Slope LEVEL -- or DIRECTION?

[spotdespot]

Pretty much. I'm looking at the D1 direction more than the D1 level. So e.g. AUDCHF still has a pretty good level but the D1 TMA trend is heading the wrong way. I'd be tempted to short it even though the TMA is positive.

I haven't been using the H4. I may go back to that for more fine-grained timing but for now I'm trying to look just at PA.

Understand I don't have a great track record to back this up so don't take my word as gospel! It just looks like the right thing to me. I'm experimenting.

Gary - posted this here as my reply isn't specific to Jem's Uber Basket thread but I am replying to something you wrote there - I hope that makes sense!! .

I think you need to consider Bob's original ideas about slope value (D1 slope value) when considering direction.

What I mean by that is that even is slope direction has changed there may still be a move left in the direction of the value - kind of a momentum move if you like - it has caught me out a couple of times. So if slope direction has changed to down for example and slope value is still >0.8 then be careful that there isn't still a move upwards to come.

An extreme example of this is UJ (below) at the beginning of March - slope direction changed to down (D1) but slope value was >2.0 and price continued upwards with momentum for another 200+ pips over the next 2 weeks before it began to follow the direction.

uj.gif

I am now looking at direction and value - to try to determine whether I am getting an early warning (e.g. the UJ case) or something more imminent - which is determined by slope value.

AUDCHF we had an early warning (from around 20th July) but are getting to the point where we are going to see something more imminent imo. And I agree - definitely not looking long.

(Black line in the indicator window is D1 TMA Slope)

audchf.gif

[cmeade2]

Its working fine with me. I am posting it again

It totally crashes my Empty4.

[cmeade2]

What do you think about this approach?


happy Trading,

NeoTrader

-

Neo

This really is the greatest, especially when combined with the TMA/CSS cross for extra confirmation. In my view, it's really valuable in a market like this where you don't have a lot of strongly trending pairs on the D1 timescale.

[harry45]

Its working fine with me. I am posting it again

May be into your indi had been written pairs we do not have .

[Zypip]

I agree with Neo Trader, with both the things.

1) We should use True Tma, so that we can backtest it manually.

2) We should should use slope of slope for right enty.


So I had mod the 10.2 Tma slope to true and for right entry, use the color coding. so even the strength is >.8 it will show when to exit and when to enter in other direction.

Color Coding

Lime : UP trend
Green : Up Trend Weak
Light Green : Force is on UP Side

Red : DN Trend
FireBrick : DN Trend Weak
HotPink : Force is on DN Side

Please Play with It

Bharat

Hi Bharat,

Thanks for the work on this indi, it looks very interesting.

I am at work at the mo so will only be able to check if it works for me later today.

I have just one question if I may. Could you confirm that the chart you have posted displays the indi in TMA true mode ?

Cheers,

Zyp

[macman]

.....
So I had mod the 10.2 Tma slope to true and for right entry .....
Please Play with It

Bharat

Hi Bharat,

If you want people to "Play with it" which I assume is a euphemism for "Give me feedback" best is to supply the source file (.mq4) IMHO

That will get around the problems of it not working on other versions of Empty4 and you may even get suggestions from the hot coders around here about possible additions.

Best regards,

macman

[garyfritz]

What I mean by that is that even is slope direction has changed there may still be a move left in the direction of the value - kind of a momentum move if you like - it has caught me out a couple of times. So if slope direction has changed to down for example and slope value is still >0.8 then be careful that there isn't still a move upwards to come.

Agree 100%. That's why I was tossing out ideas like "put it into the 'buy' column when Slope is increasing rapidly AND/OR it's > 0.4 or whatever."

Slope is a range-limited oscillator so it can't keep increasing or decreasing forever, but price can. Slope is calculated by looking at the last N candles -- usually 20. Slope can change direction as soon as price changes direction, because now your lookback period includes some "up candles" in with all the "down candles" from the previous downtrend. But once the Slope lookback period includes only the time since the trend change, you have all "up candles" in your Slope calculation. Slope can't keep increasing unless the price move continuously accelerates, which won't happen. Once you include only the uptrend in your lookback, Slope will flatten even with a strong sustained up trend. It will pull back if the move slows down at all.

So yes, we can't look JUST at direction. I think direction change is a good way to catch the START of a move. Level is a better way to stay in a sustained move. We just have to figure out a good way to combine the two.

[garyfritz]

What do you think about this approach?

Neo, I like this idea a lot. It makes good sense to me as a way to capture the directional TMA moves. It probably won't work to stay in the sustained moves, as I mentioned above, but it might do a great job of grabbing bites out of a move.

I can't seem to get my setup to look quite as clean as yours. Notice how the MA on the D1 TMA hugs closely to the stairstep TMA values. Yours seemed a lot smoother. Did you set up your MA differently, or did you just use a longer MA on that chart? Mine is 21-bar EMA. Even if I use a 21-bar SMA, the average ends up touching the D1 TMA on almost every stairstep. Which actually makes sense, since this is an H1 chart so there are 24 H1 candles per D1 stairstep. Did you set it up so the MA looked at the last 21 D1 levels instead of the last 21 H1 candles, or did you use a length of 24*21 on the D1 MA, or...?

USDJPY.gif

[bharatchopra]

I got the problem, It will work fine now, the problem was of default MACD provided by crims, which is not right MACD VALUE, so copy both these indicator in indicator folder and it will work fine.

YES IT IS TRUE TMA, NO (SO CALLED REPAINTING).

BHARAT

[NeoTrader]

hi Gary,

Neo, I like this idea a lot. It makes good sense to me as a way to capture the directional TMA moves. It probably won't work to stay in the sustained moves, as I mentioned above, but it might do a great job of grabbing bites out of a move.

Yep, that is exactly how I use it....grab bites out of the move...and if the tma levels of the TFs rise in the same time you can also stay longer in the move and go for a larger bite....

I can't seem to get my setup to look quite as clean as yours. Notice how the MA on the D1 TMA hugs closely to the stairstep TMA values. Yours seemed a lot smoother. Did you set up your MA differently, or did you just use a longer MA on that chart? Mine is 21-bar EMA. Even if I use a 21-bar SMA, the average ends up touching the D1 TMA on almost every stairstep. Which actually makes sense, since this is an H1 chart so there are 24 H1 candles per D1 stairstep. Did you set it up so the MA looked at the last 21 D1 levels instead of the last 21 H1 candles, or did you use a length of 24*21 on the D1 MA, or...?

Hehe...thats my Magic...

No...there is no Magic...just take a closer look at my chart...there is no D1 TMA on my Charts because I experienced the same behavior like you did and therefor it is not really useful...
My setup is M15,H1,H4 on a M15 or H1 Chart with 21 SMA or EMA. No other special settings...

If you want to have the D1 TMA shown, attach it to the H4 TMA as a second TF (in my MTF Indicator). The MA will use the first TF (H4) for calculations and the D1 will not have a MA attached to it.
Another nice feature of this combination is that you can see when the H4 is crossing the D1 which also can be used as a directional trend bias.

happy trading,

NeoTrader

-