PYenner

I found it "healing" to read some threads that fizzed out. Many ideas fizz out and to see other people getting the same "failure" as I got, its so real world, it makes you feel better about keeping on trying. So I'm glad the "embarassing" stuff doesn't all get deleted. ADX is pure momentum, cci is a mix of momentum and price, so some things they do the same and some things different and maybe both might be useful. For now, a problem with cci is that it doesn't have a "normalized" output (0 to 100% or -100% to +100%). It needs to be normalized somehow before it can be used properly. You might have to beat me with a stick repeatedly to make me compromise leading for anything else, bone headed me...This is the area I am wanting to work on now, so maybe you wont have to wait much longer. Yes heat is a biggy, most things you try to reduce heat just shift the heat to later in the trade. We may have to talk more about tolerance. I am not clear about it yet. To me, the cyan line crossing zero gives a late entry. To me, the first short entry is when cci crosses zero. You scalp out on the first cci dip or you live with the pullback heat. After the pullback you get a second entry, maybe the big trade entry, hard to say with cci because the second ladder step looks shorter on cci that it does on the price chart. There is a second pullback and a third possible entry, maybe not so good a trade though. I would like to have a pullback level type indicator, like this pullback is small 20% maybe not good scalp entry, or this pullback is 50% ok for scalp entry, or this pullback is 100% (price crosses 5min magenta=perfect entry but need to have this showing on a 1min indicator) or pullback goes past 100% and is actually a reversal or exhaustion, danger danger. When you talk about tolerance I get my own ideas popping into my head and getting in the way. You may have to beat me with a stick to make me understand a different view. Part of the problem is that I see a couple of possibilities and I am trying to fit what you are saying into one or other of the possibilities I see. When you talk about a third possibility, I stuff up.. Just keep beating... Agree but still need a clearer picture. Maybe we should be exchanging pictures of what it should look like..

>>Trading is simple, but it is not easy. There are only three choices, up, down or sideways. So how hard can it be? Yeah right...

One nice article Doc >>This type of thinking borders the delusional. Now this I can do. The bigger problem is sorting out what isn't delusional. The learning curve is not just steep but also deep at times. >>Trading is simple, but it is not easy. That is one of the most confusing aspects. >>one must progress on a path, which is mental, emotional, physical and spiritual. For many, this is the most difficult journey ever taken. It surprises me how comprehensive that re-education gets, stuff you don't expect to be an issue becomes a stumbling block. It finds the self-deception little by little. >>Every trader must climb the four rungs of the Ladder of Learning, and must do this one step at a time. You cannot skip a step, but if you let your guard down, do not continue to study and practice, you can and will fall down a step or two. Exactly what is happening, bouncing between competence and incompitunce. Seems like every time I learn something new, it throws me off track for a while. Yet the sand is forever shifting. If you fail, its because you didn't think. But it is also often that fact that you did think that caused you to fail. It wasn't the right thought or it wasn't the right time to be questioning. Then its back to "what do I know, what do I rely on for trading?". Clear out the distraction and get back on track. As you say, its simple but its not easy. How many strands are there in this can of spaghetti?

