Middle analysis - Taking home the Pips
Moderator: moderators
Here is an update on the position calculator - redefined the pip spread to reflect more accurately stdev of pairs. Enjoy
- Attachments
-
- forex_buy_position.xls
- (23.5 KiB) Downloaded 311 times
Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.
Thank you for your support.
- RicG
- rank: 150+ posts
- Posts: 274
- Joined: Tue Jul 24, 2007 11:44 pm
- Reputation: 8
- Location: Nashville, TN
- Gender:
Hi all,
I've been paper trading Pitboss's system as outlined in this thread now for seven weeks. I've been "trading" 10 lots of each of the four pairs (USD/JPY, GBP/USD, EUR/USD, USD/CHF) when they meet the criteria of the system. For the seven week period ending last week the system went 56-34, for a 62.2% winning percentage. The 10 lot size netted $9503 not including commissions.
I also paper traded the system without using the Pitboss "filter" that factors the middle by 1.002 for a short and .998 for a long. Trading it that way ended up at 83-56, a 59.7% winning percentage. The 10 lot size netted $13,352, not including commissions.
It certainly seems that he's come up with a winner here. The only downside I see is that during the seven week period there were some fairly drastic drawdowns. It would lose heavily for 2-3 days in a row (during strong trends), but then would bounce back with a vengence. That's where trading psychology comes in - you have to trust the system even when your account is getting hammered.
I'm not recommending that anybody trade this system (standard disclaimer) - I just wanted to give everybody the stats from my experience.
Great trading to all,
Ric
I've been paper trading Pitboss's system as outlined in this thread now for seven weeks. I've been "trading" 10 lots of each of the four pairs (USD/JPY, GBP/USD, EUR/USD, USD/CHF) when they meet the criteria of the system. For the seven week period ending last week the system went 56-34, for a 62.2% winning percentage. The 10 lot size netted $9503 not including commissions.
I also paper traded the system without using the Pitboss "filter" that factors the middle by 1.002 for a short and .998 for a long. Trading it that way ended up at 83-56, a 59.7% winning percentage. The 10 lot size netted $13,352, not including commissions.
It certainly seems that he's come up with a winner here. The only downside I see is that during the seven week period there were some fairly drastic drawdowns. It would lose heavily for 2-3 days in a row (during strong trends), but then would bounce back with a vengence. That's where trading psychology comes in - you have to trust the system even when your account is getting hammered.
I'm not recommending that anybody trade this system (standard disclaimer) - I just wanted to give everybody the stats from my experience.
Great trading to all,
Ric
(Disclaimer - This post is for educational purposes only. Always consult a licensed investment professional before taking any trade. Any trade you take is at your own risk.)
"Does "$ Risk Each" on the spreadsheet mean you set a SL at that price level? "
Was an approximation of average losses. Have since refined with a stop loss and take profit. The stop loss is 1.65 standard deviations which represents 90% of probable senerio outcomes. The take profit is 1.96 standard deviations which represents 95% of probable senerio outcomes.
My statistical analysis shows that the results are 'virtually' identical. Eliminates the far extremes which annoy the crap out of me
"Also, can you explain how you decided on which number to use for "Pip Spreads"?"
The pip spreads use the same factors as the stop loss and take profit factors. The pip spreads are actually used to calculate the sl and tp.
Revised and updated forex positioning excel sheet available for download.
You can experiment with an analysis program that I have put online:
http://forex.azrewards.com
* * * IMPORTANT NOTES * * *
This is not a make a million overnight. It only realizes its true advantage after recording many-many decisons.
The results posted by RicG are very-very high relative to long term expectation but certainly not un-realistic in a short term run.
Think about this: If you maintained a 5% advantage, to net an average of $100 per day you would require a minimum bankroll of $40,000.
( 100 / .05 / .05 )
Was an approximation of average losses. Have since refined with a stop loss and take profit. The stop loss is 1.65 standard deviations which represents 90% of probable senerio outcomes. The take profit is 1.96 standard deviations which represents 95% of probable senerio outcomes.
My statistical analysis shows that the results are 'virtually' identical. Eliminates the far extremes which annoy the crap out of me
"Also, can you explain how you decided on which number to use for "Pip Spreads"?"
The pip spreads use the same factors as the stop loss and take profit factors. The pip spreads are actually used to calculate the sl and tp.
Revised and updated forex positioning excel sheet available for download.
You can experiment with an analysis program that I have put online:
http://forex.azrewards.com
* * * IMPORTANT NOTES * * *
This is not a make a million overnight. It only realizes its true advantage after recording many-many decisons.
The results posted by RicG are very-very high relative to long term expectation but certainly not un-realistic in a short term run.
Think about this: If you maintained a 5% advantage, to net an average of $100 per day you would require a minimum bankroll of $40,000.
