The Fozzy EA | Page 6 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Mouteki Vs Fozzy 8 replies

The Daily Fozzy Method 2,814 replies

Fozzy Alt... one more approach to the Daily Fozzy Method 40 replies

My Daily Fozzy Trades 337 replies

Fozzy, Pip Boxer, Flyer 4 replies

Attachments: The Fozzy EA
Exit Attachments

The Fozzy EA

  • #101
  • Oct 23, 2006 12:20pm Oct 23, 2006 12:20pm
Quoting tradeigel
stockwet, I see you fixed the problem. That's good. Congrats on this .
Thanks. It was a big pain, though.
  • #102
  • Oct 23, 2006 12:24pm Oct 23, 2006 12:24pm
I have some interesting results on these EA's. First, the EA that did not have the calculation fix in it performed the best over anything so far. That's a little scary since I don't know how the RSI MA's were being calculated. Once that was fixed, Fozzy, whether normal or using BB's mods, performed poorly. In fact, the account was blown within the first few months.

So, I went to work and started incorporating some of miscon's aspects into the EA. Namely, I worked on setting entries above and below the respective previous high's and low's. That method performed relatively well. At least it left me in the positive.

I'll try to run all these through the gauntlet and post some reports. I'll also do some code cleanup and a bit of versioning so that we know what is what.
  • #103
  • Oct 23, 2006 1:41pm Oct 23, 2006 1:41pm
Interesting developments today with the EA. Thanks for your efforts, stockwet.

Quoting stockwet
I have some interesting results on these EA's. First, the EA that did not have the calculation fix in it performed the best over anything so far. That's a little scary since I don't know how the RSI MA's were being calculated. Once that was fixed, Fozzy, whether normal or using BB's mods, performed poorly. In fact, the account was blown within the first few months.


So, I went to work and started incorporating some of miscon's aspects into the EA. Namely, I worked on setting entries above and below the respective previous high's and low's. That method performed relatively well. At least it left me in the positive.

I'll try to run all these through the gauntlet and post some reports. I'll also do some code cleanup and a bit of versioning so that we know what is what.
  • #104
  • Oct 23, 2006 2:21pm Oct 23, 2006 2:21pm
Quoting stockwet
The first post is now edited with the latest fozzy EA. The calculation problem has been solved.
Backtested new (version 10-23-06) on GPBUSD 10k wentg below 2k

on version just prior to this one, 10-17-06 version, 10k went to over 23k

Something is wrong here, I believe
  • #105
  • Oct 23, 2006 11:19pm Oct 23, 2006 11:19pm
I have a question on today's entries. No trying to rock the boat at all, so please don't take my doodling intending any ill will, just doing some tests to see if we can all profit from them. I have set the MA method to zero as this would be an SMA signal line instead of the default EMA Fozzy. Now i have two identical MT4 demos though the second one has the BBs set at 40 instead of the default 20. Both entered into EurUsd Sells at 1.2545, and on the GBP the BB40 in a Sell at 1.8722 and the BB20 in a Sell at 1.8723 . I would like to compare notes with anyone that may have had their EA set at the default Ema (Per Default Fozzy) with the above two trades if taken.

Also, a curiosity, The BB20 tiggered a Buy on the UsdJpy at 119.34 today, but no similar entry on the BB40 test.

Thanks in advance. The only reason i don't know is that i've got 10 MT4 platforms running concurrently on my secondary PC and i think that's about maxed out for now.

Cheers,
Thom
  • #106
  • Oct 24, 2006 1:42am Oct 24, 2006 1:42am
Take a look at post 102. I've visually confirmed that the new version is, in fact, taking correct fozzy signals. That's the bad news. The good news is that the fozzy alt method, after preliminary tests, looks highly effective at weeding out negative trades.

One of the big problems with fozzy, as it appears, is that it doesn't do well in trending markets. It will throw signals counter to the trend after a retracement has taken place, but then reverse and resume the trend. Setting orders above and below the previous high/low seems to be an effective way of avoiding this most of the time. I have a fozzy alt expert I'm working on now.


