Slingshot 30M & 1H - The 2nd Coming...... | Page 64 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Slingshot-30 Min. PA System 394 replies

ECN && STP && Scalping && Hedging - Broker Review - Oct 2009 24 replies

question about "slingshot" indicators 1 reply

Advice/Criticism on Indicator Settings for 30M and 2H 0 replies

SlingShot Method A 12 replies

Attachments: Slingshot 30M & 1H - The 2nd Coming......
Exit Attachments

Slingshot 30M & 1H - The 2nd Coming......

  • #1,261
  • Mar 12, 2009 7:06pm Mar 12, 2009 7:06pm
Damon: Thank you. I appreciate the information. Steve
  • #1,262
  • Mar 12, 2009 7:43pm Mar 12, 2009 7:43pm
Damon: I see the following filters in the Input menu: MMF, SF, ZF, MAF, and HF. Which of these correspond to MA and HMA. I assume MA is MAF.

Thank you.

Steve
  • #1,263
  • Mar 12, 2009 11:04pm Mar 12, 2009 11:04pm
Whoops! I think I got it. It helps to re-read the notes to figure out the alphabet abreviations. Steve
  • #1,264
  • Mar 13, 2009 8:01am Mar 13, 2009 8:01am
Your are correct Damon........all the ma trades were correct!

Gary

Quoting damonl
Actually Gary, it looks like the 4th trade should have been taken. The cross had occurred,but before the bar closed. The EA would not have known about the cross until the bar closed in which the EA made the trade. It looks like the MA portion of the EA is working correctly, and I am almost willing to leave the HMA part alone until further testing is complete.
  • #1,265
  • Mar 13, 2009 11:18am Mar 13, 2009 11:18am
Slingshot is now running on my system. I am using the MAF (Moving Average Filter (?). Thank you for your assistance.

I thought I would try the Metaquotes Strategy Tester using Eur/Usd 30 minute time frame. This is sort of a look-see exercise.

The Strategy Tester started on September 2, 2008 and ended March 13, 2009 at 12:30 for a total of 794 (profit + loss) trades. The profit was a grand: $25.72, a virtual wash. The percentage of profitable trades was respectable at 64%. Evidently the size of the loss trades overcame the advantage that one might expect with 64% profitable trades.

The graph of the results showed deep peaks and valleys. It was erratic. This would likely give a trader many sleepless nights in real time.

I don't know enough about this business to know how adequate the Strategy Tester is.

The results suggest that modifying the stop loss strategy might improve the results.

Comments welcome.

Steve
  • #1,266
  • Mar 13, 2009 12:58pm Mar 13, 2009 12:58pm
Hey Steve -

It is funny that you should bring this up. At the moment I am optimizing the EA for GbpUsd 30M chart for the default MA settings for both s/l nd t/p since Oct. 08.

The plan is to find optimal settings for both that give the best results. Once this first run is done, I plan and even focusing in more by redusing my steps to 1 where right now they are by 10. Optimizing definitely takes time.

The 2nd stage would be to optimize the MA settings once we get this done. Gary likes the results he is getting, and I think the MA's are too far apart and don't mimick price action close enough. Maybe they will be good where they are because the slope direction line does mimick price action closer and between these two, we may have what we need.

Anyway, just wanted to let you know that I am working on it, and I believe that Gary will also do some optimizing this weekend as well.

