| stevehopwoodforex.com https://www.stevehopwoodforex.com/phpBB3/ Print view |
|
| TMA Discussion Thread https://www.stevehopwoodforex.com/phpBB3/viewtopic.php?t=587 |
Page 4 of 17 |
| Author: | scartree [ Tue Jun 26, 2012 10:12 pm ] |
| Post subject: | Re: 10.2 for Dummies |
Thanks NeoTrader and NNBob for your responses to my suggested 'system'. I had not realised that the TMA repainted so much but I was looking at its use incorrectly. Bob, your simple precis of the system was very helpful thanks, I now have a better appreciation of the use of the TMA. I shall return to my reading having dispelled from my mind much of the complexity which had built up by so many and varied contributions to the forum. Karl |
|
| Author: | garyfritz [ Wed Jun 27, 2012 1:43 am ] |
| Post subject: | Re: 10.2 for Dummies |
The problem is that people (like scartree in this case, but it happens constantly) don't realize how much the indicator changes. They think they can look at a chart and understand how the indicator works in realtime. The indicator causes confusion because it changes -- and as far as I can tell, there is NO benefit to this, only confusion. As I said in Baluda's thread, the future-looking TMA makes the chart "pretty" by adjusting it so it "knows" what's going to happen in the future. By smoothing the true realtime values, it destroys valuable information about what you actually would have seen in realtime. Why not use the "true" or "no future looking" mode, so a chart accurately shows what you would have seen in realtime when you were trading it? Bob, is there some way that you actually USE the smoothed values from this indicator, in a way that you couldn't with an unsmoothed indicator? If not, WHY USE IT? I'd much rather have a bumpy indicator that shows me what I actually would have seen when I was trading, than a smooth and pretty indicator that only existed 10 or 20 bars after the fact. That smooth indicator is a total fiction and it is terribly misleading. Steve, delete this if you think it's inappropriate, but I think this is an important "dummies" topic. I think darn near every person who first looks at the NB10.2 system gets confused by this, and I don't think there's any reason for it. We could streamline the learning process for newbies and avoid answering the same confused question over and over if we just got RID of the fictional future-peeking indicator. |
|
| Author: | nanningbob [ Wed Jun 27, 2012 6:03 am ] |
| Post subject: | Re: 10.2 for Dummies |
The problem with the TMA true versions is they are not a TMA true but use LWMA in the code instead. I can post those if you want to see them and how they change the code to solve the problem. ZZ in his thread posted a version without the code and said I didnt know what I was talking about. TMA trues are not using ATR for their bars because that is what ATR does, it smooths out and filters out the noise. I dont know why you are having a problem with this because to me its simple, greater than 4 look for one good trade, greater than 8 look for multiple trades or hang onto the trade longer. TMA trues do not give that same signal because if you look at the numbers on the far right they will be different than the normal slope. This throws off the 4 and 8 line and how your read them. I have written after some of your posts several times and I can see you still dont get the use of the indicator. You seem to be still trying to use it as your trade signal and I dont use it for that. I only use it for strength of move to determine how long I wish to stay in a trade. I use pivots, S/R lines, and MA to make my entrances. Weekly pivot below monthly pivot--look to sell. Where do I enter, either the cross of the pivot/S/R line or trendline break on a rally or dip. How long do I stay in? Greater than 4 probably one good trade, greater than 8 ride it out, add to the trade, or do multiple trades. Dont need the LWMA True slope indicator for that. I want the ATR based one for that because that is how it tested out better. |
|
| Author: | nanningbob [ Wed Jun 27, 2012 6:11 am ] |
| Post subject: | Re: 10.2 for Dummies |
Here is the code he had to change to imitate the TMA so he could call it true: IB_TMA = iMA(NULL,0,xi_Period,0,MODE_LWMA,PRICE_CLOSE,i); Here is the same code line in the Normal TMA that I use: atr= iATR(NULL,0,atrPeriod,inx+10)*0.1; Then to try to make the thing work you add the 4 and 8 lines but if you look to the far right the range numbers of the indicator are different so the 4 and 8 lines do not have the same meanting the true slope as they do in the normal slope. That is why you are confused with the TMA true and it doesnt seem to work. They are not the same indicator, use the same math, have the same range settings. So TMA true doesnot work with the system. How do I know??? I put them side by side for two weeks and traded with them. I saw the difference and so threw the TMA true out of the system. That is why. |
|
| Author: | NeoTrader [ Wed Jun 27, 2012 7:28 am ] |
| Post subject: | Re: 10.2 for Dummies |
hi bob,
With my utmost respect I must say that you are wrong here... The ATR is used in both modes in the GetSlope Function to calculate the slope. See Code below Code: Select all and as you also can see how the slope is calculated it uses two Values and two different calculations also in TmaTrue Mode - The CalcTmaTrue calculation is as you said based on the LWMA but the CalcTmaTruePrev calculation is based on the normal TMA calculations. I'm sorry to say that again, but this is not true. I'm watching both true and normal tma for weeks now and the value of the last (essential) bar is always identical. You can try it yourself with the attached and modified version of the TMAslope. You can switch the tmaMode in the Parameters. Please compare it to your TMAslope...I think your assumptions and your view of the TMAtrue is based either on an old or wrong version of the TMAtrue slope. The actual version shows no difference in the values of the last bar. It only doesn't repaint. I don't want go there to ask who is confused here... I respect you as a Individual and a tremendous teacher... and I value your opinion. I will only ask that you please compare the actual versions of tma and tma true that we are discussing here. Because this is really a topic that confuses the People the most. You say that the values are different...and People like me and Gary say the Values are identical...we should really settle this topic once and for all... happy trading and a nice vacation to you NeoTrader - |
|
| Author: | nanningbob [ Wed Jun 27, 2012 11:22 am ] |
