Percentage: WIN: 88.5714285714 | LOSS 11.4285714286



EXTENDED ACCOUNT REPORT: ------->>>>>>: USD/CAD
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: GBP/JPY
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: NZD/JPY
Trades: 3
Profit: 3 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/CAD
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: USD/CHF
Trades: 3
Profit: 2 | Loss: 1
WIN/LOSS Percentage: WIN: 66.6666666667 | LOSS 33.3333333333

EXTENDED ACCOUNT REPORT: ------->>>>>>: NZD/USD
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: AUD/JPY
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/AUD
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: NZD/CAD
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: USD/JPY
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: GBP/CHF
Trades: 2
Profit: 1 | Loss: 1
WIN/LOSS Percentage: WIN: 50.0 | LOSS 50.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/GBP
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: AUD/CAD
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/NZD
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: GBP/USD
Trades: 3
Profit: 3 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: CHF/JPY
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/CHF
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: USD/DKK
Trades: 1
Profit: 0 | Loss: 1
WIN/LOSS Percentage: WIN: 0.0 | LOSS 100.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: CAD/JPY
Trades: 2
Profit: 2 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/USD
Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

EXTENDED ACCOUNT REPORT: ------->>>>>>: EUR/SEK
Trades: 1
Profit: 0 | Loss: 1
WIN/LOSS Percentage: WIN: 0.0 | LOSS 100.0

NEXT WEEK TEST PLAN:

will still continue my 4 test accounts with refining some configuration on my main Stochs system (which I also use live).


next to this I have 3 test accounts with a more EscalatorToPIPs config.



The real tests on the new 'EscalatorToPIPs' configuration are planned the week after next: Starting on 2nd May.

Than I plan to stop all tests related to my current MTF Stoch system: And are having more resources free for 'EscalatorToPIPs' tests.

What I see now - does not look too bad - and might work out with enough testing.

Got enough trades so that I can play bit with more filtering rules



Also look how much I could increase my TP. And there is quite a good number of trades like that - which will help in the win/loss ration. But too early to say.

But to show it in a prober light: the real plain bad ones:


One can see: if it goes bad - it really goes wrong -- so the question will be: can this be filtered - if not the next question will be if there are still enough good trades to make it worthwhile.

Well, the new tests are not doing to well so far.

But as I said -- too early to make any conclusion - still a lot of good signs among it.

Hi M_j,

It's a couple of weeks that I've come across Spud's MTFS Strategies. After reading through 101 pages on PMTFS thread, I found myself with several technics/strategies with different rule-sets.

Just wondering which one did you find the best and used in your first bot.

Could you please explain your best findings in shape of some rules that when you enter a trade and when you exit?

I appreciate your time in advance.

PipPicker

hi PipPicker,

First: congratulation for making it through the 101 pages


-------------------------

I got recently a number of similar questions: so my short thoughts.


1. The 'real' right person to asked would be: Spudfyre (but it seems that he is since a few month not around)

2. The purpose of this Journal: my Idea as seen from the first post was to have a 'joined effort' of a 'proven' transparent trading history.

For different reason it did not work out.

I than changed it and thought - as I was 'anyway going to to some testing' Why not post the trades transparently.

http://www.forexfactory.com/showpost...4&postcount=72

the purpose is only to show transparently the profitability - I will not go into exact setups as there is enough basic info on Spud's threads as well on my different postings here.


This is grounded in a couple of reason:

a. Spud did such an excellent job that anyone can start playing with the underlying idea. (no point to repeat all again)

b. Many traders are 'strategy jumpers' and as such they are looking for a 'fast, free 100% system' without taking the burden to 'really invest time if there are reasonable ground that something could work out. (I'm not saying that there should not come once a time where one makes an decision to abandon a testing strategy - I wrote about that previously too)
But I'm an advocate of 'own thinking and testing' if there are grounds to belief that a system could work out. One reason e.g. my Broker does not offer 4 hour charts.
So for instance I have NO 'rule-set' which includes 4 hour time frames


__________________________

So yes, Spud gave enough background that one can start out. I hoped to give the 'prove' with an transparent trading log -- that it can work out (at least during the ca. 3 month of my Journal tracked account: about 10 %).