Sundowner I know nothing about GE and it might be next year before I get through with adx and cci with vma. But constructive discussion is our best hope of finding the "common sense" path towards trading tools that work better and better. So the longer the thread lasts and gets ongoing input, the better. When someone does drop a pot of gold idea on us I will probably be the last person to see it, too deeply into my own "thread" to change gears fast. So please persist and be patient with me, I'm a slow poke. Forgive me if I use your chart as an example of an issue I am thinking about. Maybe I should also tell people where my thoughts are going. In your 15min chart one line lags behind the price by about an hour and the other lags by about 2 hrs. This is what averaging does, it produces time lag. You get the news but way too late. Now bemac used time lag from averaging to also produce price lag. The HE starts late after a minor reversal and it also starts short of the current price. That price lag is useful when the price reverses fully, gives a good visual sign that a reversal is possible and maybe shows how far the price has to come back towards the HE line before it counts as a solid reversal signal. My efforts to now have been about reducing the time lag so the vma lines could be useful in the here and now, up to speed with the price. But the next step is to start working back towards Bemacs strong HEs but to do it with the smallest time lag and maybe to have control over having more or less price lag as needed for particular purposes. There are other ways of getting price lag but to date I have not explored them. This is stuff that is inside the adx waiting to be fiddled with. Maybe it will work out, maybe it won't. It I can get strong HEs with minimum time lag and with user control over any price lag then we may have a whole new ball game for defining trend direction and maybe for other stuff. In particular vma1 on 5min does not deal well with a period of exhaustion after a strong move. It is taking too long for vma1 to "forget" about the past and deal with the present and its long slow magenta line is a real pest in that case because it lags so far behind reality. Once the strong move is over, the exhaustion period has to be treated as a new begining, forget about the past, it is gone. If I could reset the fantail or rainbow to zero thickness once the price gets into exhaustion, that would be more realistic than the hours of waiting we now have waiting for 5min magenta to get real and catch up to the current price action. While 5min vma1 is out of touch with reality, we only have 1min vma3 to rely on for trend emergence and we have no trend definition based on 1min vma3. New indicators might help there. Progressively the problem areas can be worked on. For the main trend definition, 5min vma 1 is the best. 5min magenta also has valuable uses at times but it also is a misleading pest at other times. Then there is cci on its own or cci+adx, neither has been explored far yet. Snipers cci version gave a good clean signal compared to regular cci, so that difference needs to be understood and maybe exploited, just needs a different approach. So if I seem to have nothing to say, or seem to be moving at snails pace, it does not mean that the light has gone out. This thread has a long way to go yet. I don't know whether to try to share my thoughts like this and appear a rambling fool to no purpose, or just get on with following my own path and wait for comments on the indicators that come out of it progressively. It is taking a surprising length of time to soak up what Walter has found and figure out how it does what it does. What bits are an advantage and what bits maybe could be done a better way. Of the last 3 indicators I did, line1 and line2 are intended to be useful but at this stage line3 needs to start having stronger HE before it will be useful, then line4 more HE etc. Work back towards Bemacs HE strength but with minimum time lag and maybe minimum price lag, so the HE line is right in the middle of the choppy little price swings during exhaustion or consolidation. When a trend does emerge from a tired period, vma also sometimes has the "Randy candle" problem, it starts off in the wrong direction then switches back on you. I dont know if this problem has a solution but it needs some thinking about because reversing a trade is not what you want to be doing. So there is a lot of stuff being tossed about and most of it comes from other peoples inputs so please keep it up, eventually something may come out of it. Construction communication can be priceless and it is one of the surprising and pleasing features of this forum.

Clym On a 1min chart, an ema line that comes near being like 5min vma magenta, would need to be like 500 periods on a 1min chart, wait 10 hours for it to start? I dont know how to define trend direction on 1min. Using a couple of ema lines that cross when the 1min trend reverses would be ok for a start. But would they be any use or just telling you what you already see in the chart? But they will also be telling you that a pull back is a trend in the opposite direction and it is only when this switches back to the previous direction that you have the entry signal for the bigger trades. So it may be about comparing how long the strong trend has lasted, compared to how quick the pullback takes to switch back to the strong trend. The longer the main trend holds, the closer it gets to being a 5min type trend definition and the more tradable it becomes and the bigger pullback can be before it should be taken as a trend reversal. I'm not saying this well. But that pullback thing is so useful that it needs to be developed for 1min charts maybe. Its a dynamic type of definition, its not like 10 pips or 20 pips, it depends on the strength or energy of the main trend as to how many pips of pullback should be used for defining trend reversal. Kinda needs a thread on its own too. Arggh.

Thank you Clym Big job done well as far as I can see. It doesn't have to be exactly the same as any previous version, it only has to be doing something useful. You may have to use slightly different rules for trading with it but that is ok. An example of one different rule might be the pullback towards 5min magenta rule which is a very good entry signal for the best trades. Firstly your magenta line might be closer to the price or further away from it, so your version of "price pulls back close to magenta" might need to be more or less close than Walter is using. Just need to work out your own rule for how close to magenta before it means a good big trade. Now an excel version of a 5min chart would need to run for nearly 1hr before it starts to give chart data. Bit of a pest so you will probably want to stick with 1min vma as you have. On 1min, a good 5min type pullback entry signal will go way past 1 min magenta, so you will need to work out your own rule for how much pullback is a good entry signal or when is that pullback actually the start of a trend in the opposite direction, meaning you want to be trading in the new trend direction, not against it. I don't have an alternative for Walters 5min trend definition. I have tried using the crossing of two ema lines on a 5 min chart and it gets close to Walters rule but not as good, it gives less certain results and you have to wait a bit longer to make sure its reliable. Now I have not tried to do it on a 1min chart, which is where you need to be trying out two ema lines for trend definition maybe. It might not be as good but it would still be useful. On 1min chart you might have to use sma instead of ema, not sure, needs experimenting. I notice that excel market feed is in quarter pip steps but platforms round this out to the nearest pip. I prefer the quarter pip data, more info. So expect some small differences. Excel may become a thread on its own yet because it is capable of complex stuff and may become very useful. One advantage is that it is not tied to any one platform. Thanks for breaking the hard ground for others. Cheers pyenner

On the 1hr chart it looks like audusd has been in consolidation chop all this week. May be why the trades have not been developing so well this week.

