Lost in the Tunnel... | Page 15 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

One hour tunnel method 2 replies

Tunnel Vision. Independent Signals & Positioning Recommendations. 189 replies

4 hour tunnel confusion 8 replies

I am confused about all the different Tunnel methods out there...Clarificatio n please 3 replies

Questions re: MT4 and VWB 1 Hr Tunnel 0 replies

Attachments: Lost in the Tunnel...
Exit Attachments

Lost in the Tunnel...

  • #281
  • Dec 30, 2005 11:01pm Dec 30, 2005 11:01pm
Quoting Antra
Yes James......have you checked your email?????
And Trey's thread????
i would have expected the mouse family to have grown by now.
  • #282
  • Dec 30, 2005 11:19pm Dec 30, 2005 11:19pm
Quoting james16
i would have expected the mouse family to have grown by now.
Two is my limit!!!!! :
  • #283
  • Dec 31, 2005 12:26am Dec 31, 2005 12:26am
Quoting firehorse
Hi,I'm floundering a bit here.

The original 4 hr tunnel should be profitable but I think I don't understand the rules properly because I cannot as yet duplicate the original results.

The new 20pip rule by Diallist seems even better but I can't seem to duplicate his results from the above post as yet. There's a misunderstanding somewhere.

The updated rules by vegas seems to be in development and there is nothing to compare against at the moment. I would prefer if I could understand the discrepancies above before moving onto another system along this theme.

Best regards
Alan
Wow Alan! Why is there so much difference in our charts? Could someone else compare their chart to mine and Alan's and let us know which one is right, if either? Thanks.

Dial
Attached Image
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #284
  • Dec 31, 2005 12:29am Dec 31, 2005 12:29am
Quoting keris2112
Hi all,

I've been a member here for a while, but I don't think I've ever posted. I post a lot at other forums and just never quite got around to FF very often. Wow, was that a mistake. You guys are great.

Hi Keris, and welcome!

Great post Keris! Antra shared the HMA with me last week, and I see she has posted here today.

I appreciate your contribution to this thread and hope to hear more good things from you.

Dial
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #285
  • Dec 31, 2005 12:35am Dec 31, 2005 12:35am
Quoting Antra
Hi Keris,

I too have been playing around with the vegas tunnel and trying to work out how to trade it in a more mechanical fashion.

I came across the HMA a few weeks ago and found it to be a pretty good little indicator. Seems to have very little lag.

The one I have been using is 'HMA in colour'.
It changes colour as the trend turns...but here is the problem.
I am not 100% sure that once the bar has finished that the colour will remain changed.

Anyone willing to test it?

I first thought of putting this indicator on the weekly chart in place of the 28ema that Vegas mentioned as it seemed ( to me) to be giving more direction.

Anyway this is the HMA in colour on a setting of 21
Hi Princess! So good to see you here! Did you get my reply to your PM?

Everyone, please welcome Factory Princess Antra!

After ya'll are through bowing and scraping, then pay heed, for Antra is a true tunnel guru!

Great to see you my fellow pink mousekateer! (JIM! Where's your avatar so I can slap a pink mouse on it? )

Dial
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #286
  • Dec 31, 2005 3:26am Dec 31, 2005 3:26am
Hi Diallist
Quoting diallist
Wow Alan! Why is there so much difference in our charts? Could someone else compare their chart to mine and Alan's and let us know which one is right, if either? Thanks.

Dial
Thanks for taking the trouble of posting a chart.

Guess what! We're both right

I shifted my 'start of day' to 2:00 rather than 0:00 and now my 4 hr bars have the same shape as yours.

Interesting point to note
1) My close on the bar 8 is 1.2567 so would have entered at 1.2547.
2) my low on your exit bar is 1.2410 not touching the fib144 at 1.2408 so the trade would have exited later.

Attaching picture but I haven't changed the wording.

I'll go back and recheck the other trades.

Thanks again for your clarification.