( 100 / .05 / .05 )
- Attachments
-
- Updated_forex_positioning.xls
- (240.5 KiB) Downloaded 252 times
- RicG
- rank: 150+ posts
- Posts: 274
- Joined: Tue Jul 24, 2007 11:44 pm
- Reputation: 8
- Location: Nashville, TN
- Gender:
Hi Pitboss,
Thanks for the additional information. I have a question about how you calculate your Stop Loss and Take Profit. Is your standard deviation based on Average True Range? If so, over what time period?
I'm getting ready to go "live" trading the system with a smaller amount of units than I have been paper trading.
I'll keep everybody posted on my results.
Thanks,
Ric
_____________________________________________________________
Have since refined with a stop loss and take profit. The stop loss is 1.65 standard deviations which represents 90% of probable senerio outcomes. The take profit is 1.96 standard deviations which represents 95% of probable senerio outcomes.
My statistical analysis shows that the results are 'virtually' identical. Eliminates the far extremes which annoy the crap out of me.
Thanks for the additional information. I have a question about how you calculate your Stop Loss and Take Profit. Is your standard deviation based on Average True Range? If so, over what time period?
I'm getting ready to go "live" trading the system with a smaller amount of units than I have been paper trading.
I'll keep everybody posted on my results.
Thanks,
Ric
_____________________________________________________________
Have since refined with a stop loss and take profit. The stop loss is 1.65 standard deviations which represents 90% of probable senerio outcomes. The take profit is 1.96 standard deviations which represents 95% of probable senerio outcomes.
My statistical analysis shows that the results are 'virtually' identical. Eliminates the far extremes which annoy the crap out of me.
Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.
Thank you for your support.
The time period used to calculate the standard deviation was 8 years of daily data, little over 2,000 data samples each pair. The range was each day's middle to it's actual high and low.
Again a word of caution. Make absolutely certain that your 'bet' doesn't exceed your bankroll * advantage. Most people 'gamble' too much and even with a winning system still get themselves broke or at to the point where they become afraid.
Again a word of caution. Make absolutely certain that your 'bet' doesn't exceed your bankroll * advantage. Most people 'gamble' too much and even with a winning system still get themselves broke or at to the point where they become afraid.
- RicG
- rank: 150+ posts
- Posts: 274
- Joined: Tue Jul 24, 2007 11:44 pm
- Reputation: 8
- Location: Nashville, TN
- Gender:
Here's an update; I've now been paper trading the system 10 weeks. Using 10 units of the four major pairs has resulted in a 74-50 record (59.7%), netting $8406 before commissions. Not using the filtering portion nor the SL-TP as outlined by pitboss, has yielded a 119-80 record (59.8%) or $18,850 before commissions.
Great trading to all,
Ric
Great trading to all,
Ric
(Disclaimer - This post is for educational purposes only. Always consult a licensed investment professional before taking any trade. Any trade you take is at your own risk.)
- RicG
- rank: 150+ posts
- Posts: 274
- Joined: Tue Jul 24, 2007 11:44 pm
- Reputation: 8
- Location: Nashville, TN
- Gender:
The last two weeks the system has really taken it on the chin - as the four major pairs have all been in extremely strong trends. Over that period the system went 5-18 (21.7%) losing $10,390 dollars based on 10 units each trade. That brings the 13 week total to a record of 81-72 (53.8%) with a profit of $125.
(Disclaimer - This post is for educational purposes only. Always consult a licensed investment professional before taking any trade. Any trade you take is at your own risk.)
-
- rank: 50+ posts
- Posts: 110
- Joined: Wed Mar 14, 2007 2:15 pm
- Reputation: 0
- Location: Lahore
- Gender:
MAY WE ADD A FILTER
Pitboss,
I acknowledge: the simpler the better.
But a filter may help avoid those aweful drawdowns.
May I suggest Kaufman's Adaptive Moving Average with Standard Deviation Bands?
Khalid
I acknowledge: the simpler the better.
But a filter may help avoid those aweful drawdowns.
May I suggest Kaufman's Adaptive Moving Average with Standard Deviation Bands?
Khalid
-
- rank: 50+ posts
- Posts: 110
- Joined: Wed Mar 14, 2007 2:15 pm
- Reputation: 0
- Location: Lahore
- Gender:
RicG's persistent effort encouraged me to write a quick EasyLanguage code do my own analysis.
Then I did a test on emini S&P500 2700 plus bars of combined session daily data, the EXCEL reoprt of which is attached. I have removed the many report sheets containing lot of un-necessary details to enable attachable file size.
Win percentage is fine at 50%; the problem is winning trades are smaller than losing trades.
Khalid
Then I did a test on emini S&P500 2700 plus bars of combined session daily data, the EXCEL reoprt of which is attached. I have removed the many report sheets containing lot of un-necessary details to enable attachable file size.
Win percentage is fine at 50%; the problem is winning trades are smaller than losing trades.
Khalid
- Attachments
-
- MIDDLE ANALYSIS_ @ES# Back-Testing Strategy Performance Report.xls
- (161 KiB) Downloaded 222 times
Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.
Thank you for your support.