MJ - MTF Stochastic

I decided to open a new journal - related to my MTF Stochastic testings.

To see related threads started by Spudfyre:

http://www.forexfactory.com/showthread.php?t=37111
http://www.forexfactory.com/showthread.php?t=36574
http://www.forexfactory.com/showthread.php?t=119483


To see related threads started by myself:

http://www.forexfactory.com/showthread.php?t=144270


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

To start with:

1. I use only Stochs and Price actions: no other indicator

2. at the moment I use 10 Test accounts ( might change over the test period) each with slightly different rules.

3. in each test account I test 18 Pairs

 

1. AUDJPY
2. AUDUSD
3. CADJPY
4. CHFJPY
5. EURCHF
6. EURGBP
7. EURJPY
8. EURUSD
9. GBPCAD
10. GBPCHF
11. GBPJPY
12. GBPUSD
13. NZDJPY
14. NZDUSD
15. USDCAD
16. USDCHF
17. USDDKK
18. USDJPY

4. All of that is auto-traded - no manual interaction at all.

--- IMPORTANT ---

1. Much of the background can be found at the 4 Threads mentioned in the beginning - so for anyone interested it is best to read them completely.

2. No point in asking for any MetaTrader indicators or trading stuff: I'm NOT using MetaTrader but an own programmed system.

3. This are new tests so they may fail completely - that has nothing to do with Spud's MTF theory.

4. This should be mainly seen as an addition to the First Journal I started nearly a year ago.

I still have no plan to post exact setups - but just report some additional infos of one individual trader using a highly complex auto-trading system - basic of what Spud shared

I was thinking a bit how I will do that:

1. my plan is to post once a day or so a small statistic of the best overall account (based on Balance and NetAsset Value).

2. and from Time to time a overall statistic.

3. Screen Shots are usually GMT+1


BTW: Start Amount of all 10 accounts was yesterday late night 100.000 EURO. ( from time to time I might reset them)




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

Below is my best account so far: more than 3 % +

Debug Log: Tue, 10 Nov 2009 00:44:35 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: USD/DKK <-<-
Debug Log: Tue, 10 Nov 2009 01:01:36 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: GBP/USD <-<-
Debug Log: Tue, 10 Nov 2009 02:35:09 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: CAD/JPY <-<-
Debug Log: Tue, 10 Nov 2009 04:34:08 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: EUR/NZD <-<-
Debug Log: Tue, 10 Nov 2009 06:12:08 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: AUD/USD <-<-
Debug Log: Tue, 10 Nov 2009 06:39:08 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: USD/CAD <-<-
Debug Log: Tue, 10 Nov 2009 07:10:10 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: NZD/USD <-<-
Debug Log: Tue, 10 Nov 2009 09:11:07 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: AUD/USD <-<-
Debug Log: Tue, 10 Nov 2009 09:34:08 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: EUR/AUD <-<-
Debug Log: Tue, 10 Nov 2009 09:35:10 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: USD/JPY <-<-
Debug Log: Tue, 10 Nov 2009 09:44:10 >-> BUY ExecuteNewOrder NEW TRADE !!! Pair: GBP/JPY <-<
Debug Log: Tue, 10 Nov 2009 11:36:06 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: USD/JPY <-<-
Debug Log: Tue, 10 Nov 2009 11:36:05 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: AUD/JPY <-<-
Debug Log: Tue, 10 Nov 2009 11:36:08 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: NZD/JPY <-<-
Debug Log: Tue, 10 Nov 2009 11:48:09 >-> SEL ExecuteNewOrder NEW TRADE !!! Pair: NZD/USD <-<-


Below one of the entries

So I did my first day.

1. No Account in negative.
2. 3 Accounts had no Trade till now


BEST account REPORT: a Bit more than 6 % +

had 27 Trades: only one was a real disaster (this still happens



Some thoughts:

* Guess it was one of the better days for the system (don't know as this is a new testing time)

* 1/2 of the Entries where very exact - because of that I could have made much more with a different Balancing of the Units ( I split each trade into 4 sub-trades and the first one has just about 20 % of the units. With an exact entry just this small one is executed. if it works that well - I might allocate a bit more to the first entry)

* Notes on Number of trades: 27 trades are excellent but one has to consider that some of them are quite similar: like on the post above at about 11:36

In the future I might limit such to a singular entry - which will reduce the profit but also reduce a possible loss.