Best regards
Alan
Attached Image
  • #287
  • Jan 1, 2006 7:18am Jan 1, 2006 7:18am
Hi

First, many thanks again for Diallist for taking the time to do a quick backtest for me so that I have some sort of reference point.

I've updated the eurusd spreadsheet with my results.

By my calculations, there appears to be a few non-mechanical trades.

23/01/04 Diallist has a short trade where I calculate momentum appears to be up.
02/02/04 Program picks up an extra mechanical trade
09/02/04 Program picks up an extra mechanical trade
11/02/04 Diallist has entry on 2nd bar after change in MA 8 slope whereas mechanically, the program only enters on the first bar after change
20/02/04 Secondary trade, Diallist had full points
25/02/04 Secondary trade, Diallist had full points
01/03/04 Secondary trade, Diallist had full points. Also had an fib377 exit for a fib233 exit
09/03/04 Diallist had a trade but 4hr momentum by my calculation appears to be below 50

Taking all of Diallist trades, my total of the same trades is 1497.5
Taking strict trades only, my total is 1351.5

The good news is that for the period 1/23/04 to 21/5/04 I now get a strict total of 2307.5, better than 2075 in the original vegas document.

However running the program for a longer period:
Strict trades for 2004 is 4981.5 in 110 trades
Strict trades for 2004 to 1/7/05 is 4838 in 148 trades
Still a little short on the 7500 in the original document.

Divide 4838 by 3 to get the number of points per 'risk on trade' => 1612.6
Minus 3pt spread per trade (444) => 1168.6
Over 18 months => 65 points per month

Max drawdown is in the 1500pt range, or 500pt if divided by 3.

Note my mechanical method of stoploss is the lowest low, or highest high of the last 3 bars.

Backtesting program is still subject to error and tweaking although it appears to be in the ballpark figure now and catching the right sort of entries and exits.

Note that the results are without any other considerations of support/resistance, fib retracements, price action etc.

Also remember the result is only for one currency.

I haven't yet put the results through a spreadsheet to calculate p/l on cash.

Thanks again to Diallist for putting in the time to give me something to compare against

Best regards
Alan
Attached File(s)
File Type: zip EURUSD_Sell_Backtest(2).zip   6 KB | 482 downloads
  • #288
  • Jan 1, 2006 4:40pm Jan 1, 2006 4:40pm
I was also searching for a 'smoother' alternative to the SMA for use in the tunnel method. I've been fiddling with the HMA for a little while now and it seems that it would provide fewer unprofitable signals. It does tend to be much smoother than SMA.

One thing I would LIKE to try (but don't have the software to do as I don't believe there is a release for any of the free charting software I use) is to try the Jurik MA. JMA may be even better than HMA - has anyone got the facilities to check this?

I haven't traded using the HMA yet but I plan to start in another few months. I look forward to seeing the material that will be presented in the PF soon

Cheers, Linc.
  • #289
  • Edited 10:25pm Jan 1, 2006 7:37pm | Edited 10:25pm
Quoting lwoo034
One thing I would LIKE to try (but don't have the software to do as I don't believe there is a release for any of the free charting software I use) is to try the Jurik MA. JMA may be even better than HMA - has anyone got the facilities to check this?
I have a copy of JMA for MT4, but it's not an official version. Officially they don't make one for mt4. I believe they sell a dll that's supposted to work with MT. The mq4 file that I have is supposed to be based off of the original source and from what I've heard is pretty much spot on compared to the official software.

I'd post it, but I don't know what the forum policy is in regards to this since it's not an official version. Of course, if it was official, i.e. licensed, then I couldn't legally share that either.

If Merlin ok's it, I'll post the .mq4 file for MT4.

BTW, both JMA and HMA are good. HMA is a little faster and JMA is can be a little smoother.