------------------------


That was just some general background:

BUT more directly to your and other peoples questions:


1.

Quoting PipPicker

Disliked

Hi M_j,

I found myself with several technics/strategies with different rule-sets.

Just wondering which one did you find the best and used in your first bot.

Ignored

I basically where a bit lucky - that when I came across this thread Spud had come back after about 1 year or so absence.

So I did get into some personal interaction and started out on a similar question path like you.

Especially because there is a basic idea ( which he expressed in different ways and rules-sets I thought it could work out)

---
which one did you find the best and used in your first bot

Probably I give you a not satisfying answer now: NONE

I started out in a simple way to implement some aspects - with much more flexibility.

I did 3 or 4 bot systems each with other aspects of what he shared.

But by now I focused on the basic idea and implemented it in such way - that I actually can configure nearly 'all rule-set' what Spud presented - But also many others and freely combine them.

What I'm saying: what I'm doing is NOT possible in manual trading. It is just too complex: I did not count in the end how many parameters I track if I configure all.

But there are some basic thinks (I implemented) which I can share:

* I can configure 6 time-levels flexible - I found that more than enough (mostly 4 or 5 essential are sufficient in my testing)

* each or the 6 time-level: has at least implemented (but might not have it configured all)
a. PriceLine
b. OriginLine
c. some kind of the Rope

This is very much simplified the basic.

(for instance I can weigh the counting strength for each rule ect .. also I can for instance configure this to represent something similar as the 'EscalatorToPips'.)

#########

Probably may main question would be: Do you plan to trade such MANUALLY or program an automatized version??


1. For manual trading: I personally would go with a screen maybe similar to the picture:
http://www.forexfactory.com/showpost...&postcount=751

But note that I do not mind the rules or time frame:

1. Basically if I do manual trading it: PriceLine, OriginLine, Rope: and looking at all three on different time frames. No fixed 'rules'.

But be aware: it is a bit tricky -- there are many good signs -- but sometimes even with all this automatic trading I programmed: There are just PLAIN WRONG 'signals'.

______________________________________________

2. For automatized trading: If that is your plan: I would suggest something similar to what I did (actually I think I made it right).

Highly configurable - with nearly all possible option freely to combine.

So one can play with 'strict settings'


2.

Quoting PipPicker

Disliked

Could you please explain your best findings in shape of some rules that when you enter a trade and when you exit?

Ignored

I hope some of what I said above answers this question partially (but I know not to your satisfaction - sorry).

I can say a bit more of my current approach:

I had implemented in one of my early bots the idea of opening an trade(according to some rules) and existing by moving from one time frame group to the next. There is actually in the other thread a very detaild explanation what I had implemented.

My personal experience was: that it was not the best in my automated system. (maybe if one trades manually that would be still useful as Spud demonstrated a view times).

So, I settled for a TP and SL option. As soon as I enter a trade: I enter it with a defined (flexible) TP and SL. So, my exit is really simple: SL hit or TP hit.

As seen in the number of pictures a few posts ago: in my 'NEW' configuration tests more like the Escalator To PIPS' I would have much more room to increase my TP settings in most of the cases.

e.g. when I first started testing I used only EUR/USD and started with TP 2 pips and I think it was 100 PIPs SL.
after a good number of trades I saw that most of the time I could get more PIPS TP on the same setup - so I increased which improved by the same Win/Loss ration my profit (or better decreased my loss

Now I implemented a much more flexible way of autocalculating. So I do not have any more a fixed PIP setting. Same goes for SL.

---------------

mainly the question would be: manual or automatized - which would determine my approach to 'a kind of rules-set'.