ALL TOGETHER an excellent start - but I would not expect it to continue like that.



Below are the Account Statistic and the disaster trade.

BTW: still open position as I wrote.

Well, over night until midday I had an Internet connection breakdown.

So this messed up some bits.


I used this to implement that option to filter similar trades execution at the same time. Will test that also in comparison.


Had also some SL: The Account from Yesterday went down to 104857.43 Still extremely well.






An Other Account took the overall lead: with all together over 7% +

Not much to say.

The implementation of my Similar Pairs Filter was more complex as first thought.

So I programmed the whole day. After the Internet break down I also did not start my trading bots anymore.

So no changes.




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

I did reset all of the Bots to 100.000 EUR Start Capital




I configured for all 10 Bots the new Similar Pairs Filter.

Hope I did not introduce too many new bugs. Will see after this night.

All bots are now up and running.

I aslo did some small other reconfiguration: splitting each trade into 6 sub trades and gave the first one more weight


Anyway - through the new Similar Pairs Filter I expect less trades - will have to see how that works.

Found already a couple of new bugs - fixed them.

removed also 4 Pairs from my list - reason quite an increase in the spread over the European night.:
Removed: AUDCAD, EURAUD, AUDCAD, EURNZD which means I just track 18 pairs per account.

3 Accounts - No Trade
3 Accounts slightly in Minus
4 Accounts in Plus


BEST Account: nearly 2 % up

TOTAL ACCOUNT REPORT: START Balance:100000 END Balance: 101954.90



Similar Pair Filter seemed to have worked quite well so far.


Had a new idea for slightly adjusting the Units size based on a different Risk count option

fixed a few other bugs.

Implemented the new idea: additional new Risk Count option - still test for obvious bugs.

Depending on configuration - This will further reduce the Unit Size but hopefully increase the overall profit.

If that works out as planned - I'm very satisfied with the process so far - and can start serious testing of the new stuff.

Some more programming - found a few more bugs which I fixed.

Have a good feeling most of it should be working again.


BTW: All negative Accounts moved into Profit.

Well, found some more bugs to fix. That's life.

Anyway, this week were to many changes so I have to disregard any statistics.


Hope next week to restart Testing in a more proper manner.



All together I'm very satisfied with this weeks work. Have laid a nice base for the new test engine.

Closing down everything for weekend.


Next week will be more fun.


Actually I'm quite confident and will apply a subversion of the new stuff to one of my Live accounts too.



Below the Last Trade of this week - excellent entry so that not even the 2. Sub trade was triggered.

AIM - FOR THE EXPERIMENT (per Best Account)

* RETURN: 150 % a year PROFIT
- Which calculates down to about 1.8 % per week compound.

* a minimum of: 15 Trades per week - more trades are better

* First Test Run 3 weeks till 4. Dec: Required End result: 105497.78
on best account.

Start 100000
Week 1 101800
Week 2 103632.4
Week 3 105497.78

Mj,

Will follow along. I picked up Spuds Predictive MTF about a year ago. Demo trading and tweaking to my style of trading up until about August. Since August I went live with it and still demo trade also (very aggressive on the demo). I believe your 150% is do able. The shorter time frames is what I have been successful at. 15,5,1m

hi swenk,

I'm honored to get you very first post

Seems we both are on a similar way. I also came across Spud's thread about 1 year ago.

Actually was quite lucky that at that time Spud had start posting again after a year of absence.

After mixed results in the beginning - I got my system to a quite profitable way.

The main thing I do not like is that I filtered so much that I got just very few trades - which requires a high leverage.


So with my new testing I hope to keep a good win/loss ration and increase the number of trades.

Also a did a lot of re-programming and clean up of my coding.