Later....Damon
"Keep your eyes on the helpers" - Mr. Rogers
  • #1,267
  • Mar 13, 2009 1:10pm Mar 13, 2009 1:10pm
Here's the bad news chart on the Gbp/Usd,but not is all lost. I am using Alpari-us demo so my results may be a little different than others. The first chart is the M30 of the GBP/USD with the 4th filter MAF being used.
In the first part of this week when i was looking into why and when a failed trade took place,I was looking at shorter time frames, like M15, M5, and M1 I noiced that the M15 seem to have a lot more sling shot bars with good distance between open and close. So I played around with it this week and had very good results. Sorry I didn't save the charts but here is todays M15 Chart. I know this is not part of the test program but the results have been to good to egnore. Will keeping posting the M15 chart along with the M30 charts if it is ok with Damonl and Gary.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: m30 maf_3_13_09.gif Size: 15 KB
Click to Enlarge Name: m15_maf 3_13.gif Size: 19 KB
  • #1,268
  • Mar 13, 2009 1:47pm Mar 13, 2009 1:47pm
I thought we were done at GMT 17 but it does trade until GMT18 on Firdays. So we had one more trade at GMT 1750 on the M15 chart and one more trade at GMT 1757 on the M30 chart. The M30 lost one more pip while the M15 gained another 9 pips.
Also Gary if you look at the upper left hand corner of my chart you see the hma is still posted instead of the ma filter and the gmt off set is not working for me either. No big deal at this point more important fish to fry first. Thanks for all the great work everybody is doing with this system......Richard
Attached Image(s) (click to enlarge)
Click to Enlarge Name: m30 maf_3_13_v2.gif Size: 15 KB
Click to Enlarge Name: m15_maf 3_13v2.gif Size: 23 KB
  • #1,269
  • Mar 13, 2009 1:55pm Mar 13, 2009 1:55pm
Hi Skybiz007,


That second chart of yours looks good!


Good work Gary!
  • #1,270
  • Mar 13, 2009 2:21pm Mar 13, 2009 2:21pm
Great job... Looks to me like M15 for GbpUsd for MA deserves some optimization as well. Thanks for the post

Quoting skybiz007
I thought we were done at GMT 17 but it does trade until GMT18 on Firdays. So we had one more trade at GMT 1750 on the M15 chart and one more trade at GMT 1757 on the M30 chart. The M30 lost one more pip while the M15 gained another 9 pips.
Also Gary if you look at the upper left hand corner of my chart you see the hma is still posted instead of the ma filter and the gmt off set is not working for me either. No big deal at this point more important fish to fry first. Thanks for all the great work everybody is doing with this system......Richard...
"Keep your eyes on the helpers" - Mr. Rogers
  • #1,271
  • Mar 13, 2009 2:43pm Mar 13, 2009 2:43pm
Damon: Glad to hear from you. Good work. Steve
  • #1,272
  • Mar 13, 2009 2:52pm Mar 13, 2009 2:52pm
Richard: The 15-minute charts you posted look promising. Thank you. It looks like you are taking profits 1,2 or 3 bars out from your entry bar. What are your thoughts about this?

Steve
  • #1,273
  • Mar 13, 2009 4:47pm Mar 13, 2009 4:47pm
Taking my inspiration from Richard, I ran the MQ Strategy Tester on the EUR/USD 15 minute chart. There were 708 trades between 12/1/2008 and 3/13/09. The result was a profit of $60, consistent, albeit it somewhat higher, with the diminished result of testing the 30 minute chart. Interesting, however, is that the profitable trades edge up to 66%. 66% is a respectable result. This again suggests that work with setting stop losses might improve performance.

Steve
  • #1,274
  • Mar 13, 2009 7:44pm Mar 13, 2009 7:44pm
Hi skybiz -

Great feedback! When a time is 17 like for this Friday close, it means the window closes at 17:59:59.......hope that helps a bit. As far as the comment in the upper left, I haven't change it yet.....the slope indie we are now using is really just a non-repaint version of hma any way......but I'll change it on the next version.

Gary

Quoting skybiz007
I thought we were done at GMT 17 but it does trade until GMT18 on Firdays. So we had one more trade at GMT 1750 on the M15 chart and one more trade at GMT 1757 on the M30 chart. The M30 lost one more pip while the M15 gained another 9 pips.
Also Gary if you look at the upper left hand corner of my chart you see the hma is still posted instead of the ma filter and the gmt off set is not working for me either. No big deal at this point more important fish to fry first. Thanks for all the great work everybody is doing with this system......Richard...
  • #1,275
  • Mar 13, 2009 7:49pm Mar 13, 2009 7:49pm
Thanks for the results Steve -

The defaults were picked without any evidence that they were correct for the ma or slope(hma) filters......it was a starting point to test the technical aspects of the ea........now that the ea is performing as planned, it is time to optimize the setting. These two filters show the most promise out the the ones tested. What we are looking at is default settings that have winning percentages, not necessarily resultiing in large positive pips......not we can look for better filter settings that will improve pips gained.....the winning percentage was a gage to get us started in the right direction. Damon and I are doing lots testing this weekend as well trying to find the best settings.