A other thought I just had:

I for instance removed the 'elastic trades' (PriceLine opposite from OriginLine). I had implemented them at one time but removed it for different reason:
1. simplicity in programming flow (but that was not the main reason)
2. I reasoned that the other trading setup would be 'safer': e.g. in the light of a possible Escalator To Pips setup. I think Spud 's writing do imply something along this lines - (but see also my next post)

As you see - playing with the basic idea is what I do: and I'm sure Spud does trade quite differently. (in fact he surely does as he can not trade manual what I do automatically)

But the basic idea is the same - adapted to a 'hopefully' working set: in my case: No elastic trades can be configured (I removed such option), I could configure theoretically configure a 4 hour time-frame (I can set any time frame by seconds) - but practically my broker does not offer it.

And such ADOPTION PROCESS is at times boring, annoying, frustrating with mixed results at TAKES TIME -- but might be rewarding in the end

And one more thought: if you did not do till now: you might also like to look at this thread Spud did refer in one of his writings:

http://www.forexfactory.com/showthread.php?t=13321

Thanks M_j,

Such a great and long answer, also I could easily hear frustration from your words which might come from some stupid traders expecting a trading system with no loss.

I'm not one those just complaining about systems and never stick to principals. I'm a technical guy and because of current life situation, I can't spend time trading manually which has led me to automate it into an EA.

Your intensive effort over long time is what I really admire along transparency you have deliverd.

I also do appreciate Spud's thoughts and ideas. They are the best analysis I've ever been across, no discussion.

But as you know once trading comes to programming an EA, you have to mention every single bit, simplify some rules and/or add some additional factors that will get into action automatically by human brain a.k.a. Brain Indicator.

Your experience with combining rules and putting them into code is what my actual question is about. You have already answered many of them. Just one more here If you don't mind:

How do you calculate TP and SL (on fly)? what's the formula and/or factors you are using in this regard?

About that thread, I will definitly look into it soon even though it's 160 pages . If you have found something interesting in addition to Spud's system, please mention them.

PipPicker

hi PipPicker,

No problem at all asking

I'm not sure which system you use to program the EA.

So: So if you plan to do it for MetaTrade - I don't know enough - but I doubt it will be as flexible as what I do.

BUT having said that - it HAS NOT to be as flexible as mine -- there is no fixed rule how to play with the underlying basic idea

###########

1.

Quoting PipPicker

Disliked

Thanks M_j,

I'm a technical guy and because of current life situation, I can't spend time trading manually which has led me to automate it into an EA.

Ignored

Welcome to the 'club'. Automatizing a trading system has surely some great advantages (like I can enjoy the nice sunshine while still making (or losing) some PIPs).

But I also found it to be an quite burdensome thing: One can see that in Spud's thread when I tried to but into 'programming terms' what HE did more through human observation and experience.

2.

Quoting PipPicker

Disliked

SO YES: I understand what you mean by:

But as you know once trading comes to programming an EA, you have to mention every single bit, simplify some rules and/or add some additional factors

Ignored

And that's the headache to think of what could go differently, where do I need an extra parameter to avoid so many things which could happen -- a manual trader will most of the time not even 'think consciously about such'.

just to give a small idea: my current implementation of Spud's MTF stochs is a code with about 23.000 lines -- not very compact written and also includes a good amount of 'info' and some explanations which are not directly code.

(but that does not include much of other more general code: like stochatic calculations - general trading routins like placing orders - canceling orders MoneyManagement ect.)


##########
3.

Quoting PipPicker

Disliked

How do you calculate TP and SL (on fly)? what's the formula and/or factors you are using in this regard?

Ignored

I don't want to go throught the exact details as this is quite complex.