Quoting wwwin
Backtested new (version 10-23-06) on GPBUSD 10k wentg below 2k

on version just prior to this one, 10-17-06 version, 10k went to over 23k

Something is wrong here, I believe
  • #107
  • Oct 24, 2006 1:44am Oct 24, 2006 1:44am
What time did the expert generate the signal? Which version of the expert are you using?

Quoting accrete
I have a question on today's entries. No trying to rock the boat at all, so please don't take my doodling intending any ill will, just doing some tests to see if we can all profit from them. I have set the MA method to zero as this would be an SMA signal line instead of the default EMA Fozzy. Now i have two identical MT4 demos though the second one has the BBs set at 40 instead of the default 20. Both entered into EurUsd Sells at 1.2545, and on the GBP the BB40 in a Sell at 1.8722 and the BB20 in a Sell at 1.8723 . I would like to compare notes with anyone that may have had their EA set at the default Ema (Per Default Fozzy) with the above two trades if taken.

Also, a curiosity, The BB20 tiggered a Buy on the UsdJpy at 119.34 today, but no similar entry on the BB40 test.

Thanks in advance. The only reason i don't know is that i've got 10 MT4 platforms running concurrently on my secondary PC and i think that's about maxed out for now.

Cheers,
Thom
  • #108
  • Edited 7:23am Oct 24, 2006 2:30am | Edited 7:23am
Quoting stockwet
One of the big problems with fozzy, as it appears, is that it doesn't do well in trending markets. It will throw signals counter to the trend after a retracement has taken place, but then reverse and resume the trend. Setting orders above and below the previous high/low seems to be an effective way of avoiding this most of the time. I have a fozzy alt expert I'm working on now.
Hi Stockwet , i don't understand very well your point here could you give us an example on chart ?
From what i understand Fozzy method is a trend follower system so it's supposed to performing well when a strong trend is in place.
Looking at chart for some manual backtest most of time signals seems to lead to profit trades and really few loss occurs even if atr is used as stoploss.
On other hand I notice how backtesting with MT4 Strategy Tester is completely unrealiable, infact most of time it opens orders one day later that it should be losing most of profit or in worst case turning in a lose trade when it shouldn't be at all.

Image removed, please post no larger than 600x600
  • #109
  • Oct 24, 2006 5:10am Oct 24, 2006 5:10am
All,

I realized that I did a poor job of separating out versions on this last update. I've updated the front page which includes version 2.1 (without the calculation fix) and version 2.2 (with the calculation fix.)

Version 2.1 is the one that performed quite well, while version 2.2 is the most accurate with the system. If you grabbed version 2.1 yesterday, I apologize for the confusion. That version should actually be 2.2.

stockwet
  • #110
  • Oct 24, 2006 5:46am Oct 24, 2006 5:46am
Skyline,

It's pretty evident from looking at the backtesting charts that the method has difficulty in strongly trending market. Take October to December 2004 as an example. The EURUSD experienced a strong bull trend. This caused 2 fundamental problems.

1. The EA would throw signals on minor retracements, causing trades to get stopped out.
2. The EA would not throw bull signals because the bull crosses would often happen above the BB median line.

During that period, there were approximately 11 signals. Of those, there were 4 negative trades. Not so bad, except that each negative trade was about 3 times the average win of the remaining 7. Taking the BB filter off seemed to do even worst.

If you'd like, you can see for yourself by just looking at how the EA performs in strongly trending markets. Others have mentioned the same thing in other posts, I believe.

Regards,

stockwet

Quoting skyline
Hi Stockwet , i don't understand very well your point here could you give us an example on chart ?
From what i understand Fozzy method is a trend follower system so it's supposed to performing well when a strong trend is in place.
Looking at chart for some manual backtest most of time signals seems to lead to profit trades and really few loss occurs even if atr is used as stoploss.
On other hand I notice how backtesting with MT4 Strategy Tester is completely unrealiable, infact most of time it opens orders one day later that it should be losing most of profit or in worst case turning in a lose trade when it shouldn't be at all.
  • #111
  • Edited 8:03am Oct 24, 2006 6:33am | Edited 8:03am
Quoting stockwet
Skyline,

It's pretty evident from looking at the backtesting charts that the method has difficulty in strongly trending market. Take October to December 2004 as an example. The EURUSD experienced a strong bull trend. This caused 2 fundamental problems.

