Journal related to: Spud's MTF Stochastics

Eliminated a view configurations and reconfigured 42 accounts. I plan to let the same configuration run for 2 weeks in a row without adjusting.

This should produce about 2000 - 3000 more trades (of course many of them a very similar).


If everything goes overall well - I properly will apply one version of it to a small live account in March.

The other plan is: if it goes so far well I will move on to test one other pair for a couple of weeks.

Until now I always checked manually how many trades hit TP or SL.
This took me more than 3 hours each day. (depending on the number of Accounts running - next week I have more than 80)

Well over weekend I wrote a excellent helper program which will simplify that a lot. Will crawl by itself through a list of accounts and extract the infos.

The program needs just about 5 min to extract the data for over 80 accounts http://www.forexfactory.com/images/icons/icon14.gifhttp://www.forexfactory.com/images/icons/icon7.gif.

Still insert manually into a spreadsheet.

Nice reults again MJ

Looks like you're getting closer to achieving your goal - Which of Spuds methods are you using in your bots

hi CindyXXXX,

Most of Spud's methods are really the same just differently explained.

I implemented it in my program in an advanced way - so I can flexibly configure most of it.

I think at the moment I put most emphasis on different variations including the rope.

I also skip still a lot of trades: all the elastics -- not because they are not profitable but because of a bit 'lazyness' on my side to implement them.
But still I might look at them later on -- also in the light of: bringin sexy back - stochastics http://www.forexfactory.com/showthread.php?t=13321

There is one thing I still might want to implement: At the moment the lowest defined time frame is used to decide the trades direction: which has advantages and disadvantages.

One of the disadvantages is that it does not nicely fit into Escalator to Pips. So I might still pull myself together to do a work around -- but this is quite some work.

Bottom line: it would not be possible to trade manually the current implementation. too complex.

TOTAL TRADING PERIOD: START 28.January 2009

START Balance: 100000.00
END Balance: 104511.06

Trades: 20
Profit: 16 | Loss: 4


only small max DD till now < 2% -- never turned below initial deposit.

hi mrinnox,

I see you quite often around here and have a question: do you trade Spud's MTF?

When it runs it runs: No new trade on the tracked account but overall this first half week again excellent. I'm still not sure if too much filtering will really greatly improve the end results.

Even though the win/loss ration by same TP/SL settings is higher but there are just about 10% of the trades.


Well, I decided already on the next currency pair I'm going to look at: USD/JPY.

10 First USD/JPY accounts configured.

Will be interesting to see if it works out similar to the EUR/USD.

First night was very good. See by the filter some similarities to EUR/USD testing.

Misconfiguration in 2 accounts. So I will reset all USD/JPY and restart later on.

Anyway much to early to say anything as it was just over the Asia session.

But it produced in nearly all accounts about 1 % profit.

Had an idea for an additional TP adjustment option which I plan to implement to test out.


Some statistic incl weekend:

EUR/USD: 1 year move in PIPs: 3711 Average daily move in PIPs:158
USD/JPY: 1 year move in PIPs: 2357 Average daily move in PIPs:120
GBP/USD: 1 year move in PIPs: 6896 Average daily move in PIPs:199

To implement the new additional TP /SL option was more work than expected. Hope I did not introduce to many new bugs.

Well, I will have to reconfigure all bots as I removed some old options and reimplemented them differently.

I do not expect major result changes from this changes: but will test alongside the old settings if there is any improvements.

The new additional option allows for a more dynamic adjusted TP / SL settings based on recent movements.

This could have certain advantages but also disadvantages like when there was a strong move and it suddenly flattens.

Found already a good number of new introduced bugs.

Also had a Internet problem over night.

So this week - was too much interruption. Still hit the New High of the tracked account.

This is one of this days: Internet goes off and on.

I waited 2 days for a signal on the tracked Account - and when it came I had a calculation error (new bug from the changes yesterday http://www.forexfactory.com/images/icons/icon8.gif) so the Unit Size was very small and Profit only a couple of EUR. Correctly I would have jumped over 105000.



Anyway New High: http://www.forexfactory.com/images/icons/icon14.gif

TOTAL TRADING PERIOD: START 28.January 2009

START Balance: 100000.00
END Balance: 104539.23


only small max DD till now < 2% -- never turned below initial deposit.

TOTAL TRADING PERIOD: START 28.January 2009

START Balance: 100000.00
END Balance: 104539.23

Trades: 21
Profit: 17 | Loss: 4
WIN/LOSS Percentage: WIN: 80.9523809524 | LOSS 19.0476190476

PS: because the transaction gets longer I attach it as a exported excel:

Because of the new additional TP /SL option, which allows for more exact comparison of different pairs (at least hopefully).

I added for testing a number of new pairs:

* Main Pair is still: EUR/USD (only one on the account I track here)

* New Test Pair: USD/JPY

and then for a start 8 additional pairs with just one account each.

-- so there are 10 pairs with similar settings -- for comparison reason.

One of the new Test Pairs: just post it because it had a nice run.

4 times very nice short trade: USD/CHF

Just for anyone who is interested: Not sure if I mentioned that before: I use for this testing 'Price and as only indicator the Stocastic'. No additional indicator to filter like Spud's last addition: The Dynamik Filter (http://www.forexfactory.com/showpost...postcount=1328)

In my case it is pure Price and Stochastic which I use to decide when to enter and exit.

Funny day - normally I get on EUR/USD about 10 - 15 trades with low filtering: today only 2.


To contra the nice post from USDCHF:

USD/JPY test is really bad today: some accounts down about: -6.5%.
(well I have to say there is not much MoneyManagment applied nor much filtering - but still bad:

One of the worst one:USD/JPY

REPORT

START Balance: 100000
END Balance: 93639.48

Trades: 18
Profit: 8 | Loss: 10
WIN/LOSS Percentage: WIN: 44.4444444444 | LOSS 55.5555555556



It looks better for the more filtered accounts:USD/JPY


REPORT

START Balance: 100000
END Balance: 100458.78

Trades: 1
Profit: 1 | Loss: 0
WIN/LOSS Percentage: WIN: 100.0 | LOSS 0.0

New High: http://www.forexfactory.com/images/icons/icon14.gif

TOTAL TRADING PERIOD: START 28.January 2009
START Balance: 100000.00
END Balance: 105051.06


only small max DD till now < 2% -- never turned below initial deposit.


Was a very nice safe entry.

Well, I had an idea how to improve some parts of my base automatic trading system.

so I've started to work on that. This is more in the underlying base system - not so much in Spud's trading implementation.

But it will (if it works out as planned) improve
*certain speed issues as well as giving
* some new scheduling option (which for some bot system might speed up the programming)
* And for some bots reduces data fetch overhad (which is the main point why I do it)