| Post subject: | Re: 10.2 for Dummies |
Well the one you posted and the one I used are not the same. I started this late last Feb early March and began talking about it in April. I then posted my test versions by May. The one you posted came way after that. It may work but since I had already discarded the TMA true I have not used any since. If you look at the pic below is the NRP version of the normal TMA slope. Notice several things: 1. The bars are not smoothed but up and down jagged. Having watched these bars over periods of months the regular TMA I use does not do this. They are not jagged from time period to time period on a high time frame, therefore the bars on the so called true are erroneous. They follow price action like a MA bar would. The regular TMA does not do this. 2. Look at the numbers to the far right, they are very different. #1 top range is 1.536 etc. The so called true range #2 is 3.075 etc, or almost double the numbers of the regular TMA. Since the range of the indicators are vastly different, almost double, the 4 and 8 lines have very different meaning in the indicator. See the pic below, there is the evidence clearly marked, that the current numbers are the same does not make them the same. The reference numbers are vastly different so the indicator has a different approach. |
|
| Author: | nanningbob [ Wed Jun 27, 2012 11:38 am ] |
| Post subject: | Re: 10.2 for Dummies |
Just to be fair I loaded the TMA true you had posted and compared the three. To also be fair I would have to watch this one for a couple of weeks to judge how it compares with the origianl designed with the system. I am not opposed to change if this is better version. Some thoughts: 1. Obviously this is an improved version over the one I originally tested. 2. However the range parameters are different but this time about half of the original. The original 1.536.... see #1 and this one is 0.846.... see #3 My first reaction to that would be it would throw off the 4 and 8 numbers as markers because the range is now different. To be far though i will have to watch them work together side by side and see. See pic below. |
|
| Author: | nanningbob [ Wed Jun 27, 2012 11:42 am ] |
| Post subject: | Re: 10.2 for Dummies |
Just to give you an idea of my testing I now have 10 different versions of the TMA slope. Some I downloaded and some were adjusted by programmers who helped me with the system. I do have some idea of these things, maybe not perfect but I do. I also reset the lines to the True to .02 and .04 which is half because the range given is approximately half. Just a quick look gave me similar reads as the 4 and 8 on the regular. In other words when price goes beyond 2 = one good trade and if it goes past 4 then you can look at two or more good trades. If that holds true then you might find it useful. |
|
| Author: | NeoTrader [ Wed Jun 27, 2012 12:47 pm ] |
| Post subject: | Re: 10.2 for Dummies |
Hi bob, ok...I understand at least now what you refer to as the numbers on the right... I always assumed that you meant the current value of the actual bar...but you are referring to the range of the whole slope...and that this ranges are different. Now something to think about... since the range on the original repainting TMA slope version is also based on the fictional calculated values of the past bars and the range of the TMAtrue is solely based on real values how they occurred in real time shouldn't it be more relevant that the current value of the bars are more relevant...and actually equal on both versions? If I grasp your system and your use of the TMA slope correctly, mainly the current value of the TMA slope is of relevance and the angle of the slope itself. Since the TMAtrue values are real values as they occurred in real time and are not manipulated afterwards as in the original TMA slope this TMAtrue slope is more accurate and also allow to see how the TMA slope really behaves and also shows the real range of the TMA. If this means that we need to reassess the levels. OK, that should be doable and actually easier than with the original TMA slope since you can reliably backtest the TMAtrue slope visually. I still do not see any value in the smoothening (beautifying) of the slope afterwards since it does not bring any value of information on the table...in contrary it only confuses most of the people, especially newcomers. The angle of the slope is also available in the TMAtrue version and actually more accurate than the smoothened version. I myself use the TMAtrue slope according to your rules with 0.4 & 0.8 levels and it still works like a charm. But i also will reassess the levels like you described and will see if it make any great difference...and maybe also try some different levels too Please see my points as constructive critic and observations that I made. I don't want to insult you or be disrespectful. This is your baby and you put a lot of work in it. I belief in your strategy and also in you. I also think that it could be a bit easier to the newbie trader if they would see what is really going on with the TMA slope and don't be irritated by this repainting issue. happy trading, NeoTrader - |
|
| Author: | NeoTrader [ Wed Jun 27, 2012 1:17 pm ] |
| Post subject: | Re: 10.2 for Dummies |
hi paul, I don't think that we need to redefine TMA slope because it is correctly defined as you defined it. Why name the TMAtrue slope differently as the indicator that already exists (TMA True v2.1 is also used in NB's strategy on H1) and how the originator of this zznbrm named it ... we don't need more confusion than already exists.. True is also more suitable to this slope as the true value of the current bar is preserved in the slope. I don't think anyone really qualified it as useless...only useless for (visual) backtesting. I don't know how do you come to the conclusion that this is the question what we talking about. The value of TMA slope was never in question...as all the different strategies that evolved here prove...every single one of them use the TMA slope as an essential base of there strategy. The main topic that comes up all the time is the repainting (beautifying) of the past bars of the slope and if or why it is necessary. hmmm... we are here in the dummies section and the TMA slope is an essential part of the 10.2 strategy and also something that newbies struggle sometimes (because of the repainting I think it could stay here...but let steve decide if he wants to move it elsewhere.. happy trading, NeoTrader - |
|
| All times are UTC | Page 4 of 17 |
|
Powered by phpBB® Forum Software © phpBB Limited |
|