Either extreme pullback for super icon, or I just got burned again. 1hr vma1 up trend near peak, 5min vma1 downtrend than extreme pullback or genuine reversal? 1min has vma3 lines.

Before Fantail came along I had noticed in GBPJPY that there seems to be three degrees of pullback, ordinary (probably short of magenta), double (a bit past magenta) and extreme (triple or worse). The extreme case looks like clear trend reversal but it isn't, just an extreme pullback before a big icon payout. I saw this last night in major icons and was very impressed with how reliable vma1 trend definition was in audusd. I used to call those big pullbacks "confidence" tests, the bigger the test the bigger the payout, usually, but very testing to be trading without a fantail for a guide, easy to get drawdown or stopout on entry. Can't expect to ever have a 100% answer but even a 50% improvement is a big advantage to have on your side.

VMA1 is very well suited to audusd, it is not so well suited to some other pairs such as eurjpy where a stonger HE and a different definition of trend reversal might be needed. Vma1 is a remarkably good place to start and aususd is a good place to be using it. In time we may get to a better understanding of the "rules" but for now you have said that an expanding fantail is where we want to be and in that case where the market was unfriendly the fantail didn't get much wider than 5 to 10 pips, so that was a warning maybe. There will always be cases where the more aggressive entries fail more often and then it is time to be more conservative, but the last few days looked like quite hostile majors and I was impressed with how well the vma1 audusd trend definition survived the conditions. An excellent place to start in one of the most difficult markets to trade.

On the other hand, it was hardship that gave us Hitler.

Walter Are you happy with the trend definition you are getting from vma1 or do you want a version with more adjustments so you can monkey about with the settings?

Thanks sundowner It seems like 3 steps forward and 2 steps backwards but slowly we progress, with a bit of luck... Cheers

Seemed like quite a fast bear burn at 0.8900 too. Aggresive for AU.

Also watched it slide by, thought it was too slow...

Pix of sniper in sub window also appearing to not keep up with live ticks. Thats CCIVMAsub1 on 5 min, 1 min seemed ok...?

>>>>Bruce, did you test this one with live data yet ? would it plot same as historical ? cheers Walter. On my slow XP it does not keep up. Live different from historical. Your idea of a vma of a cci output could be a biggy. That would be like price --> cci --> adx --> vma Just shows the big rises or falls, ignores the smaller ones.

CCIVMA2 This is as far as I got with speeding up Sniper CCI in-chart version. It uses a lot of multiplication, squares, cubes etc. Very processor intensive. Seems to miss ticks when used as an in-chart indicator, yet seems to keep up as a sub-window indicator???? Might work ok on fast machines but my XP MT4 has slow down problems and M$ can be expected to slow XP down more with their updates now that Vista is selling. CCIVMA2.mq4

ADXVMALine3 Didn't explain the settings... ADX_Bars3=4 The others are set to 2 or 3, 4 helps give a "longer view". Weight3=3.0 Input smoothing, usually doesnt help to change this. ChandeEMA3=4.0 What do you call it? Works like an ema on the vma. Bigger number lags it and it doesn't swing so far on the peaks. Can also try all 3 lines with ChandeEMA3=1.0 for minumum lag. What is needed maybe is a strong HE line for the consolidation chop. Problem is this stuff can be 4 bar cycle, 15 bar cycle or even slower and thats makes it complicated for adx and maybe too for cci. I had hoped that cci might have more to offer here than adx which kind of struggles here. The other problem is the size of the chop, can be 5 pip can be 20 pip. Probably needs an indicator build for just chop HE and no good for anything else. Just wondering..

ADXVMA5Line1.mq4 ADXVMA5Line2.mq4 ADXVMA5Line3.mq4 Had to go back to using arrays to fix the live bar problem. So now its one line per indicator and 3 indicators per chart. They have been preset to be some use for scalper exits. They can be set up other ways. If you want to fiddle with more lines, just add more versions of ADXVMA5Line3 and try different values and line color. These are aimed at minimum lag, so dont expect smooth curves or long strong HEs. ADXVMA5Line1.mq4 ADXVMA5Line2.mq4 ADXVMA5Line3.mq4

CCIVMA3.mq4 This one is based on the MT4 CCI function. Snipers has so many multiplications in it that it misses ticks, then the live chart differs from historical. This one seems to keep up. CCIVMA3.mq4

You guys are a step ahead of me. Yes the adx vma might be fed with the cci signal instead of price. Might give solid HEs. Sheesh.

A vma of a vma of a cci? Heavy....

Walter There is also the possibility of combining two CCI signals that use slightly different speeds (bar counts) to reduce uncertainties that you get from just one cci. Don't know yet.

There is a lot of different information in the CCI curve, may need different approaches to bring out different things. Yes, hard to say yet what is possible or useful, but it seems to have more going for it than ADX. Better place to start from maybe.