1. The EA would throw signals on minor retracements, causing trades to get stopped out.
2. The EA would not throw bull signals because the bull crosses would often happen above the BB median line.

During that period, there were approximately 11 signals. Of those, there were 4 negative trades. Not so bad, except that each negative trade was about 3 times the average win of the remaining 7. Taking the BB filter off seemed to do even worst.

If you'd like, you can see for yourself by just looking at how the EA performs in strongly trending markets. Others have mentioned the same thing in other posts, I believe.

Regards,

stockwet
Stock, take a look at this (it's a backtest on EURUSD with latest fozzy starting on oct '04) and i'm still convinced that backtest with strategy tester isn't realiable :

Image removed: Please repost no larger than 600x600

I marked 4 days : 4 Oct, 8 Oct, 12 Oct (i'm not sure about this), 14 Oct
Isn't supposed that EA fozzy had to open a trade on these days? I think so based on what i understand on fozzy method but Strategy tester doesn't and all trades were winning if opened. Infact it opened first order later on 9 Nov that leads to first loss. From this evidence i cannot say that EA performed so bad in strongly trend market because backtest didn't take care about those winner trades (or open a trade a day later eating most of profit). Isn't it ?
  • #112
  • Oct 24, 2006 6:59am Oct 24, 2006 6:59am
Yeah, I have the same problem. I found this post on MetaQuotes website: http://www.metaquotes.net/forum/2458/. I initially thought that the problem might be related to there not being enough data to calculate the 20 period BB lines. But, I'm not sure.

Quoting skyline
Stock, take a look at this (it's a backtest on EURUSD with latest fozzy starting on oct '04) and i'm still convinced that backtest with strategy tester isn't realiable :



I marked 4 days : 4 Oct, 8 Oct, 12 Oct (i'm not sure about this), 14 Oct
Isn't supposed that EA fozzy had to open a trade on these days? I think so based on what i understand on fozzy method but Strategy tester doesn't and all trades were winning if opened. Infact it opened first order later on 9 Nov that leads to first loss. From this evidence i cannot say that EA performed so bad in strongly trend market because backtest didn't take care about those winner trades (or open a trade a day later eating most of profit). Isn't it ?
  • #113
  • Oct 24, 2006 7:24am Oct 24, 2006 7:24am
Sorry for a dumb question: I thought that the Fozzy Daily Method indicators was automatically installed when I open the EAFozzy2-2, but there is not one indicator there.

Where can I find the complete indicators for this Fozzy Daily Method(EA)?
  • #114
  • Oct 24, 2006 7:29am Oct 24, 2006 7:29am
Hi Stockwet, the version i'm testing in your original version 2.1 that i downloaded last Thursday or Friday and these are the first trades of the test. Each trade says they were triggered at 00:00 on the 24th.

There are two other areas that the setups differ that i just realized i had not included in my notes in post #105. The BB20 version has BB filter=false, and GMT=false. The BB40 version has both at true. My main point in my post 105 was if anyone had similar trades triggered, and at what price the entries were, using the exponential MA signal line instead of the simple MA as i've got it set to.

: ) Thom
  • #115
  • Oct 24, 2006 7:30am Oct 24, 2006 7:30am
Quoting LFN
Sorry for a dumb question: I thought that the Fozzy Daily Method indicators was automatically installed when I open the EAFozzy2-2, but there is not one indicator there.

Where can I find the complete indicators for this Fozzy Daily Method(EA)?
Constructing EA's versus indicators is a bit of a different beast. You can get information on how to construct the indicators from the original fozzy thread. It's not really a custom indicator either, just a matter of placing indicators on indicators.
  • #116
  • Oct 24, 2006 7:34am Oct 24, 2006 7:34am
OK. Thanks, Thom.