Thanks again - Gary

Quoting Steveg100
Taking my inspiration from Richard, I ran the MQ Strategy Tester on the EUR/USD 15 minute chart. There were 708 trades between 12/1/2008 and 3/13/09. The result was a profit of $60, consistent, albeit it somewhat higher, with the diminished result of testing the 30 minute chart. Interesting, however, is that the profitable trades edge up to 66%. 66% is a respectable result. This again suggests that work with setting stop losses might improve performance.

Steve
  • #1,276
  • Mar 13, 2009 7:50pm Mar 13, 2009 7:50pm
Quoting Steveg100
Taking my inspiration from Richard, I ran the MQ Strategy Tester on the EUR/USD 15 minute chart. There were 708 trades between 12/1/2008 and 3/13/09. The result was a profit of $60, consistent, albeit it somewhat higher, with the diminished result of testing the 30 minute chart. Interesting, however, is that the profitable trades edge up to 66%. 66% is a respectable result. This again suggests that work with setting stop losses might improve performance.

Steve
Hi guys, keep up the good work.

Just wanted to make a "heads up" concerning the use of the Strategy tester, since you don't seem to be aware of that (sorry if I'm wrong):

First of all, it isn't very reliable;
But for it to be more reliable, you have to take some actions, like downloading M1 data from your broker for the pair you want, uploading it to the M1 database, and then convert it to the various timeframes (M5, M15, M30...) in order to obtain a +90% of test quality (visible in the upper bar of the final report). There are tutorials available for the procedures to take;

Finally, I advise you to choose 0.1 lot size when testing USD pairs (when account is in USD), so that the money earned equals the amount of pips earned - so that you can count profit in pips, rather then in money (that's what matters knowing).
  • #1,277
  • Mar 13, 2009 7:51pm Mar 13, 2009 7:51pm
Hi sng -

We're getting there.......looking forward to the testing this weekend so we can try some new ma pairs next week.

Gary

Quoting silverngold
Hi Skybiz007,


That second chart of yours looks good!


Good work Gary!
  • #1,278
  • Mar 13, 2009 8:02pm Mar 13, 2009 8:02pm
Hi All

I've attached charts from todays trading with the ma and slope(hma) filters. Overall a losing day, but not unexpected in the whipsaw market of today. Ma was 4/7 (-48) and slope 1/3(-28). Ready to work on testing now to optimize the setting to improve the results. However, even with the defaults the ma filter for the past week or so has been very good. Not enough results to draw conclusions on the slope(hma). Looking forward to next weeks trading. Have a wonderful weekend!

Gary
Attached Image(s) (click to enlarge)
Click to Enlarge Name: gu_ma.gif Size: 12 KB
Click to Enlarge Name: gu_hma.gif Size: 13 KB
  • #1,279
  • Mar 13, 2009 8:05pm Mar 13, 2009 8:05pm
Thanks tonirai -

I've used the tester a lot and don't care for it much......all your suggestions are good, I use it along similar lines but get the 90%.......i use it on weekends to get into the ballpark but really almost entirely on forward testing as you can see in this thread. Appreciate the feedback

Gary

Quoting tonirai
Hi guys, keep up the good work.

Just wanted to make a "heads up" concerning the use of the Strategy tester, since you don't seem to be aware of that (sorry if I'm wrong):

First of all, it isn't very reliable;
But for it to be more reliable, you have to take some actions, like downloading M1 data from your broker for the pair you want, uploading it to the M1 database, and then convert it to the various timeframes (M5, M15, M30...) in order to obtain a +90% of test quality (visible in the upper bar of the final report). There are tutorials available...
  • #1,280
  • Mar 13, 2009 8:15pm Mar 13, 2009 8:15pm
The only time the strategy tester is worth a dang is if you are doing something simple. When you add indicators, etc, then forget it. I used a system that was based purely on an ma live for a year, then decided maybe I would code it. I did, compared the totals to mine and it was exact, but it was based on 4hr tf and had over 200 pip stops, etc, so it did not need a lot of little tf data. The tester is only good to get an idea if your ea is working right and if it is resulting at least a positive number. Forward testing tells all.
0 traders viewing now