Here's some pics showing the differences.
8 sma
25 hma
14 jma
25 jma

They also have the 55 sma and 233 hma:

http://img329.imageshack.us/img329/6...ajma1su.th.gif

http://img329.imageshack.us/img329/8...ices2rw.th.gif

http://img359.imageshack.us/img359/2...term6pw.th.gif
  • #290
  • Jan 1, 2006 8:12pm Jan 1, 2006 8:12pm
Hi Guys,
Bit of a misconception that I am a tunnel Guru....far from it and what you have all done so far on this thread is really great.
I like the Vegas tunnels a lot and had a lot of success trading them demo but I then ran into difficulties with discretion.
It is like a project in progress for me to try and formulate a tunnel system that is purely mechanical.
If anyone wants the JMA just send me a pm.
I will add what I can to this thread but it really seems you have done a pretty good job.
Back to the drawing board.....
Happy New Year Everyone!!!!!!!
  • #291
  • Jan 1, 2006 9:14pm Jan 1, 2006 9:14pm
Quoting keris2112

BTW, both JMA and HMA are good. HMA is a little faster and JMA is can be a little smoother.

Here's a pic showing the differences.
8 sma
25 hma
14 jma
25 jma

http://img353.imageshack.us/img353/9...ajma6mj.th.gif
Since 8/55 crosses are such an important part of the system I was wondering if you could add the 55 SMA to the chart so that it can be seen where they cross and how much of a difference there is between each line.
  • #292
  • Edited Jan 2, 2006 12:09am Jan 1, 2006 10:20pm | Edited Jan 2, 2006 12:09am
Quoting volan
Since 8/55 crosses are such an important part of the system I was wondering if you could add the 55 SMA to the chart so that it can be seen where they cross and how much of a difference there is between each line.
I added the 55 sma to my above pic as well as the 233 hma, which I tend to think keeps us small during ranging markets. I don't know that replacing the 55 sma with the 233 hma will better, it's just something I've been playing with.

I also added 2 pics with the prices removed, showing a larger time period.
  • #293
  • Jan 2, 2006 12:01am Jan 2, 2006 12:01am
Gentlemen with all do respect I believe were starting to stray away from the version that Vegas either posted or updated us with in the email.

Then again I may be wrong because the above posts are confusing with new techniques implied and backtested.
  • #294
  • Edited 12:23am Jan 2, 2006 12:15am | Edited 12:23am
Quoting firehorse
23/01/04 Diallist has a short trade where I calculate momentum appears to be up.
02/02/04 Program picks up an extra mechanical trade
09/02/04 Program picks up an extra mechanical trade
11/02/04 Diallist has entry on 2nd bar after change in MA 8 slope whereas mechanically, the program only enters on the first bar after change
20/02/04 Secondary trade, Diallist had full points
25/02/04 Secondary trade, Diallist had full points
01/03/04 Secondary trade, Diallist had full points. Also had an fib377 exit for a fib233 exit
09/03/04 Diallist had a trade but 4hr momentum by my calculation appears to be below 50
Great job Alan! Thanks for taking the time to be so thorough in your backtesting.

Now as to the errors. They're mine, and I take full credit for them!
Quoting firehorse
Divide 4838 by 3 to get the number of points per 'risk on trade' => 1612.6
You lost me here. Why are you dividing by three?

Dial
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #295
  • Jan 2, 2006 12:22am Jan 2, 2006 12:22am
Quoting volan
Since 8/55 crosses are such an important part of the system I was wondering if you could add the 55 SMA to the chart so that it can be seen where they cross and how much of a difference there is between each line.
Hi volan,

Um, I'm curious as to why you think 8/55 crosses are an important part of the system. I've seen nothing in the original document that indicates crosses are used at all, much less being important. I think that is main beauty of the system is the lack of ma crosses.

Perhaps I missed something in an earlier post of yours that is keeping me from understanding.

Thanks volan for your continuing contributions!

Dial
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #296
  • Jan 2, 2006 12:37am Jan 2, 2006 12:37am
Quoting diallist
Um, I'm curious as to why you think 8/55 crosses are an important part of the system. I've seen nothing in the original document that indicates crosses are used at all, much less being important. I think that is main beauty of the system is the lack of ma crosses.
I think Volan is referring to the position sizing aspect of the 8/55 cross. When the 8 crosses the 55 and turns, take a full position. When the 8 turns without having crossed the 55, take 1/2 position.
  • #297
  • Jan 2, 2006 12:59am Jan 2, 2006 12:59am
Quoting diallist
You lost me here. Why are you dividing by three?