For next year I plan to switch all my trading to MTF Stochs systems.


hi swenk - do you keep statistic of your trading - like number of trades, success, failure and so on ...

MJ,

I will compile my stats from the demo and live accounts over the weekend and post back here by Sunday evening US time zone. I know the challenge for me has been the lack of time that is needed in front of the screen to catch all the trades manually. I trade 8 pairs actively and watch 4 more to total 12. I know the strike rate is very high and consistent with the setup the MTF gives me. Due to other obligations (day job) the Asian session is where the bulk of my time was spent during the first months of trading the MTF setup. Although I could clearly see and record all the setups during the Euro and NY session I didn't have the time to invest in those areas. Since August I have been able to capture a few key hours of trading those areas where
more volume is present. From what I have traded and recorded it is very satisfying to see the performance no matter what type of market direction it is.
Will be back Sunday

swenk Thanks,

looking forward to your data so far.

My case is a bit different than yours: I do not work a other job and my trading is fully automatized.

Which has advantages and disadvantages.

One of the disadvantages is that my 'system' lacks the brain indicator and as such has to work on pre-defined complex rules for entry and exit. Because of that I also miss out quite a lot of profit because 'human judgment' of a particular situation is missing.

The positive side is: I can trade 24 hours and still sleep like a normal person. Also the rules and filters can be so complex that it would not be possible doing it manual - this is actually the case.

Recently I did implement something like an alert - which I could use to get me going in a similar way but with manual interaction.


Anyway, my aim was always fully automatized trading and Spud's basic idea has proven to be quite workable.



Wish you all the best for next week

and maybe we can profit from our different approaches of the same basic trading idea.

Coded a complete new Additional Time Scheduler.

This is not immediately needed but gives me a new flexible way of completely shutting down for a certain time period - like news time or or weekends. Before I just had Weekends execution block but not full shutdown.

I hope it works as planned as I can not test it sufficient on weekends. Will see tomorrow.


ANYWAY -- looking forward to the coming week.

MJ,

Here are some quick stats from the 8 weeks I have been live with the MTF

150 trades taken
135 wins
15 losses
account positive 19%

Demo account which I stated was traded aggressively
12 weeks
217 trades
210 wins
7 losses
account positive 395% (I don't personally believe this is true account of the type of trading I usually do with the live money.) I had a lot more risk in these trades. Only used for testing certain settings and filter work that is evolving and is always a work in progress.
This week I will post a few trades taken live (time, entry, and results). That will allow you to see some of my results. Hopefully this will improve or satisfy some unknowns with in this system. My your week be prosperous.

Thanks swenk,

If you don't mind I have a couple of questions:

1. which pairs did you trade
2. How come you have live in 8 weeks 15 looser and in demo in 12 weeks just 7
What makes the difference.


All together you seem to have a slightly better win/loss ration than I ( which is expected by manual trading).

Hey MJ,

Primary pairs: AU, AJ, EU, EJ, GU, UCAD, UJ, EGBP
Minor: GJ, NU, NJ, UCHF,

Since my trading time is sometime limited to Asian I trade AU, AJ, EJ, EU

I have a tendency to micro manage my live account sometimes (bad habit) opposed to the demo I set a 15 to 25 pip TP and move on. My stop loss is always set 10 pips above the high of the candle that was part of my entry setup.

Trades that where taken live at opening of market
UCAD 0:15 S @ 1.0535 TP +17
EU 2:05 S @1.4955 TP +4 Exited due to lack of momentum at 4:05
EU 2:25 S @1.4954 TP +5 Same as above
Sunday evening trading can be tricky at best sometimes with lack of momentum. MJ has that been a factor in some of your win or loss rate?

Other entry's marked:
AU 02:00 S @.9349
AU 7:25 S @.9348
UCHF 0:15 S @ 1.0089 poor momentum on this pair Time of the market.

One other observation in trading this MTF is when you have a established trend I am more likely to shorten up the TP and SL when going against the trend along with the placement of the longer tf stoch. Will check back later this evening