But I have a number of different ways I can configure:
e.g: FIXED PIP,
SL based on FIXED Pips added to the PricePeak (similar to what Spud did refer sometimes)

At one time I had implemented a different PIPs calculation based on probability strength: e.g I have configured to track 100 parameters:

Some are essentials, some are confirmation only etc... each of this parameters (rules) can have asigned different count strength.

If a valid signal exists I check how many percent of the total possible count strength are reached - and based on that I set different TP / SL.

OR - I use different levels of 'essential' rules:

if only essential rule 1 is reached: I set a TP so much, if essential rule 1 and 2 is reached by entering a trade I set such TP.


To make it easier to apply to different pairs: I have also a percentage settings: e.g. I take as my base movement the last 20 (or 100 or 1000 ...) 5min candles (or 1 hour or whatever timeframe candle).

I calculate the average movement in such time period. based on that figure I calculate a certain percentage for TP and SL.

SO what I'm saying - there are many many ways one can do it - and I have a good number implemented.


##############


I don't know how good you are at programming, what platform you program for - what does such platform offer ...

BUT if you are 'REALLY' new to forex and trading as such - and also new to programming I would suggest starting simple.

##############

For instance I would suggest: (I'm personally in the process of doing this as a new 'system' variation'):

Implement a simple ESCALATOR to PIPs approach to start with.

http://www.forexfactory.com/showthread.php?t=36574

Get some additional parameters: like to be aple to configure more than the 4 time frames what Spud suggests.
Mabe some additional parameter. like being able to use other timeframes.

I actually think (but not tested) that something along this line could work out with patience. And based on this approach alone - it would be much much easier to program than what I did.

(only a couple of OriginLines) Stochs levels and a view other parameters to start with.

Mybe one thing would be than to add an option to also configure a PriceLine (or on all timeFrame levels additional PriceLines )...



Like I said don't know the surrounding circumstances of your broker api, programming skills, trading experience as such ... But maybe this helps a bit.

I do it somehow along this lines: and than I test with different settings and observe (statistics ect) if I find more favourable settings....


##########
4.

Quoting PipPicker

Disliked

About that thread, I will definitly look into it soon even though it's 160 pages . If you have found something interesting in addition to Spud's system, please mention them.

Ignored

Not sure if that other thread is that much important in relation to Spud's MTF stochs approach. But I skimmed over (not all 160 pages in the beginning because it is in somewhat related.

Well, I have many interesting addition which Spud never mentions (like I shared a bit above). BUT basically I did implement quite closely based on Spud's presentation. (with a more complexed base of defining flexible rules - as you correctly said: if one wants to automatize such thing - one needs so many other things to think about - a experience human does not need hundreds of parameters for quite obvious situations)

ANYWAY: wish you all the best in your efforts. I'm surely interested in any progress and especially in any statistics or so.

If I can give you a suggestion: if you are seriously going to do that - open your own Journal - that might help you also to focus as well as others to 'learn' from your success or failures. (that does not mean one has to expose all 'parameters in detail'



Just to show what I mean by testing:

THIS week: Escalator To PIPs related worst account.

TOTAL ACCOUNT REPORT: START Balance:100000 END Balance: 82491.03

Trades: 57
Profit: 39 | Loss: 18
WIN/LOSS Percentage: WIN: 68.4210526316 | LOSS 31.5789473684

But in-spite of that I still see enough reason to invest more time into trying to make it work.

The only trade this week (so far) on my older tested MTF configurations: (similar to my live one)

A quite nice (relatively secure) trade:

BTW: no transaction log at the moment as my 'new test EscalatortoPips related' are not yet advanced enough to make it useful.


##############

FINISHED for this week: The second trade this week on my older tested MTF configurations: (similar to my live one)

A relatively secure trade too:

FAZIT: two nice trades - both successful.

At the moment no point to post the transaction log: as I'm in too early tests for a method based on EscalatorToPips: Best account this week in plus with about 1.4 %.

All others are negative.