I've made a bit of a mess with versioning. Yesterday, I posted a version 2.1 that should have been 2.2. I changed all that this morning. The version you have should be a nearly correct 2.1.

As for the trades, I didn't see the same things come through. I'm not good to guage against though since mostly I'm working on fixing bugs, testing, or adding new features. I'll look on my other machine though and see what was triggered on those days.

Quoting accrete
Hi Stockwet, the version i'm testing in your original version 2.1 that i downloaded last Thursday or Friday and these are the first trades of the test.

There are two other areas that the setups differ that i just realized i had not included in my notes in post #105. The BB20 version has BB filter=false, and GMT=false. The BB40 version has both at true. My main point in my post 105 was if anyone had similar trades triggered using the exponential MA signal line instead of the simple MA as i've got it set to.

: ) Thom
  • #117
  • Oct 24, 2006 7:35am Oct 24, 2006 7:35am
Quoting LFN
Sorry for a dumb question: I thought that the Fozzy Daily Method indicators was automatically installed when I open the EAFozzy2-2, but there is not one indicator there.

Where can I find the complete indicators for this Fozzy Daily Method(EA)?
Hi LFN,
attached you can find template that i'm using in demo Fozzy EA
Attached File(s)
File Type: zip fozzy_template.zip   1 KB | 813 downloads
  • #118
  • Oct 24, 2006 7:40am Oct 24, 2006 7:40am
I can see a lot of indicators posted here on the pictures in this tread, RSI/MA and ATR.
You have to put in these indicators manually even if you have the EA..?..correct?

This is from Fozzys first message in the other tread -
"On each chart I have an 8 period RSI. I also have an 8 period MA of the RSI and Bollinger Bands with a 20 period setting, also on the RSI"

I cant see anything about the ATR indicator in fozzys indicators.

Quoting stockwet
Constructing EA's versus indicators is a bit of a different beast. You can get information on how to construct the indicators from the original fozzy thread. It's not really a custom indicator either, just a matter of placing indicators on indicators.
  • #119
  • Oct 24, 2006 7:59am Oct 24, 2006 7:59am
LFN,

Welcome to Forex Factory. This is a great place to be with a lot of smart and respectful people. I think you'll enjoy it.

Please don't take my comments negatively - it's just constructive criticism. It would be polite if before you posted questions about a particular method, that you take the time to get yourself up to speed. Otherwise, you're asking those of us ahead of you to do extra work to come back and give you the answers.

Your question was already asked in the original fozzy thread. Please read the entire thread first. I could go and search out the posting for you with your answer, but, that would rob you of the chance to learn from all the other posts there. Once you're up to speed with the method and have read through this thread, then fire away on your questions.

Regards,

stockwet

Quoting LFN
I can see a lot of indicators posted here on the pictures in this tread, RSI/MA and ATR.
You have to put in these indicators manually even if you have the EA..?..correct?

This is from Fozzys first message in the other tread -
"On each chart I have an 8 period RSI. I also have an 8 period MA of the RSI and Bollinger Bands with a 20 period setting, also on the RSI"

I cant see anything about the ATR indicator in fozzys indicators.
  • #120
  • Oct 24, 2006 8:12am Oct 24, 2006 8:12am
I´m really glad you answer me. I´ve spent the last two days reading throught both threads but it takes some time when English isn´t my home language. But I will start over from page 1. Thanks



Quoting stockwet
LFN,

Welcome to Forex Factory. This is a great place to be with a lot of smart and respectful people. I think you'll enjoy it.

Please don't take my comments negatively - it's just constructive criticism. It would be polite if before you posted questions about a particular method, that you take the time to get yourself up to speed. Otherwise, you're asking those of us ahead of you to do extra work to come back and give you the answers.

Your question was already asked in the original fozzy thread. Please read the entire thread first. I could go and search out the posting for you with your answer, but, that would rob you of the chance to learn from all the other posts there. Once you're up to speed with the method and have read through this thread, then fire away on your questions.

Regards,

stockwet
1 trader viewing now