Dial
I was wondering the same thing, although the Vegas doc does suggest choosing a unit size and then trading 3 units (or 1 1/2). I think this is the wrong way of describing it, as it implies that you determine your position size and them multiply it by 3, thus, on each trade, risking 3 times what your money management method tells you to do. That's probably not what they meant, but that's kind of how it sounds.

I think we should do just the opposite. Determine our position size, based on whatever money management method we use, and then divide by 3 to determine how much to take off at each fib level. With trading platforms offering micro lots now (or any-size lots at oanda), this is easy to do. We don't have to trade in 3-lot (or mini lot) multiples.
  • #298
  • Jan 2, 2006 12:59am Jan 2, 2006 12:59am
Quoting keris2112
I think Volan is referring to the position sizing aspect of the 8/55 cross. When the 8 crosses the 55 and turns, take a full position. When the 8 turns without having crossed the 55, take 1/2 position.
Thanks keris! I suspected that might be what volan meant.

Dial
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #299
  • Jan 2, 2006 1:11am Jan 2, 2006 1:11am
Quoting keris2112
I was wondering the same thing, although the Vegas doc does suggest choosing a unit size and then trading 3 units (or 1 1/2). I think this is the wrong way of describing it, as it implies that you determine your position size and them multiply it by 3, thus, on each trade, risking 3 times what your money management method tells you to do. That's probably not what they meant, but that's kind of how it sounds.

I think we should do just the opposite. Determine our position size, based on whatever money management method we use, and then divide by 3 to determine how much to take off at each fib level. With trading platforms offering micro lots now (or any-size lots at oanda), this is easy to do. We don't have to trade in 3-lot (or mini lot) multiples.
That's exactly my view as well keris. In fact, I like to be even more conservative by rounding down the money management trade size to a multiple of three before dividing. Example: if the MM says you can trade 17 lots, I would round that down to 15 first and then divide by 3 to give a "unit" size of 5 lots. A half unit would be the 5 divided by 2 = 2.5 rounded down to 2 lots per half unit.

Doing the half units by rounding down the 17 to a multiple of 6, which is 12 and then dividing by three would give a half unit size of 4 lots which is too high. Better in my opinion to go with the more conservative 5/2 rather than 5/4.

Keris, I am really enjoying your contributions to this thread regarding the use of HMA and JMA. Thanks for that!

Dial
sxaxlxvxaxtxixoxnxbxyxgxrxaxcxexdxoxtxoxrxgx
  • #300
  • Edited 1:35am Jan 2, 2006 1:33am | Edited 1:35am
Hey Keris, thanks for posting the images comparing SMA/JMA/HMA. It is interesting to compare them. I might upgrade to some charting software that can support the official JMA at some point, although it looks as though HMA might be just as good. It is an interesting trade-off: HMA gets you in earlier but will have more of those jittery close-outs when the trend changes, while JMA will get you in slightly later (a less optimal start?) but will give you a smoother ride (and potentially a greater range that you can pull your profit from).

Just by eye-balling the charts, I would say that HMA would perform better than a SMA to trigger a trade and keep you in the trade. I also looked at some other 'modified' MA such as T3 but decided that HMA would provide a better result than the others.

Another similar indicator that I picked up somewhere was the "Hull Trend" indicator for MT4. If you were going to use the change of slope of the HMA to determine the trend (as HMA moves upwards the trend is up; when the slope of the HMA changes down the trend is moving down). The "hull trend" indicator shows pretty much the same trend except for a few odd bars during a major trend that show a 1-3 bar counter-trend movement.

Thanks for the help with the JMA. The FF is great - I wish I had found this forum a long time ago

Cheers, Linc.
0 traders viewing now