------------------------

But my more tested method worked out fine again: REACHED NEW HIGH in live account: Since going live - all trades successful. (Hope it keeps like that

Thought I share my last month trading in one of my accounts using the 'older system'.


It is one month since I did reset my account: so I thought it would be nice to see how I did.

http://i40.tinypic.com/29yrqfm.png

NO LOOSER at all - about 4% - SATISFIED - there would be quite some more possible


If I WOULD HAVE KNOWN http://www2.oanda.com/ubb/wink.gif that I have no looser I would have used a 100% leverage http://www2.oanda.com/ubb/wink.gif and made quite a fortune


--------


below the last trade: wich I closed manually - (it would have hit TP later) - I was not sure and I do not like open trades over the weekend.

This week so far one trade:

Which means: since I when life mid-april - no loosing trade.

MOST RECENT: Which means - still no loosing trade since I did reset on 17 April.


Some of my new once EscalatorToPIPs based: are this week doing nicely too.

Over 10 %


~~~~~~~~~~~~~~~~~~~~~~
TOTAL ACCOUNT REPORT: START Balance:100000 END Balance: 105610.25

Trades: 17
Profit: 15 | Loss: 2
WIN/LOSS Percentage: WIN: 88.2352941176 | LOSS 11.7647058824


~~~~~~~~~~~~~~~~~~~~~~
TOTAL ACCOUNT REPORT: START Balance:100000 END Balance: 113155.68

Trades: 59
Profit: 50 | Loss: 9
WIN/LOSS Percentage: WIN: 84.7457627119 | LOSS 15.2542372881

I started to take the 'NEW' tests based on EscalatorToPIPs a bit more serious this week with some changes made to my configuration.

BEST ACCOUNT SO FAR This WEEK over 17%:

TOTAL ACCOUNT REPORT: START Balance:100000 END Balance: 117805.99

Trades: 26
Profit: 25 | Loss: 1
WIN/LOSS Percentage: WIN: 96.1538461538 | LOSS 3.84615384615






My active system: Still going nicely

Just for anyone who did follow along!


I will discontinue my journal.

I do not see any motivation for it any more - it has fulfilled it's purpose.

Basically I have a live account up with a system based on Spud's MTF stochs.


I'm still experimenting with some other approaches of what Spud shared: like EscalatorToPips based systems.


WISH EVERYONE - MUCH SUCCESS IN YOUR TRADING EFFORTS.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

JOURNAL CLOSED !!!

Hei welches rauf MJ

Me speak many languages but English is me #1.…
I see this thread is down. I fallowed some of your comments at other threads with Spudfyre.

I have been demo trading for about 4 months now. Started with MT4. Got to where I could work it with my eyes wide shut.

Couple weeks ago I found an ECM broker. They use their own platform.
I’m still learning how to use it.

I bumped into Spudfyre’s MTF Stochastic Methodology about two weeks ago.
Last Wednesday and Thursday I used it along with learning the new platform.
Got to where I could enter buy, sell, place stops and exit.
Learned how to enter and exit using Spudfyre’s MTF Stochastic Methodology as well.

Today Friday 28 Aug. I used it.
To say the least I was impressed with the results.

I realize you are using a different approach, but I thought you might be interested

My broker does not provide demo statements. So I took screen shots of my last two trades today. GBPUSD and EURUSD.

Numbers 1 to 8 are my entry points. 9 to 17 are exits. Number 18 gave me a start but I added a couple positions and exited just fine.

I did not trade number 19. It started too late in the afternoon for a Friday..
I thought..
Next Friday if a trade develops around the same time frame I will take it.

Not all of my enter and exit were as accurate as my last two that you see here.
The more I traded the better I got.

Overall results: Upteen trades and zero, nada, kein, keine, keiner, keines losses.
We’ll see as time passes….

Just a note: Even thought this thread has been closed: I just want to put some info about a follow up journal I just started.


http://www.forexfactory.com/showthread.php?t=205809